Slashdot Mirror


OpenSolaris From a Linux Admin and User Perspective

MSa writes "How does OpenSolaris, Sun's effort to free its big-iron OS, fare from a Linux user's point of view? Is it merely a passable curiosity right now, or is it truly worth installing? Linux Format takes OpenSolaris for a test drive, examining the similarities and differences between the OS and a typical Linux distro. If you want to sample the mighty ZFS filesystem, OpenSolaris is definitely the way to go."

370 comments

  1. maybe I should go and play around with this! by jacquesm · · Score: 5, Interesting

    Ever since the demise of SGI I haven't looked at anything but Linux / BSD, but this makes me wonder if there is maybe life for Solaris after all.

    Would be nice if this was more geared towards the server end of things, which is where I would expect you'd deploy solaris much sooner than on the desktop.

    1. Re:maybe I should go and play around with this! by Anonymous Coward · · Score: 5, Funny

      What do you mean, the demise of SG-1? The Apophis was defeated and the replicators contained.

      Oh, SGI. Sorry.

    2. Re:maybe I should go and play around with this! by Ruud+Althuizen · · Score: 2, Interesting

      I've been a sys admin for FreeBSD and Linux machines for a computer club for a few years now. We've had one Solaris machine in the lineup for a long while (its three PSUs loudly exploded while debugging the hw yesterday). And I must say that it is a robust OS on robust hardware; we've never had to look at it much: it just worked (really, it did). Though I never got the hang of it. The OS has some oddities here and there, stuff you need to know that are specific to the OS.

      --
      **TODO** Steal someone elses sig.
    3. Re:maybe I should go and play around with this! by Anonymous Coward · · Score: 5, Funny

      I've been a sys admin for FreeBSD and Linux machines for a computer club for a few years now.

      I bet you have to beat the ladies away with a stick...

    4. Re:maybe I should go and play around with this! by alancdavis · · Score: 5, Informative

      The review didn't address desktop vs. server and as a "lightweight" review doesn't look any deeper than the distro package for answers to the questions and objections raised.
      OpenSolaris works well as a server OS - that /is/ it's heritage. It's easier to run OpenSolaris headless and on a serial console than any of the *BSD and Linux distros that I've used over the years. All of the "standard" server packages are available to run web and net services out of the box. For truly lights-out server rooms it's still necessary to choose hardware that implements some sort of remote power cycle or remote system monitor capability.
      The ZFS filesystem is interesting for desktop installations - it does allow seamless use of the 1-2 terabyte desktop disk configurations that are now possible. ZFS was designed for the datacenter - eliminating the need for the time-honored but fragile combination of journaling filesystem over software volume manager (usually over HW RAID). It's the first real innovation in filesystem architecture since journaling filesystems were developed.
      Additional software packages are available from 3 well-known (in the Solaris community, at least) sites. Sun has it's own freeware site, blastwave.org and sunfreeware.com
      http://www.sun.com/software/solaris/freeware/s10pkgs_download.xml
      http://www.blastwave.org/
      http://sunfreeware.com/

      The package manager for blastwave.org is their own, the others use the standard Solaris pkgadd commands. The package naming convention is a long-standing convention - each vendor uses a different prefix, making it easy to differentiate the source of packages.
      OpenSolaris commands, where Sun hasn't replaced stock UNIX commands with their own, reflect SVR5 standard rather than the more Linux-ish BSD syntax.
      One of the places where Sun has replaced "normal" functionality is the init process. SMF is Sun's attempt at fixing the long-standing problems and in-efficiencies of the BSD or SVR5 init process. Apple has launchd, there's openrc and gentoo's baselayout that all have similar goals. SMF works well and there's a fair amount of support on the net for integrating non-distro apps.
      One of the "why OpenSolaris" answers is that there is value in running the same OS on the desktop as on the server. For Solaris shops OpenSolaris on the x86* servers provides a common platform that enables system management efficiencies to be extended.

    5. Re:maybe I should go and play around with this! by jacquesm · · Score: 1

      thank you!

      I'll definitely give it a spin to see how well it stacks up to 64 bit linux on opterons (that's what we're running for the most part).

    6. Re:maybe I should go and play around with this! by $RANDOMLUSER · · Score: 2, Informative

      I think that's kind of harsh - comparing Solaris to Irix. In the first place, Irix was never meant to be more than a workstation OS, and it was crufty and crappy enough that you'd never want to run it on big iron. Solaris, on the other hand, even with its roots in workstations, has always run well on the largest servers. While the Irix/Linux comparison might be valid for desktops and small servers, a better camparison for Solaris might be HP-UX, as they're both more aimed at the data center than the desktop.

      --
      No folly is more costly than the folly of intolerant idealism. - Winston Churchill
    7. Re:maybe I should go and play around with this! by Directrix1 · · Score: 1

      Coming from somebody that managed servers just plugged in with an internet connection from half a continent away. I can say running Linux headless is as easy as you can get. With regards to ZFS, I dream of there being an equivalent on Linux. It may happen some day, but until then its RAID5/6+LVM+ext3.

      --
      Occam's razor is the blind faith in the natural selection of least resistance and in universal oversimplification. -- EF
    8. Re:maybe I should go and play around with this! by Znork · · Score: 3, Interesting

      With regards to ZFS, I dream of there being an equivalent on Linux.

      Out of curiosity, what particular features in ZFS do you want on Linux? I mean, it's a large step upwards from solaris disksuite, but compared to the linux device-mapper/filesystem paradigm it's not a particularly large improvement (if it is one at all).

      Having actually run ZFS in production, there are some serious drawbacks with the remaining features (copy-on-write fragmentation, problems in SAN environments, etc), that may leave one wishing they'd implemented the ZFS features in a more stackable way so you could easily discard inappropriate layers and features.

    9. Re:maybe I should go and play around with this! by C0vardeAn0nim0 · · Score: 3, Informative

      The ZFS filesystem is interesting for desktop installations - it does allow seamless use of the 1-2 terabyte desktop disk configurations that are now possible. ZFS was designed for the datacenter - eliminating the need for the time-honored but fragile combination of journaling filesystem over software volume manager (usually over HW RAID).
      It's the first real innovation in filesystem architecture since journaling filesystems were developed.

      just karma whoring here, but it's important to mention that pretty much everything ZFS has to offer was already available on tru64's advFS: http://advfs.sourceforge.net/

      it's a shame HP killed this fine unix to keep that abominable HP-UX, so kudos to sun for bringing back the functionality of tru64 back to the datacentre AND the desktop.

      hmmm, i wonder if my notebook (presario v6210) is compatible with opensolaris...

      --
      What ? Me, worry ?
    10. Re:maybe I should go and play around with this! by Anonymous Coward · · Score: 0

      Oh, the ladies had been hit with a stick alright...

    11. Re:maybe I should go and play around with this! by khb · · Score: 1

      Blastwave is off the air. It has largely been replaced for the moment by :

      There is an elaborate story behind this (isn't there always). The solarisx86 yahoo group has discussions entitled: "[solarisx86] Blastwave gone?" and "[solarisx86] All assets of Blastwave frozen.." which may be of interest.

    12. Re:maybe I should go and play around with this! by outZider · · Score: 5, Insightful

      You've run ZFS in production, yet you can't see the improvement on Linux's model? You mean the fact that md is completely broken and LVM is unreliable and slow by comparison?

      --
      - oZ
      // i am here.
    13. Re:maybe I should go and play around with this! by fm6 · · Score: 2, Informative

      Irix was never meant to be more than a workstation OS...

      Say what? SGI sold servers too. In the late 90s it was their main business. The email server at AOL used to be an SGI Origin — running IRIX.

      SGI's business was an is (they still exist, albeit as a very tiny supercomputer/server vendor) high performance computing. Doing HPC with workstations and doing it with servers and supercomputers is not all that different. And until they shifted from MIPS to x64 and Itanium, all their systems ran IRIX.

    14. Re:maybe I should go and play around with this! by doomicon · · Score: 1

      "You've run ZFS in production, yet you can't see the improvement on Linux's model? You mean the fact that md is completely broken and LVM is unreliable and slow by comparison?"

      Sir, I wish I had points to mod this up!

      --

      Awesome!
    15. Re:maybe I should go and play around with this! by fm6 · · Score: 1

      HP didn't kill tru64, the marketplace did. This OS never ran on anything except Alpha and Itanium. Not a lot of demand for those.

      The marketplace is littered with the corpses of superior OSs that failed: CP/M, QNX, CTOS (which all ran on x86, damn it!). Thinking too much about them is not healthy.

    16. Re:maybe I should go and play around with this! by jhol13 · · Score: 1

      1) Simplicity. It is trivial to set up and administer.
      2) End-to-end checksums.
      3) Copy-on-write and always consistent disk (no need for journal or fsck).
      4) Build in compression and encryption.
      5) Snapshots, etc.
      6) NFS4.

      It is so huge improvement from Linux that my home server will be Opensolaris with ZFS and raidz2.

    17. Re:maybe I should go and play around with this! by JBdH · · Score: 2, Informative

      It may happen some day, but until then its RAID5/6+LVM+ext3.

      Right, and that day better come soon, at last before 2TB drives become mainstream in RAID5/6 setups. read http://blogs.zdnet.com/storage/?p=162 about the chance of double failure on RAID5/6 growing with the size of the disk. Happened twice on my servers, exactly with RAID5+LVM+ext3. We need ZFS (or similar) on linux servers right now.

    18. Re:maybe I should go and play around with this! by jacquesm · · Score: 1

      QnX is far from a corpse, it is very much alive and kicking especially in the embedded market.

    19. Re:maybe I should go and play around with this! by True+Grit · · Score: 1

      ZFS won't protect you either. From the article you linked to:

      As well-informed commenter Liam Newcombe notes:

      The key point that seems to be missed in many of the comments is that when a disk fails in a RAID 5 array and it has to rebuild there is a significant chance of a non-recoverable read error during the rebuild (BER / UER). As there is no longer any redundancy the RAID array cannot rebuild, this is not dependent on whether you are running Windows or Linux, hardware or software RAID 5, it is simple mathematics. An honest RAID controller will log this and generally abort, allowing you to restore undamaged data from backup onto a fresh array.

      AFAICT, this problem is inherent in the RAID Standard itself (i.e. the firmware of the RAID controller), not specific to any OS.

    20. Re:maybe I should go and play around with this! by fm6 · · Score: 1

      Alive? Sure. As is CP/M. (I'm pretty sure CTOS is quite dead.) But hardly kicking.

      QNX was originally designed as a general-purpose PC OS. When I first heard of it, around 1987, its feature list was pretty impressive compared to the leading OSs of the time. (Assuming you can call MS-DOS 2.0 an "operating system", which is debatable.) They actually claimed to do serious multitasking on 8088-based systems. No reason to be skeptical: I'd seen other OSs do it on similar hardware. But I was stuck using MS-DOS, which not only didn't support multitasking, it was actively hostile to it, due to some poor coding by Tim Patterson. There were various kludges to implement this and other "real" OS features, but they were pretty evil.

      QNX survived, but only by tightly focusing its market. As you say, it's always been big as an embedded OS. (I seem to recall that it used to be the leading OS for some kinds of POS systems.) Nowadays it's marketed as a "realtime OS". That's honest enough, but it's really a good general-purpose OS that runs on very limited hardware. I've always been frustrated that it never captured significant mindshare.

    21. Re:maybe I should go and play around with this! by Anonymous Coward · · Score: 0

      On that note, see this handy chart here.

      Yes, it comes with Angelina Jolie pics. :-)

    22. Re:maybe I should go and play around with this! by ZaMoose · · Score: 1

      Huh? Blastwave.org is still loading just fine for me...

      --
      I wish I had a kryptonite cross, because then you could keep Dracula and Superman away.
    23. Re:maybe I should go and play around with this! by m50d · · Score: 1

      Where's the broken? No, really. I'm writing this from my lvm-over-md setup, 120mb/s read speeds (and I haven't done any settings tuning). What's the problem?

      I haven't tried ZFS, because as I understand it can't do raid5-like parity (I don't quite feel I have the spare diskspace for mirroring), so I can easily believe it might have advantages, but there's not a whole lot *wrong* with the linux model that I can see.

      --
      I am trolling
    24. Re:maybe I should go and play around with this! by Kent+Recal · · Score: 5, Informative

      I'd like to give Znork some credit here.
      ZFS is really, really nice but it does have some warts and the biggest for many would be that arcane operating system that's dangling off its nutsack. Yes the solaris kernel is great, scales like a champ etc. but the userland and the lack of centralized package management (in 2008, no less!) are bad joke.

    25. Re:maybe I should go and play around with this! by Anonymous Coward · · Score: 0

      you cannot contain replicators, only the ones made by aphohis...
      replicators ..after all where made by the ancients.
      so by logic either fix the replicators, or destroy the ancient tech (including the stargates).

    26. Re:maybe I should go and play around with this! by RedK · · Score: 1

      No raid5-like parity in ZFS ? I don't think you've looked very deep into ZFS then. Try RAID Z.

      --
      "Not to mention all the idiots who use words like boxen."
      Anonymous Coward on Monday August 04, @06:49PM
    27. Re:maybe I should go and play around with this! by Anonymous Coward · · Score: 0

      If you're wondering "If there's life after all" for Solaris, you clearly havn't been paying attention.

      OTOH, I find myself wondering "If there's life after all" for Linux.

    28. Re:maybe I should go and play around with this! by ci4 · · Score: 1

      I am running it perfectly well on a HP nx6310. The Intel wireless "just works" with nwamd out of the box; had to find a driver for the wired (Broadcom) i/f and an update for the audiohd driver.

    29. Re:maybe I should go and play around with this! by jedidiah · · Score: 1

      By the time that you would have heard of it in 1987, it was not at all considered a "general purpose OS".

      Even in those days it was considered something intended for "process control" rather than desktops.

      The only people running it as such would have been demented QNX distributors.

      --
      A Pirate and a Puritan look the same on a balance sheet.
    30. Re:maybe I should go and play around with this! by jedidiah · · Score: 1

      Go to San Francisco when they're holding Oracle OpenWorld.

      That will cure you of your "wondering".

      --
      A Pirate and a Puritan look the same on a balance sheet.
    31. Re:maybe I should go and play around with this! by Anonymous Coward · · Score: 0

      just karma whoring here, but it's important to mention that pretty much everything ZFS has to offer was already available on tru64's advFS

      Data checksuming? (Which may be more important than any other feature if you think about it--how much do you trust your data integrity?)

    32. Re:maybe I should go and play around with this! by Lally+Singh · · Score: 1

      Two quick additions:

      1. I donno about the rest of them, but SMF does automatic dependency analysis for startup & restart. Something depend on a service that just sig 9'd? SMF will restart it, b/c it's probably hosed too.

      2. DTrace. Holy God DTrace. How did everyone miss dtrace?

      I don't know.

      DTRACE IS 37ee+ h4x0rz +007. By that, I mean you can finally figure out wtf your system is actually doing.

      Oh, and unlike Redhat et. al.'s support, when you call sun for support of their OS, you're calling the people that actually *wrote*the*code*.

      --
      Care about electronic freedom? Consider donating to the EFF!
    33. Re:maybe I should go and play around with this! by jabuzz · · Score: 1

      And I wish I had mod points to mod you troll. So lets see may be the lack of user quotas on a shared file space didn't fly for him. For me that is a complete deal breaker.

      Perhaps it because he has a SAN, and doing the snapshotting and other random LVM tasks in the file system does not fly for him, because he wants to do things like move the VG's from one machine to another.

      Maybe he wants a cluster file system, or one that does HSM. I know I do and ZFS as of today does neither.

      ZFS is designed for managing a bunch of direct attached hard disks in thumper or similar device. At anything else it is frankly a bit sucky.

    34. Re:maybe I should go and play around with this! by fm6 · · Score: 1

      Your memories differ from mine. Besides, what about QNX prevented it from being used as a desktop OS? Aside from that fact that they had no hope of competing with Microsoft.

      Part of my perspective comes from working for Convergent Technologies, which sold x86-based systems running their own OS, called CTOS. Like QNX, CTOS was microkernel based, had a message-passing architecture, and a really good scheduler. Now, you could use these systems for "process control," but you could also use them for ordinary office work. That wasn't overkill: these OS features made it possible to implement some very fancy user interfaces. I worked for them as a technical writer, and did most of my writing on that system. The level of support for good interactive software wouldn't be matched in the PC world for many years. And the advanced features of the OS enabled this on very limited hardware.

      Perhaps my ignorance is showing (or my memory is faulty), but it seemed to me that CTOS and QNX were very similar. The most important differences were that QNX was designed to run on generic hardware (no such thing when CTOS was invented in 1979) and to provide an API that would be intelligible to UNIX programmers. And if that's true, QNX would have been a good general-purpose OS from day one, even if it does do real-time stuff well.

    35. Re:maybe I should go and play around with this! by SETIGuy · · Score: 1

      Where's the broken? No, really. I'm writing this from my lvm-over-md setup, 120mb/s read speeds (and I haven't done any settings tuning). What's the problem?

      Well, to start with md doesn't talk with the SMART features of drives. When md gets an uncorrectable read, the proper and intelligent course of action would be recognize the error, get the correct sector values from the mirror (or calculate it from the other disk values on RAID4/5/6) and rewrite the sector which would probably prompt a relocation of the sector (assuming the write fails).

      The current md course of action is to suspend the read until the drive is able to read the sector without error. Which will tend to be never.

      "Where's the broken?" There's the broken.

    36. Re:maybe I should go and play around with this! by SETIGuy · · Score: 1

      about the chance of double failure on RAID5/6 growing with the size of the disk. Happened twice on my servers, exactly with RAID5+LVM+ext3.

      Correct me if I'm wrong, but isn't the reason for RAID6 the ability to survive a double failure. Maybe the first time you got a RAID5 double failure should have been the day you switched to RAID6.

      And are you really using ext3 for large filesystems? If I had 3 days to wait for a fsck to complete, maybe....

    37. Re:maybe I should go and play around with this! by Alpha830RulZ · · Score: 1

      You mean the fact that md is completely broken and LVM is unreliable and slow by comparison?You mean the fact that md is completely broken and LVM is unreliable and slow by comparison?

      While I haven't run ZFS to compare it to, I don't understand why you would assert that md is broken and LVM is unreliable. I have a good dozen machines that have been running for 4+ years without incident. I have nothing but good things to say about them, and the FS performance is far better than the Windows servers we replaced. In fact, we specifically needed better disk performance than NTFS was capable of, and got it.

      --
      I was taught to respect my elders. The trouble is, it's getting harder and harder to find some.
    38. Re:maybe I should go and play around with this! by 19Buck · · Score: 2, Insightful

      "Would be nice if this was more geared towards the server end of things"
      in that case what you want is Solaris 10, not OpenSolaris

    39. Re:maybe I should go and play around with this! by mrbooze · · Score: 1

      Out of curiosity, what particular features in ZFS do you want on Linux? I mean, it's a large step upwards from solaris disksuite, but compared to the linux device-mapper/filesystem paradigm it's not a particularly large improvement (if it is one at all)

      Block checksums?

    40. Re:maybe I should go and play around with this! by Anonymous Coward · · Score: 0

      I've been a sys admin for FreeBSD and Linux machines for a computer club for a few years now.

      I bet you have to beat the ladies away with a stick...

      Beat something, anyway.

    41. Re:maybe I should go and play around with this! by Znork · · Score: 2, Informative

      yet you can't see the improvement on Linux's model?

      Yah, the OOM crashes, flush hangs, dbf fragmentation and performance degradation issues really left a bad aftertaste. I didn't go into those details, as anyone else who'd run ZFS in production for some time would most likely know about them. Some are fixed now, but improvement... lets just say that any improvement I could see was rather outweighed by the problems.

      Don't get me wrong, ZFS is nice if you use it where it has it's strengths. It's a perfect filesystem for hardware like thumpers serving out NFS. Unfortunately it seems to be designed with that, and only that, in mind; even worse, technical evangelism at Sun appears to have failed to mentioned that to their tech sales, getting ZFS installed in ways it's not at all suited for.

      A more flexible architecture would have allowed ZFS to be great for many things where it can actually get in the way today.

      You mean the fact that md is completely broken

      So don't use md. Use lvm mirroring. Or hardware mirroring. Or drbd remote mirroring. See, flexibility. If a layer doesn't do what you want it to, change it.

    42. Re:maybe I should go and play around with this! by Hucko · · Score: 1

      Really?? I stopped using it after ~4 hours because of the afforementioned downtime of Blastwave. (Suns packaging is horrible!) I was hoping that Blastwave would make it useable.

      --
      Semi-automatic amateur armchair Australian philosopher; conjecture ready at any moment...
    43. Re:maybe I should go and play around with this! by Anonymous Coward · · Score: 0

      I've been a sys admin for FreeBSD and Linux machines for a computer club for a few years now.

      I bet you have to beat the ladies away with a stick...

      I wish the ladies would let my beat them with my stick...

    44. Re:maybe I should go and play around with this! by JBdH · · Score: 1

      Correct me if I'm wrong, but isn't the reason for RAID6 the ability to survive a double failure.

      You're correct, but the article I linked describes that that won't work forever either. RAID5 or RAID6 enables me to bypass LVM in certian solutions, if I use RAID1 I certainly need to lump several RAID1's together with LVM, which in my opinion is better than LVM over RAID5/6 but still somewhat flaky.

      And are you really using ext3 for large filesystems?

      Not anymore no, I use XFS for large filesystems these days. There was a time though that ext3 was the only journaling filesystems out of beta for Debian. And don't start about reiserfs on debian, long before Hans killed his wife he practiced on my servers.

    45. Re:maybe I should go and play around with this! by jacquesm · · Score: 1

      I must have been hallucinating all those Challenges and Origins then :)

    46. Re:maybe I should go and play around with this! by jacquesm · · Score: 1

      I'm not aware of anybody actively selling/maintaining CP/M for customers but then again you never know.

      However QnX seems to still have very active support and http://www.qnx.com/ is quite up to date (especially when contrasted with CP/M, which I would classify as 'mostly dead').

      In the real time arena they are very well represented, and I think that without knowing it you probably use their products at least once or twice daily.

      Think automotive, energy supply, industrial automation, medical equipment and such.

    47. Re:maybe I should go and play around with this! by pyite · · Score: 1

      ZFS won't protect you either.

      ZFS RAID-Z won't protect you from a complete double failure. What ZFS will protect you from is silent corruption that would result in an effective double failure. If you have one bad disk in a RAID 5 array and then you try to rebuild the array, but one of the remaining drives has already corrupted blocks, you're out of luck. In ZFS, it's much less likely that you'd ever have corruption as there are end to end checksums.

      --

      "Nature doesn't care how smart you are. You can still be wrong." - Richard Feynman

    48. Re:maybe I should go and play around with this! by jacquesm · · Score: 1

      Yes, what with solaris taking market share away from all the other os's lately...

      And then of course IBM, HP and all those others switching away from Linux to Solaris the tide has really turned. Even redhat is rumoured to be secretly thinking of proposing to drop Linux in favour of solaris ;)

      The only arena I think where Solaris currently can still claim significant marketshare is fortune 500 accounts.

      This:

      http://finance.yahoo.com/echarts?s=JAVA#chart1:symbol=java;range=5y;indicator=volume;charttype=line;crosshair=on;ohlcvalues=0;logscale=on;source=undefined

      also tells a pretty bleak story.

      The simple fact that SunW has been changed to 'java' should tell you what sun itself thinks of it's business and the direction they'll be taking it in.

    49. Re:maybe I should go and play around with this! by m50d · · Score: 1

      That's directly contradictory to what the md documentation says; if a read fails it will rewrite it, then attempt to read it again, and only then fail the drive if that fails.

      --
      I am trolling
    50. Re:maybe I should go and play around with this! by SETIGuy · · Score: 1

      The problem is that the drives don't typically return an error when there is a sector read problem. It returns its best guess and then marks the sector as uncorrectable and "pending" relocation. Pending sectors get relocated the next time they are written. The md driver does not detect this condition and does not rewrite the stripe, so the condition persists. Bruce Allen (the author of smartd/smartctl) says he has been trying for some time to get the author of md to incorporate recognition of this condition. If you've got a new enough drive (one that supports a "write uncorrectable" command to write a sector with bad ECC to a drive) you can try it out for yourself.

      On the other hand, good quality hardware raid controllers do seem to recognize and correct these issues, even if it is only when an array is verified. But you may have noticed that "verify" is not an option that the md driver supports. It supports two options... 1) rebuild a failed device 2) recalculate the parity blocks. The problem with 1 is that it requires the entire drive to be failed. The problem with 2 is that it assumes the data is good and the parity is bad.

      If you're managing large arrays you see this "current pending sector count" condition far too often. You end up having to fail the entire drive and forcing a rebuild on a spare. Then you use dd to overwrite the entire failed drive, check to see that "current pending sector count" has gone to zero, Use dd to read the entire drive to see that the "current pending sector count" is still zero, and then either pull the drive to use it in some less critical task, or put it back into an array if you have a good feeling about whether it's going to keep working. DO NOT USE A DRIVE THAT HAS BEEN IN THIS STATE AS A SPARE! That is asking for trouble.

      If it were possible to do a partial sync, maybe someone could write a tool to fix these problems. As far as I know, there is no way to get md to sync a single stripe (and if it were you would need to tell it which component of the stripe was bad).

    51. Re:maybe I should go and play around with this! by fm6 · · Score: 1

      According to their web site, they're in a lot of bluetooth devices. So yeah, I was overstating their defunctness. But they definitely are dead on the desktop.

      Their article on Wikipedia links an interesting (if somewhat messy) article on the ICON computer, a QNX-based desktop that was commissioned by a Canadian public school system. Soon destroyed by IBM-compatibles, of course. A depressing glimpse of unrealized potential.

      You can still buy the current incarnation of CP/M (now called Multiuser DOS) from Concurrent Controls. But I had to really hunt for that web site — there's obviously very little interest. And the web site hasn't been updated in a couple years.

    52. Re:maybe I should go and play around with this! by Anonymous Coward · · Score: 0

      Who uses md nowadays anyway?

    53. Re:maybe I should go and play around with this! by Allador · · Score: 1

      The #1 primary thing for myself and most folks is data integrity.

      End-to-end checksumming and automatic recovery. It's so huge it just blows by most people.

      Most current systems (at least in the mainstream x86/x64 world) have ridiculously high error rates and data corruption rates.

      The systems have evolved over the years with this a known factor, and so systems are reasonably robust to that decay, at least at the lower level. But its a real problems.

      System busses introduce errors. Network cards (particularly TOE) introduce errors. Disk controllers, cables, and drives introduce errors. Non ecc memory introduces huge amounts of errors.

      (When I say 'high' and 'huge' I mean at volume. When you've got machines constantly thrashing a large disk pool 24 hours a day, 7 days a week, 52 weeks a year, for the lifetime of the system, even 1 error in a trillian ops becomes quite significant.)

      Now mind you, many high end SAN systems give you nice snapshotting, mirroring, etc features. But with ZFS you get the data integrity goodness, PLUS snapshotting, mirroring, pooling, etc, for free and open source, without having to pay 6-figures USD for SAN equipment.

    54. Re:maybe I should go and play around with this! by Znork · · Score: 1

      Mmm, that's rather interesting, and one of the aspects of ZFS I find most interesting. Do you know of any research on the data integrity issue or have any good links on it? I've seen the claims too, but frankly, I haven't noticed data integrity problems as a serious issue, even in a large enterprise environment. I mean, heck, even ECC errors are extremely rare occurances that usually indicate actual broken hardware (and having had Sun techs complain about cosmic radiation corrupting memory being the cause of system crashes I'm a bit leery of claims from Sun).

      Non-ECC memory is certainly a large possibility for errors, but again, in my experience even a few defective bits in a memory chip will cause a solid crash so fast you'd notice from the once-a-minute crashes if it was an even remotely common occurance. Not exactly the same thing, but them being 'silent' just hasn't been my experience with data corruption issues.

      It would be interesting to see data from a huge installation like Google on how often their data comes back silently corrupted from one of their replicas. Is it common? Does it happen at all in the real world?

      In theory, the sheer volumes of data suggest it should happen often. In practice it doesn't appear to be. So I don't know wether I should sha1sum all my data or not :). Or if Linux would do well with an optional dm-checksummer layer in the devicemapper.

    55. Re:maybe I should go and play around with this! by Allador · · Score: 1

      I dont have hard data, no. Wish I did.

      But I've been sold on it, from the papers I've read and the experiences I've had and have read/heard about others.

      For example, there are tons of anecdotes about people using ZFS and just being blown away by how often it reports data integrity errors.

      Another bit of evidence is how often files get silently corrupted, even if only a little bit. And the file fragments and broken metadata that shows up on all drives when you do thorough checking.

      You're right in that it doesnt have a big impact in the real world. I believe this is because many of our systems are built to be robust against small errors. Many file types can take the occasional bit flips without major loss of data (though dense binary structures like the windows registry are notoriously susceptible to even this).

      So no, I cant quantitatively support it. But it does resonate with what I've seen over the past 15 years doing this kind of work. Part of it is that systems are just inherently stable.

      Even the most stable systems will occasionally just degrade, because of corrupted files or hardware problems. Often leaving file system detritus behind after.

      It is rare, but its definitely non-zero.

    56. Re:maybe I should go and play around with this! by Directrix1 · · Score: 1

      Hahaha, I feel for you with the ReiserFS murder. I have suffered a similar mutilation.

      --
      Occam's razor is the blind faith in the natural selection of least resistance and in universal oversimplification. -- EF
  2. Nexenta by Enderandrew · · Score: 3, Insightful

    I'd try Nexenta, except I don't really want to use the Ubuntu repositories for my Linux packages. I'd prefer something with a good KDE desktop.

    I'd consider it for a web-server box to test how the kernel handles I/O.

    --
    http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    1. Re:Nexenta by jacquesm · · Score: 1

      There is also 'glusterfs', which has some pretty impressive specs.

    2. Re:Nexenta by SanityInAnarchy · · Score: 2, Insightful

      I'd prefer something with a good KDE desktop.

      What, exactly, don't you like about Kubuntu?

      Or is that not among the packages ported? Because to bootstrap from ubuntu-minimal to kubuntu is fairly easy.

      --
      Don't thank God, thank a doctor!
    3. Re:Nexenta by Enderandrew · · Score: 5, Interesting

      Someone asked me this question recently. And for the sixth time I answered with a laundry list of things I didn't like about it. Agian, I was modded Troll for stating I don't like Ubuntu/Kubuntu, and then people got all in a huff.

      Like I always say, it is marketed at a certain target audience, and it isn't me.

      I suggest that you try out a really good KDE desktop (Arch's KDEMod, Sabayon, openSUSE 11, etc) and the differences should be immediately apparent to you.

      As far as whether or not the KDE packages are available in Nexenta, I'm not sure actually.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    4. Re:Nexenta by Anonymous Coward · · Score: 0

      Someone asked me this question recently. And for the sixth time I answered with a laundry list of things I didn't like about it. Agian, I was modded Troll for stating I don't like Ubuntu/Kubuntu, and then people got all in a huff.

      Like I always say, it is marketed at a certain target audience, and it isn't me.

      I suggest that you try out a really good KDE desktop (Arch's KDEMod, Sabayon, openSUSE 11, etc) and the differences should be immediately apparent to you.

      As far as whether or not the KDE packages are available in Nexenta, I'm not sure actually.

      Kubuntu includes the "KDE Desktop Environment". So... I have no idea what the hell you are saying. I am assuming that you mean another distro with KDE?

    5. Re:Nexenta by KillerBob · · Score: 2, Informative

      You might also want to try Zenwalk... it's an XFCE desktop out of the box, but there's KDE packages in the repository... I don't actually have kdebase installed on my system (the only things from KDE that I actually use are Konqueror and Kopete, which are in the kdenetwork package, and work without kdebase), but it's actually a stock, unmodified, compiled from source package that, if it's anything like every other package on the system, is about as close to what the KDE devs want it to behave/look like that you'll find.

      I'm not saying that other systems, like arch, Sabayon, or openSUSE aren't great systems. But if you're interested in tinkering and trying things out, you may want to give Zen a try. :)

      --
      If you believe everything you read, you'd better not read. - Japanese proverb
    6. Re:Nexenta by SanityInAnarchy · · Score: 1

      I suggest that you try out a really good KDE desktop (Arch's KDEMod, Sabayon, openSUSE 11, etc) and the differences should be immediately apparent to you.

      Well, let me answer this by referencing two other posts. First, the sibling to mine:

      openSUSE is a good KDE, then you are a lost cause :-)

      And the reply to that:

      Kubuntu is a poor KDE desktop compared to most of the other majors. Ubuntu is solid, but Kubuntu isn't quite up to par. OpenSUSE is certainly far superior.

      Sorry, but no, it's not readily apparent, not to me, and not (apparently) to hummassa. If you want me to see the difference, you'll have to point it out.

      Agian, I was modded Troll for stating I don't like Ubuntu/Kubuntu, and then people got all in a huff.

      Care to point me to that post?

      It's very possible that the things you didn't like really were trollish, or that you weren't able of wording them in a non-trollish way.

      But you know what? Saying "it's just bad, and I'm afraid to say why" is the real troll here.

      --
      Don't thank God, thank a doctor!
    7. Re:Nexenta by Sancho · · Score: 1

      He appears to be a skilled troll. He refuses to point out anything that he actually doesn't like about Kubuntu--just that "it isn't marketed to me" and "it isn't up to par." No specifics, and he gets people to beg for specifics because he's playing the injured party. It's a masterful stroke, frankly.

    8. Re:Nexenta by ericrost · · Score: 1

      And its his typical drama troll whenever any story remotely about linux is posted, there's enderandrew within the first few posts mentioning how he doesn't like Ubuntu. Its nearly as bad as UbuntuDupe.

    9. Re:Nexenta by pjt33 · · Score: 1

      I think he must be saying that Kubuntu's default KDE settings aren't the ones he wants. I'm not sure what stops him exporting his settings from a "good KDE" and importing them to a "bad KDE".

    10. Re:Nexenta by Anarke_Incarnate · · Score: 3, Interesting

      Why is a differing opinion always a troll? Maybe he doesn't like it AND doesn't want to have to defend himself. Linux is about choice. He/She chose to not like it. Move along.

    11. Re:Nexenta by Mistshadow2k4 · · Score: 1

      I think he must be saying that Kubuntu's default KDE settings aren't the ones he wants. I'm not sure what stops him exporting his settings from a "good KDE" and importing them to a "bad KDE".

      Try another major KDE-based distro for a while and I think you will see the difference. IT is quite apparent that Kubuntu is second-rate to the Ubuntu dev team despite any protests to the contrary, so one major problem is that glitches and issues in Kubuntu get fixed slower than they do on Ubuntu. Its second-rate status shows in other ways too. I used Kubuntu for months, thought it was cool, but then I tried Saboyan. The latter is definitely KDE-based and the the difference shows.

      --
      I dream of a better world... one in which chickens can cross roads without their motives being questioned.
    12. Re:Nexenta by Enderandrew · · Score: 1

      It isn't a matter of defaults. I customize my desktop anyway, so I don't care that much about defaults. I'm talking features. openSUSE backports features, includes patches that don't make to upstream, and then develops features of their own.

      Kubuntu's packages aren't just vanilla, they are often packaged poorly, and are extremely buggy.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    13. Re:Nexenta by ericrost · · Score: 1

      When its posted over and over and over again, without prompting or sufficient reasoning ever backing it up, just to get a reaction, that is trolling.

    14. Re:Nexenta by Anarke_Incarnate · · Score: 1

      Get over yourself. When somebody suggests they try something they have already tried and did not like, they don't owe you anything at all. If you like it, have fun with it, file your bug reports, use your package manager, do your work. Thats how I behave with my F/OSS OS of choice.

    15. Re:Nexenta by Enderandrew · · Score: 1
      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    16. Re:Nexenta by Enderandrew · · Score: 2, Interesting

      No, I said I've had the argument several times before and don't see the need to repeat it. Just look for previous posts (like the ones I've listed above).

      Trolls get off on pissing people off. I'm trying to avoid an argument. How the hell does that make me a troll?

      If anything you're trying to bait me, and frankly I could care less.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    17. Re:Nexenta by Enderandrew · · Score: 1

      In the past four or five years I've posted in probably five or six threads where I've said I don't like Ubuntu. Maybe it is because I loathe that damned distro.

      And actually I try to stay away from that discussion. I repeatedly tell people I don't want to get into those details and rehash something unnecessarily.

      Your attempt to mischaracterize me here really falls flat. Over the years, amazingly enough I've said as much positive about Ubuntu (mainly centered on Shuttleworth's grasp of marketing) as I've said negative. I don't hang out on the Ubuntu forums and troll. Instead I go to other forums and try to provide them with helpful information.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    18. Re:Nexenta by Enderandrew · · Score: 1

      Thanks. :)

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    19. Re:Nexenta by bigstrat2003 · · Score: 5, Insightful

      This is a matter of taste. "I don't like it" is sufficient reasoning for this arena, like it or not. Furthermore, you have no reasonable basis to say he's saying this "just to get a reaction". He doesn't give what you consider to be good reasons for his opinion, so he's stating his opinion just to get a reaction. Erm, no, it doesn't work that way.

      People, not just on slashdot, but on internet forums in general, love to claim that those whose arguments they disagree with must be trolling. It's fucking pathetic, and is just a sign that these people can't handle an opposing point of view with any amount of dignity. Grow up already.

      --
      "16MB (fuck off, MiB fascists)" - The Mighty Buzzard
    20. Re:Nexenta by ericrost · · Score: 1

      And in the past week, you've posted in three if memory serves.

    21. Re:Nexenta by ericrost · · Score: 1

      I have no problem with his opinions, I've stayed out of the flamewars that have been created by his tone and vigor. I'm just pointing out that repetitive posting of an unbacked-up or "opinion based" argument (especially when its unrelated and very early on each thread) is trolling.

    22. Re:Nexenta by Anonymous Coward · · Score: 0

      Well, I must say I agree with this guy.
      Kubuntu's KDE has bugs like mixing up Konqueror tabs contents, defaults I don't like and aren't easy to fix without a different installation at hand (And I wonder if some changes are still easy to revert) like the braindead "Firefox-like" layout in Konqueror, Strigi popping out when I don't care a shit about it...
      And well, some things can be fixable, something aren't, yet I wonder why I should _fix_ something that is just fucking what upstream provides the way I want.
      I'm a long time Slackware user, yet I was looking for something different, less hacker oriented, and Kubuntu wasn't the choice .It lasted a week before I wiped it out; I still use it at work, though.
      I don't want to say nobody should use KUbuntu, and the fact finding distributions I can like besides Slackware says something about me (not OpenSuse, not KUbuntu, not Fedora, and I don't care about Gnome)...

    23. Re:Nexenta by Anonymous Coward · · Score: 0

      The desktop environment in Kubuntu sucks. Full stop.

      What don't I like about it? The bloat. The crappy icons and "art". The shitty keyboard shortcuts that conflict with control characters.

      And yes, there is an apparent difference between 8.04's KDE and 7-point-whatever-it-was', let alone Debian or Gentoo or an of many other distros. If you can't tell, you must not use computers.

    24. Re:Nexenta by SanityInAnarchy · · Score: 1

      they don't owe you anything at all.

      That's a strawman. I didn't say they owe me anything. Trolls don't owe me anything, either.

      But if you're just going to throw around blanket statements, without backing them up, you're just causing a lot of noise without a lot of signal. And to top it all off, the excuse for not backing it up is "I might be modded down!" (Oh no!)

      Which brings us to what you've done here, which is to drive it offtopic, while also saying inflammatory things. This wasn't a discussion of the moderation system, it was a discussion about various OSes and distros.

      --
      Don't thank God, thank a doctor!
    25. Re:Nexenta by SanityInAnarchy · · Score: 1

      Just look for previous posts (like the ones I've listed above).

      Actually, no, you haven't. That's one of the things I was asking for -- if you don't want to repeat yourself, link me to one of your oh-so-unfairly-modded posts. I promise I'll read it.

      Trolls get off on pissing people off. I'm trying to avoid an argument. How the hell does that make me a troll?

      If that's really your intention, you are doing so pretty unskillfully.

      If anything you're trying to bait me, and frankly I could care less.

      First: Your grammer sucks -- unless that is intended to mean that you really do care a lot, and that you therefore could care less.

      The phrase is supposed to be "I couldn't care less." That way, it actually makes sense.

      And second, you've posted several times on this thread -- in reply to me, and to others. Obviously, you care enough to keep posting. You're just too lazy (or scared? Of what?) to post a real argument, so you're having a meta-argument.

      --
      Don't thank God, thank a doctor!
    26. Re:Nexenta by Anarke_Incarnate · · Score: 1

      Why does anybody have to back up "I don't like it." Nobody owes you any reason for not liking it. They don't have to back it up. If you don't like his comment, don't read it. Move on and discuss OSes.

    27. Re:Nexenta by SanityInAnarchy · · Score: 1

      Kubuntu's KDE has bugs like mixing up Konqueror tabs contents, defaults I don't like and aren't easy to fix without a different installation at hand (And I wonder if some changes are still easy to revert) like the braindead "Firefox-like" layout in Konqueror, Strigi popping out when I don't care a shit about it...

      That's a start. I guess I'll have to look at how upstream does it differently, but for what it's worth, I don't mind Konqueror at all the way it's currently laid out.

      And "mixing up Konqueror tabs contents", I've honestly never seen. Not once.

      I wonder why I should _fix_ something that is just fucking what upstream provides the way I want.

      I've never been happy with what any upstream provides, or what any distro provides. I either try to find someplace that's just a bit off of the defaults so I can remember it, or I carry around my home folder for half a decade.

      There are reasons other than KDE that I like Ubuntu, and prefer it over vanilla Debian.

      Of course, if I was more of a C programmer, and I was particularly unhappy about what upstream provided, source-wise, I would probably still be on Gentoo. But I think my (very limited) Kernel-hacking days are over.

      Strigi popping out when I don't care a shit about it...

      Citation needed. I don't have Strigi installed at all.

      --
      Don't thank God, thank a doctor!
    28. Re:Nexenta by Enderandrew · · Score: 1
      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    29. Re:Nexenta by SanityInAnarchy · · Score: 1

      Seems to range from vague:

      Fire up a Knoppix CD, or Kubuntu and you'll basically see a vanilla KDE desktop. They don't customize the packages, install addition patches, or do anything....

      (but no mention of what patches you'd like to see, what features you're actually missing...) ...from vague, to uninformed:

      They've gone out of their way to patch into their kernel sources a measure to stop you from using proprietary modules.

      Well, no, your problem was much simpler:

      it would not load the ATI module at all. It gave me an error about how it could not load the module because it was missing a restricted modules .deb package.

      So install the restricted modules .deb package. It's trivial. Or does that not work with a custom kernel?

      In fact, you're compiling your own source. Why not download the ATI drivers straight from ATI? If you're compiling a kernel, it might make sense to consider all kernel drivers to be your responsibility.

      It's worth mentioning that Ubuntu's focus is not on people hardcore enough to compile their own kernel. I think it still does a pretty good job at this.

      Kubuntu got KDE 4 shoehorned in, with what was universally reviewed as the worst KDE 4 packages out there.

      I can only assume you're talking about Hardy, in which KDE 4 was not considered stable. The stable Kubuntu Hardy featured KDE 3.5.

      Shuttleworth said it was okay to have such a buggy major release because these things get patched.

      If true, he's probably making the same assumption I am -- that KDE 4 was not for users yet, and that these problems would probably be patched by the time it actually was -- maybe Intrepid? ...from vague and uninformed, to outdated:

      Again, I think this was going on two years ago.

      That is, two years ago that you tried it. Obviously, the KDE4 stuff wasn't 2 years ago -- but then, I'm guessing you didn't try it.

      So much for that comment. Would you like to point me to another?

      --
      Don't thank God, thank a doctor!
    30. Re:Nexenta by SanityInAnarchy · · Score: 1

      The bloat.

      Be specific.

      The crappy icons and "art".

      Matter of taste, and it's a skin. Don't like it? Install your own!

      The shitty keyboard shortcuts that conflict with control characters.

      That can all be customized with K->System Settings->Keyboard.

      If you can't tell, you must not use computers.

      Ad-Hominim. Thanks for the compliment.

      --
      Don't thank God, thank a doctor!
    31. Re:Nexenta by SanityInAnarchy · · Score: 1

      I think I may have to bring back my old sig. Here's what it said:

      If you're going to reply to this post, please read it first!

      --
      Don't thank God, thank a doctor!
    32. Re:Nexenta by SanityInAnarchy · · Score: 1

      Yeah, sorry about that. Found that after I posted. (I get email notifications, which I've been following in chronological order.)

      --
      Don't thank God, thank a doctor!
    33. Re:Nexenta by Enderandrew · · Score: 1

      Did you somehow miss me saying I don't want to rehash this over and over again?

      I tried it. I hated it. I think honesty anyone of any real discriminant taste who tries several distros is likely to find that most distros are better than Ubuntu in many different ways.

      So install the restricted modules .deb package. It's trivial. Or does that not work with a custom kernel?

      Perhaps you don't understand. The restricted module .deb package is for a specific kernel. The version made for 2.6.25.15-1 (or whatever) only works with that specific kernel. If you roll your own kernel, that package does not exist. I did download the ATI driver from the ATI website, and that installed fine. However, the ATI module refused to load. The kernel threw an error saying it wouldn't load the proprietary driver without the restricted module package. I checked the source repositories and found no source to make that package myself.

      I asked repeatedly how I might make that package, and was told not to attempt to compile anything. I should be happy with defaults.

      I explained the default didn't work. I was given no help to fix the driver issue, so I was attempting to fix it myself, and was chastised for doing so.

      Even worse, the fact that the kernel has an added measure preventing you from loading proprietary modules was the final blow. Why would the Ubuntu devs add that extra measure into their kernel sources?

      Effectively it kills any chance to rolling your own kernel, and using proprietary drivers at the same time.

      Again, like many other areas I ran into with the distro, it locks you in and removes choice.

      It is exactly as I have said repeatedly. The distro is aimed at a certain market, and I am not that market.

      I can only assume you're talking about Hardy, in which KDE 4 was not considered stable. The stable Kubuntu Hardy featured KDE 3.5.

      Yes, and the stock KDE 3.5 packages in Kubuntu are still pretty terrible. But the KDE 4.0.4 packages were even worse. They built and packaged them incorrectly, to the point that they got called out by the KDE devs.

      The Hardy release had plenty of massive problems aside from the KDE 4 version, which isn't how you issue a LTS release of a major distro that is trying to promote itself as a top-tier product, and enterprise worthy.

      Shuttleworth always says that Ubuntu "just works", yet the Hardy release was a very bad example of that.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    34. Re:Nexenta by virtualXTC · · Score: 1

      WOW! I did as you said and tried openSUSE 11, and must say it's quite a bit more impressive than any debian based distro. For years I've resisted RPM distro's after experiencing RPM hell vs APT magic powers. Last time I tried SUSE was right after the Novell acquisition, and there were definitely some holes; SimplyMEPIS was far superior. Unfortunately MEPIS decided to peg it self to Ubuntu (a big mistake) and I've been searching for a new distribution since. Mint and DreamLinux had a similar bleeding edge feel, but Mint sacrifices much speed for eye candy, and let KDE development lag behind their main branches. Thanks for pointing me toward a better distro.

    35. Re:Nexenta by SanityInAnarchy · · Score: 1

      Did you somehow miss me saying I don't want to rehash this over and over again?

      And yet, here you are. If you really didn't, there's a solution: Don't push reply.

      Instead, you wrote quite a lot of rehashing here.

      Again, like many other areas I ran into with the distro, it locks you in and removes choice.

      I haven't seen anything in any of your other comments which lock you in to anything, or remove any choice, except this one, which I'm still not convinced of.

      In fact, maybe 30 seconds on Google, and here you go. Notice the section called "Rebuilding linux-restricted-modules" -- which isn't even required -- but if you really need to, it links to yet more detailed instructions on how to build them.

      Search terms were: custom ubuntu kernel howto. That was the third link down. Which raises the question: Are you even trying?

      Shuttleworth always says that Ubuntu "just works", yet the Hardy release was a very bad example of that.

      Again, citation needed. Because it did just work for me.

      What about it doesn't just work?

      --
      Don't thank God, thank a doctor!
    36. Re:Nexenta by DMUTPeregrine · · Score: 1

      Because it's open source, and we want to fix it. If we don't know what's bad, we can't submit bug reports and patches. Thus, on this forum, it's more likely a troll than a legitimate complaint.

      --
      Not a sentence!
    37. Re:Nexenta by cbart387 · · Score: 1

      Just as an FYI, Mepis is now debian driven. In fact, it uses the Debian stable repositories except for a repository that has backported packages and other custom stuff. The only reason I have moved away from it, is the oldness of stable.

      If you're looking for something more bleeding age I would suggest checking out arch. You'll have to do a little work to get it up to the usability that OpenSUSE is because it makes you build your own system. On the plus side though, it's a rolling release (so you don't have to reinstall it again) and it's comparatively fast since you choose what daemons to run.

      --
      Lack of planning on your part does not constitute an emergency on mine.
    38. Re:Nexenta by Anonymous Coward · · Score: 0

      I suggest that you try out a really good KDE desktop (Arch's KDEMod, Sabayon, openSUSE 11, etc) and the differences should be immediately apparent to you.

      openSUSE has done wonderful things with KDE. The layout of the "start" (or whatever) menu is incredibly clean and easy to navigate. The whole thing is uniform and normalized. I just wish it wasn't connected to Novell.

      I like parts of u/kubuntu. I'm not crazy about the software layout in their GUIs though. IMO there are too many menus and they are WAY overpopulated. It doesn't effect me that much as I (still) use WindowMaker, but watching new users try to sift their way through all that stuff is painful.

      I haven't encountered Sabayon and haven't tried Arch yet. I'll have to give them a whirl one of these days.

    39. Re:Nexenta by virtualXTC · · Score: 1

      WTF, MEPIS is back to debian again? not that I'm complaining.... I'll keep arch in mind next time I find a must-have app (my guess: a KDE plasma bar that actually hides) that hasn't propagated to my current distro and is too hard to compile myself. 1.5 days in, an it seems openSUSE has everything I'm looking for and even some stuff I didn't know I was missing. Its tight integration with various visualization managers is going to be hard to shake. Plus if I really get paranoid, I might check out AppArmor.

      Alas, I'll likely always run a at least one Debian based distro such as Gnusense or Gobuntu (I dual boot - until android is working on my HTC phone and distance ed no longer requires flash and WMV) as I'm a purist at heart.

  3. From an experienced Admin's perspective by Anonymous Coward · · Score: 0, Insightful

    Stay away from Solaris, unless you have to use it.

    That being said, I have never seen a large shop without 'some' SUN/Solaris machines.

    Solaris isn't bad. There are just 'linux' OS's that do most of the same jobs w/o you getting tied to Solaris.

    1. Re:From an experienced Admin's perspective by igb · · Score: 5, Informative
      I run a mixed Linux and Solaris shop, but having replaced some Solaris boxes with Linux we're swinging massively back to Solaris 10.
      • ZFS, of course.
      • SMF. Being able to start services in a dependency tree is excellent if you have a multi-processor machine. And having services self-heal, including restarting any dependencies, is good for things like mail servers that use a lot of flakey milters.
      • FMA. Hardware self-healing (admittedly, this is essentially Sun hardware only, and in my experience better on Sparc than on AMD) is good.
      • Zones. Because sometimes full-blown virtualisation is too much like hard work.
      • Binary compatibility. I've got some SunOS 4.1.1 binaries from 1989, for which the source is long lost, running fine.
      • There's probably a Linux equivalent of rcapd, to limit the physical memory use of particular groups of processes, but I've never found one.
      • There's probably a Linux equivalent of processor sets, CPU shares and the Fair Shares Scheduler, but again I've never found one.

      Horses for courses, but Solaris has much to offer even for shops that aren't traditionally tied to Sun. Hell, even my private ``1U box in someone else's datacentre'' server for my family is now a Solaris machine.

      ian

    2. Re:From an experienced Admin's perspective by GuyverDH · · Score: 3, Interesting

      Ahem...

      Actually the reverse is true...

      Since release 10, Solaris has been pretty well stomping the competition in price, performance and throughput. With Solaris supporting pretty much every type of virtualization (including some not offered anywhere else), it's hard to beat.

      Solaris as well as OpenSolaris are free, you can download and use either flavor with no cash outlay. Want support? It's cheaper to buy Solaris support from Sun than to buy Linux support from RedHat.

      There's no *tying* with Solaris, it's all about choice. I personally choose Solaris over Linux for pretty much any task.

      --
      Who is general failure, and why is he reading my hard drive?
    3. Re:From an experienced Admin's perspective by Marillion · · Score: 3, Insightful

      It all depends on the skill-set your admins already have. If you have a shop of 100% Linux admins with no Solaris experience, stay away. If your shop already has some Solaris machines on Sparc, go for it - although you should double check the license.

      From my own perspective, I've invested several hours getting it running. Granted, I was running the 200805 OpenSolaris installed on ZFS which had some bugs in the boot process which left my system unbootable a few times. Some follow up releases fixed those problems. But as a guy who's been using Linux since 1993, old habits are hard to break.

      --
      This is a boring sig
    4. Re:From an experienced Admin's perspective by Billly+Gates · · Score: 3, Interesting

      Stay away?

      What if you need real uptime with a load of 80 on a 32 cpu system? Can Linux handle the load and have years of uptime?

      Solaris just works and its made for servers. Linux seems always beta quality with its cutting edgness and is desktop oriented. I would not trust my job to it unless its Debian or RHES which costs $$$ as cutting edge features are not needed on a mission critical server. Solaris scales far better than any BSD or Linux distro out there.

      Ask any *real* Unix admin who uses both and more than likely they will say Linux is great for small jobs but Solaris is king for anything else.

      Oh and the article discussed a scarcity of third party apps. I found the opposite as most server ERP and database apps are on Solaris than Linux.

    5. Re:From an experienced Admin's perspective by Skrynesaver · · Score: 2, Interesting
      I work with both Solaris and Linux on a daily basis. I'm looking forward to Solaris features with a full GNU userland as some times the Solaris CLI is a bit clumsy by comparison (Yes i know about sunfreeware.com but our clients don't necessarily have it installed on production boxes).

      However I think this is probably a response to something I've noticed of late, in Asia and South America we don't sell support for Solaris installs any more, they've all moved to Linux, cheaper hardware, a pool of interested young (and therefore cheap) admins, and of course our wonderful software is available on Linux ;)

      While Europe and to a lesser extent the US are almost exclusively Solaris (the odd godforsaken HP-UX or AIX box as well to keep me interested) the emerging markets, where the growth is, are moving en-mass to Linux/Open source.

      --
      "Linux is for noobs"-The new MS fud strategy
    6. Re:From an experienced Admin's perspective by zappepcs · · Score: 1

      Well, there are other more technical reasons for considering Solaris. I'm not sure how this applies exactly in detail to the OpenSolaris but Solaris handles threading a little differently than Linux and a few other minor things that can make a big difference depending on your application. When you get down to the fine technical details of each OS, there are differences that can make or break your application's efficiencies.

      For the desktop, perhaps that kind of analysis is not needed but if you are planning on handling 10,000 bi-directional transactions per second, application performance is a big issue.

      I am forced to use WinXP, Solaris 5.8-10, CentOS 5.x, Fedora Core 8/9, RedHat, and Ubuntu in my daily life. When it comes to desktops, it's more or less a choice of personal style, no more difficult than choosing a desktop background picture. OOorg and Mozilla have made them all function the same for me. My favorite text editor comes in win/nix flavors also. For the most part, they all function the same. Hard core performance is where they begin to vary a lot IMO.

    7. Re:From an experienced Admin's perspective by Enderandrew · · Score: 2, Interesting

      What about hardware support? I keep hearing that the openSolaris kernel just frankly doesn't have many drivers. If I can't install it on my hardware, it isn't doing me any good.

      Also, I'd really like to see some basic benchmarks between the kernels. People benchmark the BSD kernel against the Linux kernel on IO, networking, etc.

      Show me some quantifiable numbers on openSolaris.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    8. Re:From an experienced Admin's perspective by Timothy+Brownawell · · Score: 1

      There's probably a Linux equivalent of processor sets, CPU shares and the Fair Shares Scheduler, but again I've never found one.

      Newer kernels allow you to put processes into a tree and assign priorities at each level, or assign priorities per user. Look for "CFS" or "Completely Fair Scheduler" and "group scheduling" or "fair user scheduling". Not sure how exactly this compares, my only run-in with it has been that various system cron jobs that used to run at nice 19 don't act like they're at nice 19 any more.

    9. Re:From an experienced Admin's perspective by Capt+James+McCarthy · · Score: 2, Insightful

      What, no mention of dtrace? Now that's been an excellent part of the Sol10/OpenSol movement IMO.

      --
      There are no loopholes. It's either legal or it's not.
    10. Re:From an experienced Admin's perspective by igb · · Score: 2, Interesting

      dtrace is great, but actually my experience as an administrator is that I use it less than I expect, because the kernel `just works'. I use it to attack badly behaved applications, but I've not used it for tuning anything like as much as I thought I would.

    11. Re:From an experienced Admin's perspective by TheRaven64 · · Score: 5, Insightful

      It doesn't matter how many drivers any given kernel supports. All that matters is if it has drivers for the hardware you want to run it on. If you're buying a server then you will typically buy one which comes with support for the OS you want to run and so you won't encounter driver difficulties (although you might pay a bit more).

      --
      I am TheRaven on Soylent News
    12. Re:From an experienced Admin's perspective by serviscope_minor · · Score: 4, Informative

      Solaris scales far better than any BSD or Linux distro out there.

      Yeah, you know, the roadrunner team would like a word with you, as would pretty much everyone in the Top 500. For some business loads Solaris scales better. But the claim the "it scales far better" in general is as absurd as it is patently untrue.

      Ask any *real* Unix admin who uses both and more than likely they will say Linux is great for small jobs but Solaris is king for anything else.

      Ah, and no true scotsman^W UNIX admin would run a supercomputer, right?

      --
      SJW n. One who posts facts.
    13. Re:From an experienced Admin's perspective by blane.bramble · · Score: 3, Informative

      One of our Linux servers regularly copes with a load in excess of 100. Things slow down, but nothing breaks.

    14. Re:From an experienced Admin's perspective by TheRaven64 · · Score: 2, Informative

      Solaris handles threading a little differently than Linux

      It used to be that Solaris used an N:M model while Linux used an N:1 model. Now both use a 1:1 model. There are lots of reasons for this (Matt Dillon gave a really detailed description when explaining why Dragonfly BSD went 1:1 instead of N:M). Basically, it boils down to the fact that debugging threaded C code is such a bitch that people tend not to use high levels of parallelism in C code (which is where N:M really shines). If a language has better support for parallelism then it is easy build an N:M model on top of a 1:1 model (this is what Erlang does, and I believe Java does as well in some versions).

      --
      I am TheRaven on Soylent News
    15. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 1, Informative

      "Since release 10, Solaris has been pretty well stomping the competition in price, performance and throughput. With Solaris supporting pretty much every type of virtualization (including some not offered anywhere else), it's hard to beat."

      I think you have been drinking a bit too much of the marketing cool-aid. Well, you haven't ever been in my shop. I'm a Solaris guy - Admin, instructor, blah... These days our Sun boxes just don't cut it. They are being out performed by *AIX* machines!!! (yes, solaris 10)

      "It's cheaper to buy Solaris support from Sun than to buy Linux support from RedHat."

      You had better not rely on either. Sun support SUCKS!!!! (at least since 2003) RedHat doesn't do much better either. You need to have experienced admins. Platinum support means nothing when Sun doesn't know what to do.

      "There's no *tying* with Solaris"

      Oh yes there is. Once you start developing and using a platform in production, you are tied to it. Sure, you can always move off of it - at least in your utopia, not reality very easily.

    16. Re:From an experienced Admin's perspective by Chirs · · Score: 1

      Some suggestions for linux users. (Note, I'm not a solaris user, so I'm sure there are some differences between features.)

      --OpenVZ or vservers should give something similar to zones

      --They're working on memory resource controllers for the "cgroups" functionality. There have been out-of-tree patches for years now, but I think this may be included already, or will go into 2.6.27.

      --cpusets and processor affinity masks are available, cpu shares are available as part of the cpu resource controller for cgroups, and the "completely fair scheduler" went in around 2.6.23 or so.

    17. Re:From an experienced Admin's perspective by Enry · · Score: 2, Informative

      Stay away?

      What if you need real uptime with a load of 80 on a 32 cpu system? Can Linux handle the load and have years of uptime?

      Yea, stay away. If you have a load of 80 on a 32 CPU system, you didn't design the hardware or software correctly.

      Solaris just works and its made for servers. Linux seems always beta quality with its cutting edgness and is desktop oriented. I would not trust my job to it unless its Debian or RHES which costs $$$ as cutting edge features are not needed on a mission critical server. Solaris scales far better than any BSD or Linux distro out there.

      Debian doesn't cost anything, and there's always CentOS if you want the RHEL reliability without the cost.

      Ask any *real* Unix admin who uses both and more than likely they will say Linux is great for small jobs but Solaris is king for anything else.

      Yea, these were the same Unix admins who used to ask me if I installed the latest kernel patch while they were still using sendmail (and patching it about as frequently). I didn't put a lot of faith in their opinion.

      Oh and the article discussed a scarcity of third party apps. I found the opposite as most server ERP and database apps are on Solaris than Linux.

      For the big big things, probably. Oracle? Works perfectly fine. I'm building a RAC now using commodity hardware that will probably be 1/3 the price of what it would cost to get something from Sun.

      To be fair, I haven't used much Sun equipment (hardware or OS) in the past 6 years or so. There's a number of things they get right, like the Open Firmware. But from an OS and maintenance perspective, does Sun still have patch clusters? Do I have to head over to SunFreeware.com to get useful applications installed? I can provision a Linux server literally in a few minutes, but it would take the better part of a day to get Solaris set up (have to remember to disable telnet, find the latest patch cluster, reboot, install gcc and other apps). Bleah.

    18. Re:From an experienced Admin's perspective by Abattoir · · Score: 1, Insightful

      Uptime?

      Uptime is so 1997.

      Anyone who needs serious uptime of "years" will have a high availability cluster implementation.

      Or they'll use a *real* platform known for reliability like a mainframe, not some toy running on purple plastic hardware.

    19. Re:From an experienced Admin's perspective by GuyverDH · · Score: 4, Informative

      Having been a UNIX admin for 23 years and Solaris for 10 years, I'm not sure what you're drinking, but I'm staying away from it.

      Solaris support has rocked. We've never had an issue that Sun hasn't been able to solve, and yes, we've thrown them some curves (and sliders for that matter). IBM's support has told us on multiple occasions to re-install the system as a fix for a problem. RedHat we've stumped more often than not. HP? Well - they still can't figure out how to handle more than 8 luns per target for scsi (as well as fibre)...

      Solaris performance has been fantastic - outperforming Linux, AIX, HP-UX on modern equipment.

      We've migrated workloads to and from Solaris - no big deal - as long as you know what you're doing.
      (Our misguided DBA's started migrating from old SunOS 5.8 boxes to Linux - and are now migrating back.)

      If you use tools that are available on multiple platforms, migrating isn't all that tough.

      If you are developing native language apps, porting isn't terribly difficult although finding workarounds for pesky native quirks is troublesome at times.

      So I guess it depends on what you call "experienced"...

      --
      Who is general failure, and why is he reading my hard drive?
    20. Re:From an experienced Admin's perspective by Cheeze · · Score: 1, Interesting

      ZFS - Are you really using your server for data storage? SAN or NAS should be a better option depending on your price point
      SMF - while nice, i have experienced many different kinds of errors. If one of the dependencies has a problem, the chain breaks and it is a pain to discover the problem.
      FMA - if hardware is broken, i would rather the machine be fully broken and out of service. Running on degraded hardware is too much of a risk. if a few bits get switched or some data is not written correctly, you could corrupt data.
      Zones - I still have yet to see a reason to use this except for dedicated virtualization servers.
      binary compatibility - if you are running custom code without the source, sounds like you have a setup for failure.
      rcapd - ulimit can do this per process, and there are also multiple 3rd party open source resource limiters.
      processor sets, cpu shares, fair share scheduler - Yes, there are.

      I've been in many different places with Solaris, Linux, and a few other random UNIX environments. In almost every case, the Solaris and other random unix environments could be replaced with Linux at 1/10th the cost.

      I manage some HPUX servers right now and just the hardware maintenance on each of the servers is more than a few brand new Linux servers each year.

      --
      Why read the article when I can just make up a snap judgement?
    21. Re:From an experienced Admin's perspective by Danathar · · Score: 1, Informative

      I guess you have never worked in supercomputing....where Linux is just about what everything runs.

    22. Re:From an experienced Admin's perspective by BrokenHalo · · Score: 1

      Ask any *real* Unix admin who uses both and more than likely they will say Linux is great for small jobs but Solaris is king for anything else.

      OK, you are free to doubt my reality (I am a 45-year-old old-timer when it comes to both Unix and Linux), but Linux isn't all about bleeding-edginess.

      I would stack a nice solid Slackware distro against Solaris or BSD and expect sound results. Until very recently I used this for my desktop system too, but since my preference is for Gnome (for which development of the formerly excellent Dropline distribution appears to have stalled) I have had to go shopping.

      I tried Ubuntu (again) and hated it (again), tried Gentoo for 3 weeks and not quite hated it, but found it frustrating enough to look elsewhere. Currently I'm playing with Arch linux, which is looking quite promising. Not quite as quick to set up as Slackware, but with all of its other (i.e. KISS) advantages.

    23. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 1, Informative

      Linux seems always beta quality with its cutting edgness and is desktop oriented.... Ask any *real* Unix admin who uses both and more than likely they will say Linux is great for small jobs but Solaris is king for anything else.

      Get your head outta your ass. When even Wall Street firms use Linux on their grids/farms to crunch numbers using complex modeling on very very critical jobs ($$$$), reliability, scalability along with cutting-edgeness - everything comes into picture. And they are replacing their Sun boxes for last 3 years.
      I have worked on these (as a developer, mind you), so at least I can attest Linux's readiness with its 'beta' quality. So, take your bullshit somewhere else.

    24. Re:From an experienced Admin's perspective by lokedhs · · Score: 1

      You should try using it to debug applications. I use it a lot for this, and it's amazing. It's basically like a programmable debugger that can monitor a running process without really affecting its performance.

    25. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 1, Insightful

      I guess google have no "real" admins that look after 200,000 Linux servers. Obviously amateurs compared to monkeys like yourself. If you knew what you were doing, why would you be a lowly admin?

    26. Re:From an experienced Admin's perspective by Gerald · · Score: 1

      You forgot 'lsof -o'.

    27. Re:From an experienced Admin's perspective by igb · · Score: 1

      On the ZFS point, just what filesystem do you think I put on my 20TB SAN?

    28. Re:From an experienced Admin's perspective by gclef · · Score: 2, Interesting

      I'm not the original poster, but I agree with him wholeheartedly. In my experience, tasks that can be easily parallelized work well in Solaris (web servers, polling servers, etc). However, tasks that are serial in nature (dealing with a stream of events like IDS or syslog) work *horribly* on Solaris.

      When we moved some of our log parsing from Solaris/SPARC hardware to similarly priced Debian/x86 hardware, we expected a 3x improvement in performance just due to the CPU...we actually saw a 10x improvement in performance. We attributed this largely to Solaris' aggressive reservation of CPU cycles for other threads...even when we only had one.

    29. Re:From an experienced Admin's perspective by Kennon · · Score: 2, Insightful
      Uhh...try doing a little research asshat...

      What if you need real uptime with a load of 80 on a 32 cpu system? Can Linux handle the load and have years of uptime?

      Over 85% of the top 500 super computers in the world run Linux. http://www.top500.org/ as best I can tell almost none run Solaris as most of the Unix is AIX. So all you "Linux's uptime, stability and processing power sucks compared to Unix" old ass fanboys go back to your clubhouse and cry.

      Ask any *real* Unix admin who uses both and more than likely they will say Linux is great for small jobs but Solaris is king for anything else.

      Wow...if you are a *real* Unix admin it is no wonder Linux came along and is so successful.

      --
      "All those moments, will be lost in time...like tears in rain..."
    30. Re:From an experienced Admin's perspective by phantomfive · · Score: 1

      # There's probably a Linux equivalent of rcapd, to limit the physical memory use of particular groups of processes, but I've never found one. # There's probably a Linux equivalent of processor sets, CPU shares and the Fair Shares Scheduler, but again I've never found one.

      Just starting to get added in the latest kernels.

      --
      Qxe4
    31. Re:From an experienced Admin's perspective by Wdomburg · · Score: 2, Informative

      A load of 80 on a 32 cpu system sounds like a poorly architected solution to me. :)

      As for the rest, give me a break. One of the benefits Linux is that if you want cutting edgeness and desktop goodies you can have them, but if you're looking for stability and vendor support you can have that too. And it doesn't mean spending a ton of money either - RHEL is relatively cheap and Debian is free (no idea what you're smoking there), as are a number of other options (CentOS, Ubuntu LTS, etc).

      We run hundreds of servers on Linux servicing millions of subscribers and have absolutely no stability issues whatsoever. Machines occasionally go down due to hardware faults, power incidents or kernel upgrades, but only a handful of kernel related failures over the years. We've actually had more failures with our Sparc/Solaris machines, generally exhibiting as spontaneous reboots.

    32. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      Ugh, *real* UNIX admins spell UNIX in uppercase, and run AIX on Power6 P595's, the fastest commercially available machines in the world.

    33. Re:From an experienced Admin's perspective by Wdomburg · · Score: 2, Insightful

      Not to mention that an uptime of years generally means someone hasn't been keeping their system patched properly. :)

    34. Re:From an experienced Admin's perspective by BrainInAJar · · Score: 1

      Yeah, Linux has CFS and all... but does it have it /AT THE SAME TIME/ as other schedulers?

      Solaris has 5 schedulers and they can all be running at the same time ( they all have different priorities ), no need to pick on boot or compile which algo. you want to use

    35. Re:From an experienced Admin's perspective by BrainInAJar · · Score: 4, Insightful

      Couple points:

      "ZFS - Are you really using your server for data storage? SAN or NAS should be a better option depending on your price point"
      Why not set up a server for data storage? Then you get all the ZFS checksum/auto-heal/snapshot goodness ?

      "Zones - I still have yet to see a reason to use this except for dedicated virtualization servers."
      Zones are so cheap, I run every single service in a zone so that they can be migrated between machines, any dependencies can be contained, etc. If you haven't seen a use for them it's because you haven't ever used them.

      "rcapd - ulimit can do this per process, and there are also multiple 3rd party open source resource limiters."
      And yanking the ram stick can do it per-machine. How coarse grained do you want to go before you look like a fool?

      In almost every case, the Solaris and other random unix environments could be replaced with Linux at 1/10th the cost."
      Solaris is free. Support is 1/3 the price of RHEL. It runs on cheap Dell/Supermicro whiteboxes.

    36. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      Yay! ... AC bites the dust ... I pronounce Mr. GuyverDH the winner in this duel.

      Oh! by the way, I am a different AC :) not the one above.

    37. Re:From an experienced Admin's perspective by BrainInAJar · · Score: 1

      "Yea, stay away. If you have a load of 80 on a 32 CPU system, you didn't design the hardware or software correctly."

      "it would take the better part of a day to get Solaris set up (have to remember to disable telnet, find the latest patch cluster, reboot, install gcc and other apps). Bleah."
      All that stufff is included by default now. You being inexperienced is not a fault of the OS

    38. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      Linux doesn't scale very well SMP wise. Any OS on individual boxes can be ganged together with interconnects, but that does not scale compared to a big SMP.

      Also, if you have a big dataset that needs to be seen by all processors, a big SMP is the only way to go.

      Yes, I'm on the top 500 list, used Linux on SGIs with NUMAlink and all that. I know what I'm talking about.

      Linus at this time doesn't care about scaling, because it would take a major reworking of about everything under the hood. Now that everything is SMP (and going to NUMA), and there seems to be no stopping it (Intel talks about 96 cores on one chip all the time), it might be time to focus on getting linux to scale.

    39. Re:From an experienced Admin's perspective by BrainInAJar · · Score: 2, Insightful

      Heavily modified Linux. Ship of Theseus style Linux ( Is it still linux if only the interfaces are the same? )

      Unmodified Solaris scales to > 512 CPU's almost linear

    40. Re:From an experienced Admin's perspective by digitalhermit · · Score: 2, Informative

      Funny you should mention the "desktop orientation" of Linux, considering that any recent Solaris puts on a shiny Gnome suit...

      Solaris scales far better than any BSD or Linux distro out there.

      Ask any *real* Unix admin who uses both and more than likely they will say Linux is great for small jobs but Solaris is king for anything else.

      OK, I'll bite. I've been admin'ing Unix for 12 years, and a user for 20. I've maintained everything from an UltraSparc workstation to E10Ks. Lately (past five years) I've been doing AIX, maintaining over 150 OS instances.

      For a test project I've been researching Linux and Solaris for an Oracle RAC installation. I will tell you that, hands down, the Linux network stack is faster. Raw i/o is close, but Linux wins small file writes and reads by a significant margin. BTW, the Sun.com site has a paper comparing Linux filesystems versus UFS, but take it with a grain of salt (look at the machine they use).

      Now let's talk about scaling...

      Up to the 8 processor machine that I tested (which is a *small* PC system), Linux continues to scale close to linearly for the Oracle/TomCat/Apache workload, as does Solaris. Beyond this I understand things can change, but so be it, that's not the platform I'm needing.

      Now certainly you can build a workload requirement that will put Solaris on top, but in the vast majority of installations Linux will do just fine.

      Of course there are other factors. RedHat doesn't have the most sterling support (compare it to IBM for example), but I can buy support through IBM (or Oracle or Sun for that matter) if I wanted. This may be the reason for a shop to go with Sun, but I wanted to correct the idea that Linux is not as technically valid as Solaris.

    41. Re:From an experienced Admin's perspective by C-Shalom · · Score: 1

      "ZFS - Are you really using your server for data storage? SAN or NAS should be a better option depending on your price point"

      Why not set up a server for data storage? Then you get all the ZFS checksum/auto-heal/snapshot goodness ?

      That's why there is Thumper (aka Sun Fire X4500) that can hold 48TB in 4RU.
      Considering that, it makes a 20TB SAN seem small.

    42. Re:From an experienced Admin's perspective by illumin8 · · Score: 4, Interesting

      Solaris support has rocked. We've never had an issue that Sun hasn't been able to solve, and yes, we've thrown them some curves (and sliders for that matter). IBM's support has told us on multiple occasions to re-install the system as a fix for a problem. RedHat we've stumped more often than not. HP? Well - they still can't figure out how to handle more than 8 luns per target for scsi (as well as fibre)...

      Ok, I'll call you on this one. I'm a SCSA (Solaris Certified System Administrator) and a former Sun SSE. I've worked with Solaris systems going back to 1996 on original Sparcstations (not even Ultrasparcs). I've also worked on Enterprise 10000, 15000, and 25000. We also have a smattering of Sun Fire X4600s, the new AMD Opteron boxes.

      I tested Solaris 10 and Red Hat Enterprise Linux 4 (64-bit) on the exact same hardware (X4600), and you know what? Red Hat Enterprise Linux 4 performed better on massive storage I/O than Solaris 10. I have the Oracle ORION benchmarks to prove it. We have over 50 LUNs carved from an HP EVA 8100 and presented to these X4600s, on 4x 4gb HBAs per server. They run Oracle RAC, have 4x quad core AMD Opterons, and Red Hat Enterprise Linux.

      Sorry, but Solaris used to be a good OS back in the 90s. They have fallen so far behind it's not even funny. The reality now is that I can run Red Hat and Oracle on a 32 core AMD Opteron box with a hundred LUNs on a fibre channel SAN and it outperforms Solaris now. ZFS is nice, but we use ASM (automated storage management) for Oracle anyway, so ZFS is unnecessary.

      Solaris has unfortunately fallen far behind the performance curve, and I doubt they can ever catch up. Your BS about HP not supporting more than 8 LUNs per target is absolutely BS. I can do hundreds of LUNs, and I have systems like that in production.

      On support, they all suck. Red Hat, HP, Sun, every one of them sucks. They have all been chasing the bottom and if it ever gets to the point where I'm stumped, they're going to be stumped as well.

      --
      "When the president does it, that means it's not illegal." - Richard M. Nixon
    43. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      it would take the better part of a day to get Solaris set up (have to remember to disable telnet, find the latest patch cluster, reboot, install gcc and other apps). Bleah.

      Apparently you've never heard of Jumpstart because if you were going through all that in a data center environment, the SA's would through you to the curb for wasting time.

    44. Re:From an experienced Admin's perspective by illumin8 · · Score: 2, Informative

      What if you need real uptime with a load of 80 on a 32 cpu system? Can Linux handle the load and have years of uptime?

      Solaris just works and its made for servers. Linux seems always beta quality with its cutting edgness and is desktop oriented. I would not trust my job to it unless its Debian or RHES which costs $$$ as cutting edge features are not needed on a mission critical server. Solaris scales far better than any BSD or Linux distro out there.

      Absolute bullshit. I run Red Hat Enterprise Linux (64-bit) on a 4 node AMD Opteron cluster running Oracle RAC. Each server has 32 cores, 128GB of memory, and 8x 4gb HBAs, connected to an HP EVA 8100 SAN. I also tested Solaris 10 on this same hardware, and you know what? It outperforms Solaris 10 on the same hardware. I have the Oracle ORION benchmarks to prove it.

      There was a time in the late 90s when Solaris was a far superior OS to Linux for use in the data center. I'm an SCSA, I know, because I started as a Solaris admin long before I worked on Linux. The reality is that now, Linux outperforms Solaris for I/O intensive applications like Oracle database. Why do you think Oracle themselves migrated to Linux a while back? Solaris and Sun have been losing their best and brightest engineers for a long time now, and the quality of their OS shows. It's getting dated. Sure, new features like ZFS are cool, but the core of the OS, where it really counts, hasn't been updated enough to take advantage of the large memory and CPU core footprint that new commodity servers have.

      --
      "When the president does it, that means it's not illegal." - Richard M. Nixon
    45. Re:From an experienced Admin's perspective by Anarke_Incarnate · · Score: 1

      You could also try SLES. Their init scripts are not the SYS V style, and do have dependency abilities. You don't get ZFS, but I have found that it is a finicky beast and needs lots of kernel tuning for certain roles (Databases especially, unless you don't care about system allocation and performance).
      There are Linux virtualization tools that work like zones.
      However, do what works for you. Just figured you could save some migration pains.

    46. Re:From an experienced Admin's perspective by Anarke_Incarnate · · Score: 1

      You should not put a filesystem on the SAN but rather the presented LUN. Also, if you are using a battery backed up SAN you should really turn fsync off for the ZFS filesystem and get some performance back. Also, research ARC buffer sizes as well as tuning the read block and write block size for your applications.

    47. Re:From an experienced Admin's perspective by Anarke_Incarnate · · Score: 1
      size in capacity is only the beginning of what makes a SAN a superior investment. I would love to see how many 4GB fiber connections a x4500 can maintain with decent throughput. They compete in very few overlapping markets. If you tried to run a Gemstone /S database on a x4500 I would laugh.

      That being said, I love the x4600s for running VMware/Linux.

    48. Re:From an experienced Admin's perspective by GuyverDH · · Score: 0

      Hmmm - yes - we have lots of luns - however, they increment the target on the disk in order to go beyond 8 - btw - this is on HP-UX 11i - we haven't gone further or up as the legacy software took a lot of tweaking to get it to run at this level.

      Now - the tape driver, they didn't do this "target" increment trick that they implemented on the disk driver - this means that you have to manually present a new zone, which would map to a new target within the OS to go beyond 8 tape drives.

      With disks, it runs cXtXd0 throug cXtXd7 - then it goes cXtX+1d0 through cXtX+1d7, then cXtX+2d0 through cXtX+2d7 - etc....

      As to the Linux vs Solaris - we've had so many databases get corrupted due to RAC on linux, that they've stopped using the RAC filesystem and gone to another one. We've had issues with Linux servers (3 identical systems, identical hardware, identical builds) that just behave differently, it's uncanny. One of the servers will mount about 1/3rd of their filesystems and then error out the rest as though it timed out scanning all the devices down the san before mounting. We've had Linux servers decide they just want to shutdown - no errors in the logs, nothing wrong with the hardware - just poof - down. Hardware vendor is at a loss, RedHat is clueless.

      Which filesystem were you using in your test for disk / i/o bandwidth? ufs? zfs? qfs? vxfs? some other filesystem? zfs hasn't been tuned for database performance - yet, so I wouldn't use that. I'd either use qfs or samfs - or possibly use oracle's own built in disk management system and see how that goes.

      We've been migrating a ton of web applications off of 6800s, 3800s, 280Rs, etc onto just a few T2000s (yes, a 3 year old platform) and the performance of those applications has improved dramatically.

      I can't wait to try out a few 5240s to replace a handful of T2000s.

      So... sorry to burst your bubble - but unless they've done some new tricks (and multiple calls to HP support suggest that they haven't), they can only handle 8 luns per target, before incrementing the target number.

      --
      Who is general failure, and why is he reading my hard drive?
    49. Re:From an experienced Admin's perspective by mccabem · · Score: 1

      ZFS is at the top of so many folks' list that I figure I should throw this out just for reference:

      ZFS is an available option in FreeNAS.

      I'm guessing by extension that it's available in the general FreeBSD distribution too.

      -Matt

    50. Re:From an experienced Admin's perspective by synthespian · · Score: 1

      What Sun needs to do is to educate young people by supporting user groups and presenting OpenSolaris to LUGs at Universities.

      --
      Main difference between the BSD license and the GPL license: one is from California and the other is from Massachusetts
    51. Re:From an experienced Admin's perspective by Kz · · Score: 1

      • SMF. Being able to start services in a dependency tree is excellent if you have a multi-processor machine. And having services self-heal, including restarting any dependencies, is good for things like mail servers that use a lot of flakey milters.

      sounds nice, and seems doable with scripts. i'll look around if there's something like that.

      • Zones. Because sometimes full-blown virtualisation is too much like hard work.

      OpenVZ should be comparable, but i find KVM better suited to my needs. BTW, can Solaris Zones migrate from one box to another? i don't think OpenVZ can, but might be wrong.

      • There's probably a Linux equivalent of rcapd, to limit the physical memory use of particular groups of processes, but I've never found one.

      don't know of any either, but i really think it should be some way to do it.

      • There's probably a Linux equivalent of processor sets, CPU shares and the Fair Shares Scheduler, but again I've never found one.

      taskset works very nice with nice and ionice to manage KVM guests.

      --
      -Kz-
    52. Re:From an experienced Admin's perspective by larien · · Score: 1
      Some rebuttals:
      • ZFS - every server needs some kind of storage; your choices are some kind of native LVM (Disksuite, etc) or Veritas volume manager (pricy). ZFS has a few missing features still (e.g. can't remove a disk from a pool), but it's getting there.
      • SMF - svcs -x shows faults - it's pretty clear from that where hte issue is.
      • FMA - it will report faults and Solaris has had page retiral for single bit errors for some time now.
      • Zones - if you're sharing servers, it's a boon. Also useful for a lot of dev environments on a single larger server. Another reason we're considering is for clustering/disaster recovery. A common issue is config which hasn't been copied to the DR/cluster servers; with a zone on EMC replicated storage, we just flip the zone with all its config to the DR site and away we go. Binary compatibility - yes, if you're missing the source it's a major issue, but it's very useful to have applications just work(TM) after an upgrade.
      • rcapd - ulimit will do it per-process, but not across a set of processes; i.e. set a limit of 500MB but you could run 100 processes at 500MB and kill your box.

      The replacement of Solaris with Linux for cost savings could, in most cases, be equally met by using Solaris on x86. If you're doing "real" production workloads, chances are you're running on fairly similar hardware to what you'd run Solaris x86 on and have support from Redhat/Suse.

    53. Re:From an experienced Admin's perspective by Enderandrew · · Score: 1

      There are patches to switch schedulers are boot, but Linus has fought to keep these from going upstream. He doesn't want the scheduler to be modular.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    54. Re:From an experienced Admin's perspective by segedunum · · Score: 1

      Solaris just works and its made for servers. Linux seems always beta quality with its cutting edgness and is desktop oriented. I would not trust my job to it unless its Debian or RHES which costs $$$ as cutting edge features are not needed on a mission critical server.

      This is such a sad thing to say now it's just not even funny, and I feel mighty sorry for the Sun reps and consultants who still have to deliver this very, very tired line of a 'beta' quality Linux and Solaris being a 'real' Unix OS. Sad. Plain sad. After the dot com boom at the turn of this century, and after cheaper x86 servers and Linux started delivering a lot more performance than Sun's 'quality' hardware, anyone who cared about their job and performance moved. I really don't know what it will take for those people to go back.

      Certainly, if you were an admin running quite a bit of open source software such as Zope, Python etc., and you were *trying* to run it on Solaris, you switched to Linux years ago because developers didn't give a shit about a proprietary Unix running on proprietary hardware where they couldn't work out what was happening. Too expensive to develop for, too expensive to troubleshoot. These are the people OpenSolaris is trying to get back, and it has long since been too late. Many universities and educational establishments fitted that bill back then, and they moved.

      Linux is simply a far better Unix. Open source developers develop for a Linux platform first, you find problems and solutions pretty much instantly when you run Linux as your platform, and that is the kind of 'support' Sun just cannot match. Any Linux distribution is also far less of a PITA to install, to admin and especially to install software for.

      Solaris is a *real* Unix OS because you need a consultant to sit with it on site for several days to get it to do anything. That is how Solaris is designed, from the ludicrous time needed to install the OS itself to the time needed to install software. It is designed to eat $$$ and £££ in support costs. That's the bottom line.

      Solaris scales far better than any BSD or Linux distro out there.

      Another sad and tired line from the Sun rep handbook. That's just not the experience that most people are getting, and 'scalability' so misused these days it's hilarious.

      Ask any *real* Unix admin who uses both and more than likely they will say Linux is great for small jobs but Solaris is king for anything else.

      Sad. There's that 'real' word again that you get from Sun reps and consultants out there. Do you not think people are getting 'real' work done with their Linux systems? Is that so incredible? Do you not think 'real' Unix admins want to log into a shell that doesn't suck and has actually improved since about 1988?

    55. Re:From an experienced Admin's perspective by Enry · · Score: 1

      Well that's why I admin Linux boxes instead of trying to pass myself off as a Sun admin, your snark notwithstanding.

    56. Re:From an experienced Admin's perspective by Enry · · Score: 1

      Patches are installed by default? Sweet!

      Unless you mean things like getting the 6/08 media and install that which includes all patched up until then. Then you still have to install patches since 8/08 via descriptive names as 1-1854298.zip

    57. Re:From an experienced Admin's perspective by khb · · Score: 2, Informative

      ""it scales far better" in general is as absurd as it is patently untrue."

      How so? With the exception of the SGI implementation, I don't don't of Linux distros out of the box that scale to 100+ processors well (indeed, the usual approach is to run multiple instances of the kernel on each processor or so on large ensembles). When discussing an OS, scalability is usually meant in terms of how the OS itself scales ... not how one's applications can be configured to scale.

      So I think the poster's claim that Solaris scales far better is true FOR THE OS itself.

      Now, whether THAT matters to your workload (or if your workload scales across many processors via multiple OS instances just fine) is another question entirely.

    58. Re:From an experienced Admin's perspective by Alain+Williams · · Score: 2, Informative

      If 100% is the maximum load, how are you exceeding it? Unless you are using a higher value as the maximum percentage, your math troubles me.

      The load average is the number of processes waiting to run, quite different from % CPU usage.

    59. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      Actually the funny HPC installation is ranger at TACC. It is 100% sun gear, the fancy 'magnum' IB switch and it runs CentOS not Solaris. Go figure.

      http://en.wikipedia.org/wiki/Texas_Advanced_Computing_Center

    60. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      Well I have, and I can also tell you that HPC is usually run on outdated(see heavily modified), and as sibling says, Heavily modified Linux (see outdated).

      And of course the most important point, with nodes in the hundreds who cares if one of them accidentally goes down?

    61. Re:From an experienced Admin's perspective by Jim_Maryland · · Score: 1

      SMF - while nice, i have experienced many different kinds of errors. If one of the dependencies has a problem, the chain breaks and it is a pain to discover the problem.
      SMF - The logging related to SMF is fairly robust. Any of the problems I've encountered have been easy to track (maybe not fix, but at least I find the problem quickly enough)

      Zones - I still have yet to see a reason to use this except for dedicated virtualization servers.
      Zones - Software licensed per CPU on the server may acknowledge zones as a valid way to segment the system. We've done it with Oracle and ArcSDE on a server with 4 CPUs where we only wanted those apps taking 2 CPUs.

    62. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      ZFS - Are you really using your server for data storage? SAN or NAS should be a better option depending on your price point

      You understand that you still need a file system and possibly a volume manager for SAN disk space, right? ZFS is both of those. If you really want to do complicated things, you should probably use Veritas StorageFoundation, but ZFS is pretty good for free.

      SMF - while nice, i have experienced many different kinds of errors. If one of the dependencies has a problem, the chain breaks and it is a pain to discover the problem.

      I thought SMF was a little overly-complicated at first, but once I figured out how it worked and how to register my own apps, I'm pretty happy with it.

      FMA - if hardware is broken, i would rather the machine be fully broken and out of service. Running on degraded hardware is too much of a risk. if a few bits get switched or some data is not written correctly, you could corrupt data.

      Your machine without FMA is probably at higher risk of "bits get switched or data is not written correctly". When you have a 1U server with few resources, you're right -- it's probably just as well or better for the whole machine to die. When you have a real server, FMA is your friend.

      Zones - I still have yet to see a reason to use this except for dedicated virtualization servers.
      binary compatibility - if you are running custom code without the source, sounds like you have a setup for failure.

      zones aren't perfect, but they're great for some uses. For example, I put 4 developers on one T1000, each with their own zone. They can install stuff, reboot, rm -rf /, whatever makes them happy, and the other 3 guys aren't impacted.

      I've been in many different places with Solaris, Linux, and a few other random UNIX environments. In almost every case, the Solaris and other random unix environments could be replaced with Linux at 1/10th the cost.

      I'd say almost the exact opposite -- I've seen people spend tons of switching from Solaris to RedHat. And if your answer is to use some free Linux distro with no support, then Solaris x86 can be just as free.

    63. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      If your hardware is supported. ZFS is great, but my Areca card (which works great on Linux and FreeBSD) does not work stable on Solaris.

    64. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      So what, you buy hardware based on the OS you want to run on it... Plenty of stuff works on windows and not Linux, does that mean that Windows is better?

    65. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      Who is 'we'? In who's name are you speaking?

      BTW, wake me up when Solaris has good package management. (Nexenta is doing well but limits to 1 TB storage).

    66. Re:From an experienced Admin's perspective by GuyverDH · · Score: 1

      Sorry - but the "we" remains anonymous...

      While the package management isn't the best, it's definitely not bad...

      Able to manage the packages (and dependencies) of the global zone and all running containers with a single command...

      --
      Who is general failure, and why is he reading my hard drive?
    67. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      Solaris scales far better than any BSD or Linux distro out there.

      Yeah, you know, the roadrunner team would like a word with you, as would pretty much everyone in the Top 500. For some business loads Solaris scales better. But the claim the "it scales far better" in general is as absurd as it is patently untrue.

      Ask any *real* Unix admin who uses both and more than likely they will say Linux is great for small jobs but Solaris is king for anything else.

      Ah, and no true scotsman^W UNIX admin would run a supercomputer, right?

      I designed the roadrunner cluster and many other Top10 clusters in the Top500. You are just lying.
      Road runner cluster is a beowulf cluster of linux boxes. And the choice of Linux was simply because of infiniband driver over verbs at that time which Solaris did not have. Solaris scales just as well in a beowulf environment over ethernet.

      There is an all too obvious reason why SUN did not pursue development of IB drivers aggressively.

      Please don't quote Top500 clusters for reference. They are un-realistic and cutting edge clusters with a record of very bad uptime and 0 reliability.

      They are fine for research activities but are no where suited for enterprise environments or requirements.

    68. Re:From an experienced Admin's perspective by Brandon+Hume · · Score: 3, Informative

      One of our Linux servers regularly copes with a load in excess of 100. Things slow down, but nothing breaks.

      Be careful with comparisons like these.

      Linux lumps disk I/O into the load average, whereas most "other" Unixes don't. I've seen a Linux box with a load of 300+ and idle CPU, and a Sun with a load of 2 that was near unusable because the disks were being thrashed to death.

      Comparing the two can be unfair to either side depending on the context. It's apples and laundry detergent.

      --
      Brandon Hume
      hume -> BOFH.Halifax.NS.Ca, http://WWW.BOFH.Halifax.NS.Ca/
    69. Re:From an experienced Admin's perspective by kildurin · · Score: 1

      Well, Ill say I have been running Solaris since the 3/80. Sun support has never sucked and I have worked several problems with them. We also use Red Hat and the unfortunate thing is that Red Hat refuses to make changes to Red Hat 5 when the kernel is fixed in Red Hat 6 (unreleased). They simply will not back port. No, I have made a significant amount of use of Sun support over the years and have yet to find problems that I "stump" them on. Do I end up in back line support talking directly to engineers who wrote the stuff. Yes. But I have yet to get a Linux Kernel engineer on the phone at 3AM. Oh, and I was extremely impressed over the 4th of July weekend when a lot of us (System Admins) were upgrading our servers and Sun was there. In fact, I know I got the A team as far as support goes. That is just the way Sun rolls. At least that has been my experience since 1987.

    70. Re:From an experienced Admin's perspective by Skrynesaver · · Score: 1

      there are regional OSUGs around the place supported by Sun , freebies and lectures, look them up and see if you're near one

      --
      "Linux is for noobs"-The new MS fud strategy
    71. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      Core for core... you might want to re-look at the definition of "modern equipment". Power 6 is tops right now. The new US-VII in the m9000 places it at 10% better spec-int (over the p6 595) but it takes 4x the number of cores to get it done.

      My customers are actively pulling Sun boxen and replacing them with AIX right now - for performance reasons - and for anything that is licensed by core.

    72. Re:From an experienced Admin's perspective by illumin8 · · Score: 1

      btw - this is on HP-UX 11i

      You didn't mention HP-UX in your previous post. I'm talking about Linux running on HP Proliant or Sun X4600 servers. So, apples -> oranges. I'm not sure of any LUN limitations on HP-UX.

      Which filesystem were you using in your test for disk / i/o bandwidth? ufs? zfs? qfs? vxfs? some other filesystem?

      We were using ASM and raw filesystems. ASM is Oracle's own volume manager. You present a bunch of raw LUNs and Oracle does a RAID 0 stripe across them. The data is protected on the storage array using RAID 1+0 or RAID 5.

      On Linux, with hundreds of LUNs, you get /dev/sda - /dev/sdz, then you go to /dev/sdaa, /dev/sdab, etc... Very easy to manage. No need to format, label, partition, etc, all of that BS that Solaris requires you to do before you can write to a disk. Just start writing to the raw /dev/sda and you're good.

      --
      "When the president does it, that means it's not illegal." - Richard M. Nixon
    73. Re:From an experienced Admin's perspective by QRDeNameland · · Score: 1

      If 100% is the maximum load, how are you exceeding it?

      This one goes to 110%. (apologies to Nigel Tufnel)

      --
      Momentarily, the need for the construction of new light will no longer exist.
    74. Re:From an experienced Admin's perspective by fan+of+lem · · Score: 1

      I am looking for an excuse to try out Solaris, and I thought of using it as a media server. Can it work well as, say, either a MythTV backend or front end? Or perhaps the better question is, can you run MythTV (or any other similar app) on it at all?

    75. Re:From an experienced Admin's perspective by Macka · · Score: 1

      There's probably a Linux equivalent of processor sets, CPU shares and the Fair Shares Scheduler, but again I've never found one

      Use taskset(1) to launch a program and tie them to cpu sets. or adjust them on the fly afterwards. As for your Fair Shares Scheduler, I've no idea what that does on Solaris (smells like a marketing term to me) but the Linux kernel has a FAIR_GROUP_SCHED option which likely does the same thing.

    76. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      And they are replacing their Sun boxes for last 3 years.

      I have worked on these (as a developer, mind you), so at least I can attest Linux's readiness with its 'beta' quality. So, take your bullshit somewhere else.

      I'll meet your anecdote with one of my own:

      I've worked with a Fortune 50 company that just ripped out shitloads of Dell hardware running Redhat and replaced it with heavy iron. You're the one who is full of shit.

      Talk with me when your toys are fault tolerant with 512 cores on the same box such that you can drop a fucking thermite grenade on it and it still runs.

      Let me guess: you're a PHP developer.

    77. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      What about hardware support? I keep hearing that the openSolaris kernel just frankly doesn't have many drivers. If I can't install it on my hardware, it isn't doing me any good.

      Sun (duh), Dell, and IBM all will sell you Solaris OEM (certified). HP doesn't have an agreement with Sun for their x86 gear, but all their G4 and G5 is certified and supported:

      http://h71028.www7.hp.com/enterprise/cache/492635-0-0-0-121.html

      If you need enterprise-y stuff you're set; if you're talking desktop / white box then it could be more of an issue.

      More info on hardware support:

      http://www.sun.com/bigadmin/hcl/

      And see if your system is supported:

      http://www.sun.com/bigadmin/hcl/hcts/device_detect.jsp

    78. Re:From an experienced Admin's perspective by C0vardeAn0nim0 · · Score: 1

      Zones - I still have yet to see a reason to use this except for dedicated virtualization servers.

      server consolidation is one use. if you have several small servers, and then decide to consolidate in a single, big one, having the services of each small machine on it's own zone can be a lot more flexible than hardware partitioning.

      --
      What ? Me, worry ?
    79. Re:From an experienced Admin's perspective by C0vardeAn0nim0 · · Score: 1

      if he's drinking the coolaid, you're smoking some serious shit.

      compared to the kind of service we have from IBM here at work (a 3 letter acronym that just got bought by a 2 letter acronyn. check the news.) sun's support is AWESOME!!!

      hell, last week IBM sent an "enginner" to our chilean site that didn't know how to use "sudo su -". then we all asked here: WTF ?!?

      Sun even have dedicated engineers on two of our sites in US, and they're pretty quick to answer and solve our problems. a lot better than IBM in my experience.

      --
      What ? Me, worry ?
    80. Re:From an experienced Admin's perspective by jabuzz · · Score: 1

      In a word GPFS, and then add in clustering, HSM and ctdb, and no I am not an IBM sales rep.

      Frankly though I would also be far happier using XFS on a 20TB SAN with 15 years of heritage than this new upstart ZFS.

    81. Re:From an experienced Admin's perspective by jabuzz · · Score: 1

      On 7200rpm SATA spindles only. Great if you don't need lots of IO's otherwise your a bit stuffed. Also great if you never need more than 48TB. If you however do then you are a bit stuffed with the thumper approach as it does not scale.

    82. Re:From an experienced Admin's perspective by jabuzz · · Score: 1

      ZFS is also missing user guota's, clustering and HSM. If ZFS floats your boat good for you, but claiming it does everything and is without serious problems is shear ignorance.

    83. Re:From an experienced Admin's perspective by jedidiah · · Score: 1

      Yeah, but he was talking about gear that runs on FreeBSD of all things.

      Do I really have to be the one to point out how sad this is?

      This is SERVER gear we're talking about here and Linux has been very respectable in this area for a long time.

      --
      A Pirate and a Puritan look the same on a balance sheet.
    84. Re:From an experienced Admin's perspective by NotBornYesterday · · Score: 1

      He's talking about real-world apps running on large SMP hardware, not some thousand-node HPC cluster. Scaling a single instance of an OS over 30, 60, 100 or more cpus in a single box is tricky business, and Linux doesn't do it as well as Solaris does. Maybe Linux is more stable than it used to be, but Solaris has a proven track record of top-notch stability and uptime. Top500,org is interesting, but irrelevant to the GPs point.

      --
      I prefer rogues to imbeciles because they sometimes take a rest.
    85. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      A load of 80 on a 32 cpu system sounds like a poorly architected solution to me. :)

      Late but I have to respond. A load of 80 on a 32 processor machine is fine. Just like a load of 4 on a 1 processor machine is fine. It just means that you are using your hardware. In fact, anything below 32 on his machine means that the company is wasting their money. 80 is average of like 2.5 per processor.

      We've actually had more failures with our Sparc/Solaris machines, generally exhibiting as spontaneous reboots.

      I would hate to work with you. You are a sysadmin that will leave knowingly broken hardware in production and not bother to understand what is happening to it? I have never seen a Solaris machine randomly reboot(any computer for that matter). Most likely what you are seeing is a product of a stupid user(vi'n a 5 gig log file). If Solaris was left to its own devices, it will make /tmp part of its swap. Computers are not random.

    86. Re:From an experienced Admin's perspective by jhol13 · · Score: 1

      Binary compatibility ... your "setup for failure" is true - in Linux.

      Fortunately it is not true in (Open)Solaris (or Mac or Windows or ...).

      Listen, I really, really hate that X server breaks for every minor-minor kernel update (I have switched to Ubuntu which does not have that brain damage). I really, really hate when I have to recompile (FOSS) webcam (and DVB and maybe two other) drivers after every minor-minor kernel update.

      Maybe you have time for all that. We don't - I know people who do not update their kernels anymore.

    87. Re:From an experienced Admin's perspective by Alpha830RulZ · · Score: 1

      I don't don't of Linux distros out of the box that scale to 100+ processors well

      Isn't the usual mode of scaling Linux based systems oriented towards bunches of generic boxes with a small number of processors? I don't have broad experience here, but the sites I have worked with don't want to spend the money it costs for large SMP boxes, when, by architecting the application correctly you can scale it across large numbers of commodity boxes more economically, and gain resiliance in the bargain.

      I have nothing against Solaris, but neither does it do anything for me. Our company has done well with openSUSE and CentOS/Fedora. As I said above, we haven't had any stability issues, and nothing that would force us to go to Sun. Our load is compute and file system, though, so we don't need a huge Oracle server.

      --
      I was taught to respect my elders. The trouble is, it's getting harder and harder to find some.
    88. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      I tested Solaris 10 and Red Hat Enterprise Linux 4 (64-bit) on the exact same hardware (X4600), and you know what? Red Hat Enterprise Linux 4 performed better on massive storage I/O than Solaris 10. I have the Oracle ORION benchmarks to prove it. We have over 50 LUNs carved from an HP EVA 8100 and presented to these X4600s, on 4x 4gb HBAs per server. They run Oracle RAC, have 4x quad core AMD

      Sounds like you used QLogic cards and Sun's qlc driver.

      Try the QLogic driver.

      But, as a "former Sun SSE", you should know that, right? (Although the qlc driver has gotten markedly better lately, until recently the QLogic qla driver kicked the qlc driver's butt...)

      And do you have the BALLS to try and put an "strace" on your PRODUCTION Oracle database processes running on your Linux box? With all the "-ff" and thread options? I've locked Linux boxes up - hard - with that "trick", yet I have no qualms whatsoever about dropping a truss on any process whatsoever on any Solaris box. And I've only once had any issue, and that was because of a bug in the kaio driver in Solaris 8 about 4 years ago (read the patch history of the Solaris 8 kaio driver sometime - that was ME driving all those changes...)

      An "SCSA" and "former Sun SSE"?

      I'm calling YOU on that one.

    89. Re:From an experienced Admin's perspective by Alpha830RulZ · · Score: 1

      Ask any *real* Unix admin who uses both and more than likely they will say Linux is great for small jobs but Solaris is king for anything else.

      Which is why Google runs on Solaris, right?

      Oh, wait...

      --
      I was taught to respect my elders. The trouble is, it's getting harder and harder to find some.
    90. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      On Linux, with hundreds of LUNs, you get /dev/sda - /dev/sdz, then you go to /dev/sdaa, /dev/sdab, etc... Very easy to manage.

      You think a naming scheme that provides no information whatsoever is "easy to manage"?

      Just because it's short?

      Ooooh-KAY.

    91. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      there's a GUI that auto-installs patches for you... same as ubuntu

    92. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      Does anyone truly, really want years of uptime? No kernel patches at all? If any of my HA services depended on a single machine, Solaris or not, I'd be very unhappy.

    93. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      ZFS - Are you really using your server for data storage? SAN or NAS should be a better option depending on your price point

      Sun x4550: 48 TB raw in 4U, with eight AMD cores and 64 GB of RAM, quad GigE.

      Stripe or RAID (5,6) in any combination with unlimited snapshots using ZFS.

      http://en.wikipedia.org/wiki/Sun_Fire_X4500

      Also certified to run RH/SuSE Linux and Windows.

      ZFS loves cheap disk, and you can sleep easy for data safety because every single bit on the disk is checksummed. ZFS is basically built on a Merkle (hash) tree. You know your data isn't corrupted, and if it finds corruption it can fix it if you've set up redundancy.

      Do NetApp or EMC or Hitachi have interlocking checksums that can reconstruct data? If you check the main ZFS mailing list, people have reported finding bugs in HBA firmware that was corrected by ZFS.

    94. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      RHEL is relatively cheap and Debian is free

      And Solaris is cheaper than RHEL for software support (list price, discounts may vary):

      http://blogs.sun.com/BVass/resource/SolarisRHELWinComparison.pdf

    95. Re:From an experienced Admin's perspective by Wdomburg · · Score: 1

      Fair point about load average (find a lot of people get needlessly hysterical about that without looking at other metrics), though I'm not generally fond of relying on vertical scaling to that extent.

      As for the rest, where exactly did I say we left broken hardware in production? Failed machines are pulled for diagnostic and root cause analysis. It may be relatively rare compared to some vendors, but Sun hardware can and does fail and Sun software can and does fail to handle it gracefully sometimes. We've hit undetected cpu faults, firmware bugs and of course the fucked cache modules Sun was shipping in the older Enterprise servers for a while.

      And no, it had nothing to do with user actions. The only applications we've run on Solaris machines are databases. Only a handful of highly trusted admins even have access and in none of the cases was anyone logged in.

      Academic matter, anyways. We're largely abandoning Solaris. We're getting far better price/performance on x86 hardware and the reliability advantage is minimal to non-existent. Not that we trust the reliability of a single machine anyways.

    96. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      Who cares about a package management? A good sysadmin doesn't need that. Manage your systems correctly and you won't have the need for a packaing management.

    97. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0


      roadrunner team would like a word with you, as would pretty much everyone in the Top 500.

        Huh??? Extremely few of the Top 500 are single OS image machines. (if any at this point... not going to look through the whole 500). Roadrunner has many 100s operating systems operating independently. In the "scales" perspective little different than 100s of desktop PCs independently operating. The shared resources and interconnect are different but the operating system isn't "scaling" significantly more.

    98. Re:From an experienced Admin's perspective by Wdomburg · · Score: 1

      List price is marginally lower, yes. It's an apples to oranges comparison, though, since Sun is covering an operating system and a small number of bundled applications (Apache, Bind, Sendmail, WU-FTPd, OpenSSH, Samba and a few other minor components). The subscriptions from Red Hat are far more comprehensive, covering software that Sun ships with Solaris but cover with the support contract (e.g. MySQL and Postgresl) as well as a ton of additional software that Sun doesn't even ship (Postfix, Exim, VS-FTPd, lighttpd, Tomcat, Dovecot, blah, blah, blah, etc).

    99. Re:From an experienced Admin's perspective by GuyverDH · · Score: 1

      Sorry - I still don't classify HP with the compaq gear they throw their label on.

      When I say HP - I mean PA-RISC or ITANIUM based servers - which for the most part rely on HP-UX as an OS, which I did reference in my previous post - just not in the same sentence - so I can understand the confusion.

      Migrating from an old SAN to a new SAN has been a major pain in the ass for the Linux servers - drive letters changing, size of virtual disks changing, layouts changing - and all trying to use their replication utilities.... ugh...

      ZFS has been a dream... I wrote a simple script that writes the scripts to map the old drives to the new drives, run it (a series of zpool replace commands) and done.

      I would presume that presenting the raw disks using ASM on Solaris would be similiar - although, yes, it may require that a label be present - but that's such an easy chore. I've got a script that will scan disks for those that aren't labeled, presents that list, you answer yes and it labels them.... One extra step... Not going to worry about it.

      And all you have to do is label... once labeled, use slice 2 on SMI labeled drives as your device and nothing else needed.
      With ZFS, you don't even use the slice numbers, it relabels to an EFI format and uses the disk.

      Not too much BS, and beats having to deal with drive names changing on every other boot when you've added additional disks to the zones.

      --
      Who is general failure, and why is he reading my hard drive?
    100. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      ZFS - every server needs some kind of storage; your choices are some kind of native LVM (Disksuite, etc) or Veritas volume manager (pricy).

      Ever heard of hardware RAID?

    101. Re:From an experienced Admin's perspective by Anonymous Coward · · Score: 0

      Winner? You've watched too much Olympics games.

  4. Love that they open sourced it... but... by jhfry · · Score: 4, Interesting

    I love that Sun open sourced it, however I think that the greatest benifit is not that it's open but that the technologies it offers are available to be reproduced on other nix os's. The biggest issue I have with OpenSolaris is that it's still a single vendor OS. If it forks a few times and actually develops a culture and some competition between vendors than I think it will be more appealing.

    That's actually what I hate and love about linux. It's a fragmented and ineffecient community, but because it's fragmented I don't have to worry that it's going away any time soon.

    --
    Sometimes the best solution is to stop wasting time looking for an easy solution.
    1. Re:Love that they open sourced it... but... by dreamchaser · · Score: 1

      Very insightful. One of the things I've always said was a strenght of OSS was that it provided redundancy of that nature. Even if one fork/project of a given set of code fails, if the idea is worthy it will live on.

    2. Re:Love that they open sourced it... but... by houghi · · Score: 1

      but that the technologies it offers are available to be reproduced on other nix os's.

      So why is ZFS for Linux not yet out of Beta?

      --
      Don't fight for your country, if your country does not fight for you.
    3. Re:Love that they open sourced it... but... by Kjella · · Score: 2, Insightful

      I love that Sun open sourced it, however I think that the greatest benifit is not that it's open but that the technologies it offers are available to be reproduced on other nix os's.

      Except the small detail that the CDDL is incompatible with the GPL, so you won't see things like Linux kernel-based ZFS. From what I've understood running it through FUSE (userspace) isn't all that great. I do understand why Sun doesn't want Linux to take all its crown jewels, but it's still annoying.

      --
      Live today, because you never know what tomorrow brings
    4. Re:Love that they open sourced it... but... by TheRaven64 · · Score: 4, Insightful

      He said 'other nix os's' not 'Linux'. The GPL may be incompatible with the CDDL, but the BSDL isn't, and bits of Solaris, such as ZFS and DTrace, have found their way into FreeBSD.

      Saying the CDDL is incompatible with the GPL is misleading - the CDDL doesn't say anything about code not explicitly released under it. It is the GPL which imposes constraints on third-party code. If Linux used a more permissive license then it would be able to use OpenSolaris code, and OpenSolaris would be able to use Linux code just as it used to use a lot of BSD code back in the SunOS days.

      --
      I am TheRaven on Soylent News
    5. Re:Love that they open sourced it... but... by ShieldW0lf · · Score: 1

      Saying the CDDL is incompatible with the GPL is misleading - the CDDL doesn't say anything about code not explicitly released under it. It is the GPL which imposes constraints on third-party code. If Linux used a more permissive license then it would be able to use OpenSolaris code, and OpenSolaris would be able to use Linux code just as it used to use a lot of BSD code back in the SunOS days.

      Except, of course, that if it wasn't for the GPL, there would be no OpenSolaris. There would only be Solaris.

      --
      -1 Uncomfortable Truth
    6. Re:Love that they open sourced it... but... by jhfry · · Score: 2, Informative

      Because in order for it to operate optimally it must be part of the kernel, and Linus and crew refuse to put it in the kernel due to licensing issues.

      It runs fine in userland with fuse, but it's slow.

      --
      Sometimes the best solution is to stop wasting time looking for an easy solution.
    7. Re:Love that they open sourced it... but... by jhfry · · Score: 1

      That's a tough argument... the GPL is very widely used but is hardly the only license. It's entirely possible that OpenSolaris would still exist if the GPL had never been used and something like BSD was used instead.

      --
      Sometimes the best solution is to stop wasting time looking for an easy solution.
    8. Re:Love that they open sourced it... but... by BrokenHalo · · Score: 1

      ...but because it's fragmented I don't have to worry that it's going away any time soon.

      Exactly. Evolution at work. "The Selfish Code-snippet"?

    9. Re:Love that they open sourced it... but... by ShieldW0lf · · Score: 1

      That's a tough argument... the GPL is very widely used but is hardly the only license. It's entirely possible that OpenSolaris would still exist if the GPL had never been used and something like BSD was used instead.

      No, it isn't. Before the GPL existed, Sun used BSD code in their closed source offering without releasing control in the slightest. They are control freaks struggling to remain relevant in a world that has no tolerance for such.

      --
      -1 Uncomfortable Truth
    10. Re:Love that they open sourced it... but... by Anonymous Coward · · Score: 0

      The problem isn't the license, it's patents. While the CDDL and GPL is indeed incompatible, it isn't like the source would be very useful, except to lookup concepts. Linux and solaris have so different designs that you can't just plug one into place in the other.

      Grub has a ZFS implementation (limited basic reading capabilities), developed by Sun, but it's specifically designed to not use functions that sun has patents on. I think it's fair to say that on the subject of ZFS, Sun wants to seem open but not be open.

      But nobody is going to do it while Sun is holding patents on it and seem to be doing their best not to give any indication of whether they would use them.

      Btw, I think after Redhat Sun is probably the one large corporation that does most for Open Source. So it's not like they have anything to be ashamed of, except for playing more open than they really are. They might as well admit that ZFS isn't Open Source, and get over with it.

    11. Re:Love that they open sourced it... but... by fishbowl · · Score: 1

      >Saying the CDDL is incompatible with the GPL is misleading - the CDDL doesn't say anything about code not explicitly released under it. It is the GPL
      >which imposes constraints on third-party code.

      You make it sound like the GPL can reach up from the toilet seat and grab your code.

      --
      -fb Everything not expressly forbidden is now mandatory.
    12. Re:Love that they open sourced it... but... by Anonymous Coward · · Score: 0

      So why is ZFS for Linux not yet out of Beta?

      Is ZFS for Linux even in Beta? I don't think ZFS-Fuse may be fine for hobbyists, but it's not even close to a practical solution for my primary filesystem. It is not something I would trust enough to run my business on.

    13. Re:Love that they open sourced it... but... by jhfry · · Score: 1

      Right... but Sun was not required to open Solaris. Their reasons for doing it had nothing to do with the GPL... it was purely to try and gain some mindshare and some industry recognition that Linux and the BSDs had taken over the years.

      --
      Sometimes the best solution is to stop wasting time looking for an easy solution.
    14. Re:Love that they open sourced it... but... by Kz · · Score: 1

      well designed FUSE ports are really fast (NTFS-3g, GlusterFS, and a few others). the fact that you can write your filesystem on Python or Bash doesn't mean the foundation is inefficient.

      --
      -Kz-
    15. Re:Love that they open sourced it... but... by ShieldW0lf · · Score: 1

      They're not trying to gain mindshare from Linux and the BSDs. They're trying to gain mindshare from Linux. When it was just BSD, they would have just poached the code and given nothing away, like they spent so many years doing.

      If it wasn't for the GPL, OpenSolaris would not exist.

      --
      -1 Uncomfortable Truth
    16. Re:Love that they open sourced it... but... by fm6 · · Score: 1

      I love that Sun open sourced it, however I think that the greatest benefit is not that it's open but that the technologies it offers are available to be reproduced on other nix os's.

      You're not making Sun management happy. Sun has taken a lot of heat from investors for "giving away" Solaris. Management's response: we're creating thousands of Solaris hackers. These hackers influence hardware buying decisions. Solaris runs on x64, but it also runs on SPARC, and that's still Sun's main business. Linux people probably won't buy SPARC systems and Windows people definitely won't. But Solaris people just might, especially if they need a system that scales.

    17. Re:Love that they open sourced it... but... by jhfry · · Score: 1

      Mindshare is not code... it's actual interest in the product. I could poach code day and night, but if no one uses my software, why bother.

      Sun is a corporation and they sell a product called Solaris. How do you convince someone to buy Solaris when there are free alternatives in BSD and Linux? How do you attract developers and administrators to learn to work on Solaris systems? The only way to compete with Linux for that "mindshare" to level the playing field and remove the barrier to entry.

      Solaris uses code developed by BSD and tons of GPL'd software. This is not unusual, almost all modern os's use BSD derived code and GPL tools... including MS Windows. All Sun did is open their proprietary software so that new developers and admins would help Solaris maintain it's current market share and hopefully expand upon it.

      Solaris is an excellent product and very well respected in it's niche... unfortunately with the low cost of hardware and the strides made by their competition Sun is finding it very hard to sell. Making it harder still is the fact that almost no one is "growing up" on the OS anymore. It used to be that Sun had servers running at all levels, from small departmental servers to large clusters; which allowed an admin to get their feet wet at on a small installation and gradually grow into a real Solairs sysadmin. Now Sun has lost most of the small and medium server business to Linux and MS... so how does one learn Solaris except to download it and install it on our own equipment.

      --
      Sometimes the best solution is to stop wasting time looking for an easy solution.
    18. Re:Love that they open sourced it... but... by jhfry · · Score: 1

      I'm not trying to make them happy... but even so there is something to be said for the mindshare they have generated. I know for a fact that if not for OpenSolaris I would never have touched it or even considered it if making a purchasing recommendation.

      Having seen it and used it, I may or may not depending upon the project... but the fact that I know it's there is far better than having it completely forgotten.

      Another thing to consider... if Sun hadn't opened it, it's possible that they would have needed to abandon it completely as it would eventually become impossible to keep it competitive with the decreasing sales figures. At least this way there are free developers working on parts that previously only paid Sun developers could touch.

      --
      Sometimes the best solution is to stop wasting time looking for an easy solution.
    19. Re:Love that they open sourced it... but... by fm6 · · Score: 1

      ...if Sun hadn't opened it, it's possible that they would have needed to abandon it completely as it would eventually become impossible to keep it competitive with the decreasing sales figures.

      As have other hardware companies with their own flavor of UNIX. The difference is, those companies now get most of their business from systems that use commodity processors. Sun was very, very late to that game; its x64 line, though profitable, is still a small part of the business. So Sun can't abandon Solaris, because that would mean abandoning their main product line: SPARC-based servers and supercomputers.

    20. Re:Love that they open sourced it... but... by True+Grit · · Score: 1

      it's actual interest in the product

      Interest that has been eroding since the advent of Linux...

      How do you convince someone to buy Solaris when there are free alternatives in BSD and Linux?

      yet Sun didn't have this problem when there were just the BSDs around...

      The only way to compete with Linux for that "mindshare"

      Was it a freudian slip on your part not to include the BSDs here as well (as you did up to this point)? :)

      This is exactly the point the GP is making: It is the success of Linux that is forcing Sun to open up Solaris in a desperate attempt to stay relevant, not the BSDs, which have been around almost from the beginning. If your argument were anywhere close to correct we would have seen something like OpenSolaris long before Linux, because of the free BSDs.

      Look, you don't have to like the GPL, thats ok, but pretending it is not the underlying reason why it is Linux thats gaining mindshare and usage at the expense of Solaris and most of the other commercial *nixes, and not the the BSDs themselves, doesn't do you any good in the long run... nevermind that it leads to a lot of freudian slips along the way. :)

    21. Re:Love that they open sourced it... but... by DragonWriter · · Score: 1

      Interest that has been eroding since the advent of Linux...

      And Windows NT and its spawn, both of which have squeezed traditional Unix markets. Is OpenSolaris trying to compete with Linux by being open, or is it trying to compete with Windows the way Linux has shown works? Hard to say. In any case, your argument that the GPL is the only reason Solaris is doing this requires one to accept that Linux would not have succeeded with any other open source licensing arrangement, or at least not with out substantively similar to the GPL. That's a point on which there can be endless debate, but no real resolution, since neither side can be proven correct.

    22. Re:Love that they open sourced it... but... by jhfry · · Score: 1

      I am confused, I thought the argument being made is that it's the GPL that made Solaris open up. I will concede that the use of the GPL played a role, but it's assuming a lot to place all of the honor on the GPL.

      There is nothing stopping anyone from running GPL code on Solaris. So essentially Sun could develop their own kernel (with portions pulled from BSD) and supply free GPL'd software along with some of it's own internally developed software... which is exactly what it has been doing. So what your arguement actually is, is that GPL killed BSD, in turn making it so that OpenSolaris could no longer pull free code from the BSD's to keep current. I would love for you to go on a BSD mailing list and tell them that their project is dead.

      I believe its actually hardware that hurt Sun the most. For years Sun owned the high powered workstation and server market, they had the most powerful and stable machines out there. Unfortunately the edge that Sun's SPARC processors had, and that RISC had, was slowly eliminated as Intel and AMD created better and faster x86 processors. Now, even Sun is selling x86 machines and they have ported Solaris to x86. If x86 hardware were still the underdog, Sun would still be king.

      Sun's major strenth is delivering rock solid hardware with a equally solid Posix compliant OS. As long as one of the BSD projects continue to develop a BSD kernel, Sun can continue to borrow code to develop Solaris.

      If the GPL had never been created, it is a very likely possibility that one of the BSD's would be where Linux is now... and Sun would be opening their code to remain relevent.

      It's not the GPL that hurt Sun its the fact that people like Free software more than commercial software. And, of course, that Sun's performance and stability advantages over their competitors have been greatly diminished.

      --
      Sometimes the best solution is to stop wasting time looking for an easy solution.
    23. Re:Love that they open sourced it... but... by m50d · · Score: 1
      He said 'other nix os's' not 'Linux'. The GPL may be incompatible with the CDDL, but the BSDL isn't, and bits of Solaris, such as ZFS and DTrace, have found their way into FreeBSD.

      FreeBSD licensing always seems very schizophrenic, or just unreasoningly anti-GPL - they're happy to include code with many more restrictive GPL-like licenses, even binary blobs on occasion, and then they'll fight tooth and nail to remove GPLed code from the system.

      Saying the CDDL is incompatible with the GPL is misleading - the CDDL doesn't say anything about code not explicitly released under it. It is the GPL which imposes constraints on third-party code.

      You're being very misleading - the CDDL applies restrictions to derivative works, just like the GPL - if it didn't, you could make your added code GPL and combine them with ease, like is done with BSD-licensed code all the time. In fact the reason the CDDL and GPL are incompatible is that they both have essentially the same restrictions. Incompatible goes both ways in this case, and neither license is any more to blame than the other.

      --
      I am trolling
    24. Re:Love that they open sourced it... but... by Anonymous Coward · · Score: 0

      Hmmm, I only see one copy of Linus and the kernel out there....

    25. Re:Love that they open sourced it... but... by Anonymous Coward · · Score: 0

      Saying the CDDL is incompatible with the GPL is not at all misleading, since it was intended to be GPL incompatible. In the words of Danese Cooper, former Sun employee:

      Mozilla was selected partially because it is GPL incompatible. That was part of the design when they released OpenSolaris. [...] the engineers who wrote Solaris [...] had some biases about how it should be released, and you have to respect that.

    26. Re:Love that they open sourced it... but... by neuro88 · · Score: 1

      And what happens when the FreeBSD developers decide they want to include GPL _and_ CDDL code? I think this is part of why the developers of the various BSD's have tried to remain license pure.

    27. Re:Love that they open sourced it... but... by Anonymous Coward · · Score: 0

      I'd much rather zfs and dtrace were cloned than ported. Something built from the ground up to integrate with Linux, picking the design bits that worked well and have stood the test of time and perhaps shedding others that haven't.

    28. Re:Love that they open sourced it... but... by Anonymous Coward · · Score: 0

      Because in order for it to operate optimally it must be part of the kernel, and Linus and crew refuse to put it in the kernel due to licensing issues.

      So the issue is political and ideological, and not technical?

    29. Re:Love that they open sourced it... but... by Anonymous Coward · · Score: 0

      Yes, but it might mean that you hear the words "You run our production database on WHAT? You're fired."

      Fuse-ZFS is cute, but it's not a sturdy solution.

  5. And for some reason... by Creepy+Crawler · · Score: 1

    I thought the ZFS on the "free" version was crippled down to 1 TB.

    --
  6. On VirtualBox, it blows by theoriginalturtle · · Score: 1, Interesting

    I put up Sun's free VirtualBox VM environment on a MacBook Pro, and both OpenSolaris and Solaris 10 Intel were worthless. Both achieved speeds reminiscent of PearPC.

    XP worked OK. Ubuntu was fine.

    You'd think if you were going to release a VM, at least you'd make sure your flagship OS would run on it at speeds that would compare favorably to a 20-year-old Amiga.

    --
    ---------------------------------------
    Rotate the pod, please, HAL....
    1. Re:On VirtualBox, it blows by Danathar · · Score: 2, Informative

      The OS X version of Virtual box does not support (yet) any of the processor specific virtual machine extensions that speed things up considerably.

    2. Re:On VirtualBox, it blows by BrentH · · Score: 1

      It does so perfectly for my Ubuntu Virtual machine.

    3. Re:On VirtualBox, it blows by egghat · · Score: 1

      You would think the other way round would work as well. But several of the last Virtualbox versions didn't run on OpenSolaris 08-05 :-(

      --
      -- "As a human being I claim the right to be widely inconsistent", John Peel
  7. ZFS rocks by larry+bagina · · Score: 3, Interesting

    ZFS kicks ass. Sun really raised the bar with it. There are some other FSs in development (Hammer, btrfs, etc), but they don't have the full integration that ZFS does. Maybe eventually, someone will write a patch so ZFS is just a patch and recompile away in Linux (although that approach is what made minix suck back in the day). Heh, minix will probably have ZFS support before Linux does.

    --
    Do you even lift?

    These aren't the 'roids you're looking for.

    1. Re:ZFS rocks by Enderandrew · · Score: 2, Interesting

      Someone likely could easily issue ZFS as patch directly for the kernel as opposed to in FUSE. The problem is that it would be illegal to use it a such because of the license. Sun has talked about making it all GPLv3 if Linux takes their kernel GPLv3 as well. Linux would gain native ZFS in their kernel, but Sun would gain every device driver from Linux.

      The problem is that too many individuals that you can't even contact own individual copyrights in the Linux kernel. It isn't just going to up and change to GPLv3.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    2. Re:ZFS rocks by ciderVisor · · Score: 0, Troll

      ZFS kicks ass.

      This is not a troll but how can you get excited by a file system, FFS ?

      Can you create directories ? - Check !
      Can you write files to it ? - Check !
      Can you read files back from it ? - Check !
      Can you delete files ? - Check !

      Job done !

      --
      Squirrel!
    3. Re:ZFS rocks by jrothwell97 · · Score: 1

      It depends: if it's the ZFS specification that's CDDL licensed you're screwed as kernel space goes. However, if it's only Sun's software that writes a ZFS file system (the ZFS drivers and toolchain) it's possible that an alternative implementation of ZFS could be created. Sun have said they're 'investigating' a Linux port.

      I've used Solaris (I ordered the DVD over the Internet) and I like it: it's no slower than, say, Kubuntu (KDE4), in VirtualBox, and I love ZFS. Unfortunately, I've misplaced the DVD, and additionally my only DVD burner happens to be playing up, so I'm stuck with the VirtualBox image until I get round to opening up the DVD drive and cleaning it.

      --
      Those using pirated Tinysoft signatures(TM) are a real threat to society and should all be thrown in jail.
    4. Re:ZFS rocks by TheRaven64 · · Score: 1

      Actually, it would be completely legal to use it. It would be illegal to distribute it, because it would be a derived work of the Linux kernel and the kernel is under a license which imposes a strict set of constraints on the third-party code you are allowed to link it with. Linux device drivers tend to be an unreadable mess (not helped at all by the fact that they often include changes to update them to new in-kernel APIs by people who have not looked at the overall design of the driver). FreeBSD has a similar level of driver support to Linux (better in some areas, worse in others) and Sun can already share drivers with them. Solaris has had a stable driver ABI for a very long time too, so drivers released a long time ago still work, while their Linux equivalents only work if someone is actively maintaining them and not just letting them bit rot. Drivers can't just be plugged into a new kernel. They need porting and modifying for the new kernel's interfaces. Linux has the worst layering of any free *NIX kernel when it comes to driver abstraction, and porting drivers from Linux is harder than porting them from pretty much anywhere else as a result (except possibly Darwin, but only because Darwin drivers are written in a subset of C++).

      --
      I am TheRaven on Soylent News
    5. Re:ZFS rocks by fatboy · · Score: 1

      Someone likely could easily issue ZFS as patch directly for the kernel as opposed to in FUSE. The problem is that it would be illegal to use it a such because of the license.

      I don't think the problem would be "using it", it would be with distributing it.

      --
      --fatboy
    6. Re:ZFS rocks by silentsteel · · Score: 1

      Sun has talked about making it all GPLv3 if Linux takes their kernel GPLv3 as well. Linux would gain native ZFS in their kernel, but Sun would gain every device driver from Linux.

      That is the point of FOSS, though, assuming the Linux kernel were migrated to GPLv3 and Sun migrated Solaris, ultimately, it would be a win/win situation for both sides.

      I have run both OpenSolaris and Linux and found that, like any other quality OS, both have their positives and negatives, and preferred usages.

      If both were licensed under GPLv3 individual positives, like ZFS and driver support, would then, potentially, start to be found in both OS's and individual negatives would decrease, making both Linux and Solaris better.

      --
      I cut it three times, and it's still too short.
    7. Re:ZFS rocks by gardyloo · · Score: 1

      Can you create a filesystem in the time it typically takes to create or delete a folder?

          Can you take snapshots and/or clones, quickly, so that one always has a history of each and every file on disk, and easily accessible?

          Can you create automatically-expanding storage pools?

    8. Re:ZFS rocks by gnick · · Score: 1

      That actually does seem a little Trollish (as do most posts that start with "This is not a troll..."), but I'll bite anyway.

      Disclaimer - I've not used ZFS and know zilch about it.

      But your lack of excitement over FS capabilities is disturbing.

      Does it make efficient use of the space available on the HD? Maybe.
      Does it organize files in such a way that they can be quickly found and read? Maybe.
      Can it recover from minor disk errors? Maybe.
      Does it throttle the HD by constantly having to rearrange data in order to maintain the above capabilities? Maybe.
      Etc.

      Job continues indefinitely.

      --
      He's getting rather old, but he's a good mouse.
    9. Re:ZFS rocks by caseih · · Score: 2, Informative

      ZFS does have issues with NFS though. In particular NFS writes can lock up the client. Hopefully this issue will be fixed soon. it's not really ZFS's fault; it's NFS's fault. Yet no other FS has this issue, so I'm sure a workaround could be done. In the meantime there are times when the NFS clients on our 12 TB ZFS (Solaris 10) system are unusable.

      ZFS will never come to Linux while the license remains incompatible with the GPL. I predict that one day Sun will relicense it, but not before they've really tried and failed with OpenSolaris. I think the FUSE idea is viable though. Just that the developer who was doing it is now working for Sun and has no time/inclination to do more with it. In theory all file systems should be implemented with a FUSE-type of separation from the kernel. This hearkens back to the Mach days where file systems were envisioned to just be user-space servers. Now it just might work. NTFS-3g has really good performance, really, and it is fuse. And when it crashes, it won't take down the OS.

      So far the community or lack of, surrounding OpenSolaris is pretty disappointing. I think Sun just thought it would magically happen. Long term, I'd rather see the best technologies of Linux and Solaris merge (Linux massive hardware support, scheduler, dtrace, zones, zfs), rather than continuing in separate directions.

    10. Re:ZFS rocks by Wdomburg · · Score: 3, Informative

      To be fair, ZFS is a lot more than just a filesystem. It encompasses snapshotting, redundancy and volume management as well.

    11. Re:ZFS rocks by jguthrie · · Score: 1

      ZFS appears to combine all the benefits of RAID6, LVM, and a high-performance file system all in one package. Now, I recognize the benefits of each of those technologies: RAID6 gives you powerful redundancy and reliability features and combining RAID6 with LVM on hot-swappable drives gives you all kinds of flexibility about how much storage you have and how you allocate it, and XFS on top of that gives you a file system that you can expand if you need to without umounting it. What is less clear to me is the benefit of combining all that functionality into a single logical thing. I suppose there is some performance benefit to be gained, but any performance gain would have to be pretty massive for them to be noticeable on modern processors. Besides, file systems are already massively complicated things and I would think that adding complexity to it would just make it fragile and I shudder when I think about the sort of recovery tools you're going to need to fix a hosed ZFS file system. Not that I won't ever use ZFS for anything, but I'm certainly not going to be an early adopter.

    12. Re:ZFS rocks by netcrusher88 · · Score: 1

      Actually btrfs is looking very much like a ZFS-killer, as it were - it does volume management, snapshots, redundancy... If the FS is half as good as the published concepts, it may become the Linux equivalent to ZFS.

      In fact some of the more interesting threads that made it to kerneltrap recently deal with btrfs, in some cases with the developer doing certain programming gymnastics to avoid violating layering in the kernel (I'm not sure I have the terminology right). This is, incidentally, why Reiser4 is not in the kernel - because Reiser was a dick, and among other things Reiser4 violated boundaries between parts of the kernel in a big way, and he refused to fix it. That's another interesting lkml thread that can be found on kerneltrap.

      --
      There's an old saying that says pretty much whatever you want it to.
    13. Re:ZFS rocks by Enderandrew · · Score: 1

      I'm not a kernel dev, but I read the LKML for a long time. Reiser4 works currently the way it is written, by writing a plugin system that basically only Reiser4 uses.

      Some others suggested rewriting the VFS that all file systems in Linux go through, basically placing in many of the improvements that Reiser4 offers, and allowing other file systems to load the "plugins" as well if they so desired.

      It seems at one moment, Linus was against the entire idea, and the next he'd state Reiser4 won't get in until that happened.

      Hans Reiser said he was out of money, and didn't have the means to basically rewrite both Reiser4 and the Linux VFS, so he wanted it to be included as is.

      If brtfs is running into situations where it needs to alter the VFS, then it should propose seperate patches to do that. I don't think the Linux kernel devs were against changing VFS. In fact, it seemed most of them prefered doing that, rather than having a redundant, different means of performing a task within your FS, rather than placing that code in the middle where everyone can get to it.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    14. Re:ZFS rocks by Anonymous Coward · · Score: 0

      Can you create a filesystem in the time it typically takes to create or delete a folder?

      Yes, folders are abstract file systems.

      Can you take snapshots and/or clones, quickly, so that one always has a history of each and every file on disk, and easily accessible?

      Yes, with Subversion.

      Can you create automatically-expanding storage pools?

      Yes, with folders.

      ZFS is nice, but it's "just" a new abstraction layer to get around our poor, customary use of the current abstractions.

    15. Re:ZFS rocks by GaryOlson · · Score: 1

      ZFS does have issues with NFS though. In particular NFS writes can lock up the client.

      Long term, I'd rather see the best technologies of Linux and Solaris merge...

      I prefer to keep the server space and the user space seperate. Solaris with ZFS on the storage providing iSCSI targets; Linux systems for user space using storage provided by the iSCSI connections. Linux NFS client connections to other Linux systems if required. Best of both worlds.

      --
      Every mans' island needs an ocean; choose your ocean carefully.
    16. Re:ZFS rocks by jhol13 · · Score: 1

      The problem is that too many individuals that you can't even contact own individual copyrights in the Linux kernel.

      Although I agree with this it always amuses me. You remember when SCO claims were "on the air" Linux people claimed "we would rewrite it in the matter days" ... apparently these cannot be "rewritten" :-).

      I think there is just no desire to move (I'll leave the reasons for the lack of desire for another time).

  8. I have a problem with this kind of "open source" by Spy+der+Mann · · Score: 1, Interesting

    Sure, it's Open Source and everything. But the problem is that complex programs like this are often designed with a top-down instead of a bottom-up approach. I mean, this isn't a bazaar, it's a cathedral. Oh, and OpenSolaris is not GPL. *buzz*

    There's still one company responsible and only that company will make the changes, because the codebase is so huge that it's a pain in the *** to maintain. Well, eventually many open source projects end up like that, with a huge codebase and with a company. BUT, this wasn't built by the community and it's not likely that it will get enough userbase so that a dev will become interested.

    Compare this with the Linux Kernel. Linus message, which was more or less like "Hey guys, I'm making a unix-like kernel, anyone want to join?", was followed by a stampede of developers and testers like you can't imagine.

    So, if we want competition, it's very improbable that a cathedral project such as OpenSolaris can compete with Linux, a 100% GPL'ed project built by the whole community.

    Maybe it can compete with the bsd clones, but Linux? I don't think so.

  9. Slashdotted?! Whatever. Site's fast and snappy. by Creepy+Crawler · · Score: 1

    Sun is battling hard to break into the open source operating system world with OpenSolaris. Juliet Kemp takes it for a test-drive, sampling its unique features and seeing how it fares against Linux...

    OpenSolaris is an open-source project based on (some of) the Solaris operating system code, and sponsored by Sun, but being developed independently. The main aim of the project is to create a downloadable codebase. Currently, though, there's a Live CD/install image available which gives you a full OpenSolaris distro - a project which arose after Ian Murdoch (founder of Debian) was hired by Sun in 2007 to head "Project Indiana". OpenSolaris 2008.05 was released in May 2008. It is, as advertised, a full-featured distro, which includes the GNOME desktop and the ZFS filesystem (which does snapshots and some other interesting things - more on that below).

    It's released under the CDDL (an open source copyleft license based on the Mozilla Public License), and can be downloaded from the OpenSolaris website. The majority of the source code is fully accessible, but some components are only available in binary form (under the OpenSolaris Binary License).

    The OpenSolaris desktop (click for full size)

    Installation

    Installing OpenSolaris is pretty straightforward - it's a LiveCD (as is increasingly common in Linux distros these days), with standard click-to-install. You answer a couple of questions about your location, keyboard map, and time/date, and you're also asked about disk partitioning. Solaris uses ZFS rather than ext3, which is important in that Linux support for ZFS is still in the fairly early stages.

    The rest of the install is handled for you, and worked fine for me. It picks up the network fine (at least, it does if you plug the cable in...). I didn't test it with wireless, but wireless support is supposed to be available.

    In use

    The first thing to note is that it's not lightweight. My test box is fairly old and slow, but I found OpenSolaris significantly slower than Ubuntu or OpenSUSE on the same box. So it's not really useful for putting that ageing hardware to work.

    It comes with a Gnome 2.20.1 desktop by default, and consequently looks pretty similar to more or less any Linux Gnome desktop. The usual array of applications are in place to start off, including Firefox 2.0.0.14, Thunderbird, Rhythmbox, and so on. OpenOffice isn't installed, but version 2.4.0 is available via the package manager, which uses the Image Packaging System (IFS) software.

    Is there really a need to have 'SUNW' before everything?

    There is documentation available showing how this package manager compares with apt-get - there's also a graphical option if you prefer that to the command line. Both deal with dependencies for you, as with Debian's apt-get and aptitude. There are fewer packages available than for a mainstream Linux distro, although they do have over a thousand (and certainly enough for a fully-functioning system). The package naming is slightly odd; package names begin with a handful of capital letters (eg SUNW or FSW).

    Networking works differently to Linux - ipconfig exists but has a different syntax, and eth0 isn't the standard interface. There's a graphical networking manager, but it gives an error message if started when the network management tool nwamd is configured, which is true by default. This seems wrong: either the graphical tool should play nicely with nwamd, or it shouldn't show up on the default config menu.

    Services and the starting/stopping of them works differently from Linux, as well. Instead of /etc/init.d or similar, OpenSolaris uses smf, the Service Management Facility. Services are referred to as svc:/servicetype/servicename (where service types include network, system, and application, among others) and can be started/stopped via the svcadm command. The man page is helpful, as are the online docs, but it's something that you need to get used to, and as ever there's a learning curve before you'll be comfortable with it.

    --
  10. Why is this bad? by Anonymous Coward · · Score: 0

    Note that by default [NFS] permissions are very paranoid; you have to explicitly allow access to machines or networks.

    So basically when you do it right in linux (as opposed to the openness of windows filesharing), it's bad? Please.

  11. Slashdotted... by Anonymous Coward · · Score: 0
  12. Excerpts from the article: by Futurepower(R) · · Score: 3, Informative

    Excerpts from the article:

    "... I found OpenSolaris significantly slower than Ubuntu or OpenSUSE..."

    "There are fewer packages available than for a mainstream Linux distro, although they do have over a thousand (and certainly enough for a fully-functioning system). The package naming is slightly odd; package names begin with a handful of capital letters (eg SUNW or FSW)."

    "ZFS is transactional, meaning that the filesystem is always consistent (so fsck or equivalent isn't used or needed), and snapshots are intentionally both easy and cheap in terms of disk space."

    "I'm very impressed with the concepts behind ZFS, but I'm also concerned that cross-functionality with Linux is limited."

    "I did find it frustrating to have to relearn commands that I've been using without thinking for years now (eg ifconfig), and right now I'm not convinced that for me it's worth the mental effort, especially given the relative scarcity of external software available."

  13. Mind the install. (Take backups first) by oyenstikker · · Score: 1

    I had a box with a drive with an empty primary partition at the beginning and Linux on a few extended partitions at the end. The OpenSolaris install documentation and the installer itself promised not to touch the existing extended partitions. Which it didn't. It did, however, wipe the partition table so I could not find my extended partitions and had to restore from backups.

    I will not be using OpenSolaris anytime soon.

    --
    The masses are the crack whores of religion.
  14. The syntax for ipconfig is different? by argent · · Score: 2, Informative

    That's because the Linux folks were worried about the pending USG/CSRG lawsuit so they reimplemented TCP instead of using the BSD TCP stack and utilities like almost everyone else (including Microsoft) did.

    Just about any non-Linux UNIX implementation is going to have the BSD TCP.

    On the upside the lawsuit did set SCO up the bomb. Oh, it wasn't the only thing by any means (did they actually do ANYTHING right in that lawsuit?), but one of the side effects of the USG/CSRG lawsuit was that a lot of early UNIX code code was open-sourced. Including some of the SCO claimed were examples of "infringing code" in Linux. Come on, folks, wasn't it great to have Dennis Ritchie himself point that out?

    1. Re:The syntax for ipconfig is different? by lokedhs · · Score: 4, Informative

      Actually, Solaris doesn't use the BSD TCP stack. They completey replaced the stack in Solaris 10.

    2. Re:The syntax for ipconfig is different? by argent · · Score: 2, Insightful

      Microsoft also replaced the BSD TCP stack, but they kept the same userland commands. They already had a different config command (ipconfig instead of ifconfig)... probably because they had to rewrite that to cooperate with their netbios and other stacks, but ...

      C:\WINDOWS\System32\> strings ftp.exe
      [...]
      @(#) Copyright (c) 1983 The Regents of the University of California.
      [...]

  15. It's a bit late by metamatic · · Score: 3, Interesting

    I was a Solaris admin back in the early 90s. I preferred SYSV to BSD for a lot of things. But at this point, I'm just not seeing a compelling reason to go back. Sure, ZFS sounds nice, but I don't really want a system that's slower and more RAM-hungry than Linux, and I don't want an OS with a GPL-incompatible license.

    --
    GCHQ Quantum Insert installed. If only our tongues were made of glass, how much more careful we would be when we speak
    1. Re:It's a bit late by Ash-Fox · · Score: 1

      Honestly, with LVM on Linux, I don't really see the benefits of ZFS over it (note: you can get ZFS over fuse on Linux if you really wanted).

      The only thing I ever hear about what's good about Solaris is the fact it has ZFS and dtrace. Both of which I'm not interested in. Seriously, just go look on sun.com for information about what's good about Solaris. They come up with the regular buzz words such as 'scalable' etc. But the only meat I ever find is 'ZFS' and 'dtrace'.

      --
      Change is certain; progress is not obligatory.
    2. Re:It's a bit late by shutdown+-p+now · · Score: 1

      I don't want an OS with a GPL-incompatible license.

      Why? As a user/admin, what does it matter to you? It's still open source in all meanings of the term. Even as a developer, unless you're hacking the kernel, why would you care?

    3. Re:It's a bit late by DaveV1.0 · · Score: 1

      Why? Simple, he's a zealot.

      --
      There is no "-1 offended" or "-1 you don't agree with me" mod options for a reason.
    4. Re:It's a bit late by metamatic · · Score: 2, Insightful

      Why? As a user/admin, what does it matter to you?

      I've been bitten too many times by companies killing off products. Also, my experience in the past has been that the first thing you need to do with a Solaris box is install all the GNU tools anyway, to replace the broken awk, broken sed, etc. So really, why would I want the non-GNU licensed tools available if I'm not going to use them?

      --
      GCHQ Quantum Insert installed. If only our tongues were made of glass, how much more careful we would be when we speak
    5. Re:It's a bit late by PrettyBoy_75 · · Score: 1

      Thay have the GNU tools in by default now in OpenSolaris and first in the PATH...

    6. Re:It's a bit late by daybot · · Score: 2, Funny

      Shit - my finger slipped and I modded you 'off topic' instead of the intended 'interesting'. Now I have to post this message to remove the mod and set the record straight...

    7. Re:It's a bit late by Anonymous Coward · · Score: 0

      What about when the product developer kills someone off, like with RieserFS? That product is basically EOL even though it was GPL. A project being GPL doesn't really save you from the dev flaking out unless you're willing to take it over yourself.

    8. Re:It's a bit late by shutdown+-p+now · · Score: 1

      I've been bitten too many times by companies killing off products. Also, my experience in the past has been that the first thing you need to do with a Solaris box is install all the GNU tools anyway, to replace the broken awk, broken sed, etc. So really, why would I want the non-GNU licensed tools available if I'm not going to use them?

      It is a valid rationale, but I still don't see what it has to do with Solaris being under CDDL rather than GPL. It would still be just as applicable if Sun releases the next version under, say, GPLv3 (or even GPLv2) - it would still remain a single-vendor OS.

      Tools are a different matter entirely. I install GNU tools on FreeBSD myself as well, because I'm used to extra features they provide, but I do not consider it a major negative point against the OS itself - why? I can still run my tools, that's all that matters; and GNU tools are no more native to Linux than they are to FreeBSD or Solaris (or even MS SUA).

    9. Re:It's a bit late by metamatic · · Score: 1

      I don't see the open source community rallying around to support a CDDL-licensed OS if the vendor loses interest, unlike a GPL-licensed OS.

      --
      GCHQ Quantum Insert installed. If only our tongues were made of glass, how much more careful we would be when we speak
  16. Trivial Review by semerus+(1) · · Score: 1

    This is a superficial review.

  17. My experience by Anonymous Coward · · Score: 0

    While it might be good for people who want to spend a lot of time in the OS, I can't even get that far. I installed Ubuntu on a Sparc machine to test it out. Didn't work so well since there is no good JDK for Linux on Sparc.

    So then I tried to install Solaris back on the machine. It doesn't give you an option to format the disk. Seriously.

    The entire user experience around getting the OS installed and an environment setup is so inaccessible to a person who doesn't want to make it their main focus and job. If you're a Sun shop, or if you're a sysadmin, it might be great.

    But the majority of web 2.0 style folks need a faster way to get going. If you have been used to Linux where the entire experience makes it very easy to get up and running, Solaris is like going backwards 20 years.

    Also, my pet peeve about Linux and Solaris *both* is that they still both fight the GUI vs. CLI war. 99.999% of functions should be accessible from a GUI. Period. End. Of. Story.

    1. Re:My experience by MrNaz · · Score: 1

      It's not so much them believing that CLI is better, I think it's more a case of CLI tools being far easier to develop. Writing a good GUI is non-trivial, and writing a good GUI for something incredibly technical and involved will often result in a harder to use solution than just making the CLI.

      Furthermore, admins are typically highly proficient with a keyboard, and can manipulate the CLI faster than they can a mouse.

      It's a case of tool for the job. Many tasks are just better suited to CLI due to complexity, target audience and design experience of the developers.

      --
      I hate printers.
    2. Re:My experience by PenGun · · Score: 1

      Very funny

      "But the majority of web 2.0 style folks need a faster way to get going."

        Uh Pooky the Web 2.0 thing just separates the morons from the merely slow. You figure out which group you belong in.

    3. Re:My experience by Culture20 · · Score: 1

      99.999% of functions should be accessible from a GUI. Period. End. Of. Story.

      Seriously? So when you're needing to do X, Y, or Z, that takes five programs and pipes data from one program to the other, do you put one GUI window on top of another one so that it can act as a filter, and then another filter on top of that ad nauseum? Even on MS Windows or MacOSX machines, my first action whenever I want to do any real admin work is to open up a command prompt.

      GUIs are like rebuses; they're fun to use, and sometimes can be used by people who can't read, but when you want to get a complex idea across, text is sooooo much better.

    4. Re:My experience by kwabbles · · Score: 1

      ...the GUI vs. CLI war. 99.999% of functions should be accessible from a GUI. Period. End. Of. Story.

      Okay, I'd really like you to tell me how something like that is even remotely possible.

      The only way to give a GUI the same power as the CLI is to dumb down the CLI.

      --
      Just disrupt the deflector shield with a tachyon burst.
  18. Narrow the choice.. by doomicon · · Score: 2, Insightful

    If your main concern is whether or not it runs KDE? Then stick with Linux.

    --

    Awesome!
    1. Re:Narrow the choice.. by doomicon · · Score: 1

      If you work in an environment that has the luxury of dismissing products and solutions because they are NOT GPL...

      stick with Linux :-)

      --

      Awesome!
    2. Re:Narrow the choice.. by True+Grit · · Score: 1

      You can use non-GPLed apps on Linux, clearly, Linux wouldn't have gotten to where it is today otherwise, so what exactly are you trying to say here?

      Do you hate Linux so bad that you'll post troll comments without a lick of logic in them? I mean, heaven knows there are plently of plausible things you can bash it for, but what, you just couldn't wait for the next opening? This is /. after all, its not like reasonable openings for Linux bashing are rare or infrequent...

  19. Solaris has packaging design issues by nobodylocalhost · · Score: 3, Informative

    Trying to harden Solaris is a nightmare. Mostly because so many packages in the Solaris install are interdependent. It is either install 90% of the packages or install nothing. Why do they even bother breaking the software packages if this is the end result? Getting rid of RPC can create so many problems it isn't even funny. Both BSD and Linux offer the option of only installing the base package and only choose the services you want with little to no other packages to depend on. This however absolutely cannot be the case for solaris because a single needed software package will require you to install nearly all services.

    --
    Where is the "Ignorant" mod tag?
    1. Re:Solaris has packaging design issues by pedestrian+crossing · · Score: 1

      With experience, you can figure out a pretty slim install. Yes, it takes some effort. Once you do, I think it is easier to harden than Linux.

      --
      A house divided against itself cannot stand.
    2. Re:Solaris has packaging design issues by alsta · · Score: 1

      You'll get a good start with:

      # svccfg apply /var/svc/profile/generic_limited_net.xml

      You can then further manage the services you want running with modifying `site.xml' and apply that:

      # svccfg apply /var/svc/profile/site.xml

      --
      Wealth is the product of man's capacity to think. -Ayn Rand
    3. Re:Solaris has packaging design issues by Anonymous Coward · · Score: 0

      $ sudo /usr/sbin/netservices limited
      Password: *********
      $

      Wow! You're right, that was a nightmare!

      For those of you without a Sun handy to RTFM ...

      When netservices is invoked with the limited command-line
      argument, all network services except the secure shell dae-
      mon, sshd(1M), are either disabled or constrained to respond
      to local requests only

      Of course, removing unnecessary packages is a good idea ... but I think RPM dependencies in RH are at least as bad or worse than some of the package dependencies I've seen in Solaris.

    4. Re:Solaris has packaging design issues by nobodylocalhost · · Score: 1

      I suggest you read the documentation you posted yourself. Your method isn't only flawed but is more or less hiding your head in the sand.

      First and fore most, there are services responding to local requests, that means they are actually running, and are reachable by using any sort of local reflection attack (e.g. "hey, i broke into this php app, but i am jailed... Time to run all these RPC exploits!"). This is right off the bat a security concern.

      Secondly the existing vulnerable binaries are on the hard disk, as well as vulnerable libraries. Although people can be told not to run certain binaries, building a series of apps and configuring them not to link certain libraries and still keep all the functionalities are more of a challenge.

      I am not trying to plug for linux, I just think solaris definitely has room for improvements, and from all the replies, it seems they are being addressed as we speak or type.

      --
      Where is the "Ignorant" mod tag?
  20. Funny you should ask by shanen · · Score: 1

    I just installed it a couple of weeks ago. Open Solaris starts in a GNOME shell and feels quite like Ubuntu in that way. Main difference is that the booting is much slower, but that's not unreasonable for a server OS that isn't supposed to get rebooted very often. I haven't really used it that much, but mostly it seemed to be okay. (Reference basis is that I'm a heavy Ubuntu user, though my company distro is a custom version of RHEL5, and I've experimented with about half a dozen of the live CD versions.)

    However, overall I still have to rate it as rather betaish. The first major upgrade tends to be fatal when it tries to update GRUB, and I wound up reinstalling pending the fixes. It wasn't just the lack of basic testing that bothered me, but also the unhelpful attitude in the newsgroups: "That's a well known problem." Gee, thanks, so how about a hint of how to fix it? (Yes, I eventually found the description of the fixes, but by then had run out of motivation... I prefer to be virtuously lazy in the Perlish sense and just wait for a more mature product.)

    Disclaimer or statement of limitations or something: I'm running it with the VMware Player, and it's only my fourth client OS, and I certainly can't claim to be an expert in that environment.

    --
    Freedom = (Meaningful - Coerced) Choice != (Speech | Beer^2), and sad sock puppets' bad mods avail them naught.
  21. I love the ReiserFS, Myself by Illbay · · Score: 0

    It's KILLER.

    --
    Any technology distinguishable from magic is insufficiently advanced.
    1. Re:I love the ReiserFS, Myself by Icegryphon · · Score: 0

      wow, that was lowblow. I wonder how many got it.

  22. Solaris is great. So is choice! by jregel · · Score: 1

    As someone who uses Linux at home and work, and also uses Solaris at work, I'm very pleased to see what Sun are doing. Solaris is a great operating system and I'm a bit bemused by the attitude that some Slashdotters have of "Why bother when I've got Linux?". I thought we were supposed to be geeks here and fascinated by interesting technology!

    The biggest grumble I usually hear is that the default Solaris commands are not as feature rich as the GNU equivalents. The easy answer is that the GNU tools are most probably already installed in /usr/sfw and that to use them requires nothing more than a minor tweak to /etc/profile.

    The reality is that, while Linux is great, it's not the only decent open source operating system out there and there are plenty of reasons to look at the alternatives (try it and see for yourself).

    1. Re:Solaris is great. So is choice! by rnturn · · Score: 1

      The easy answer is that the GNU tools are most probably already installed in /usr/sfw and that to use them requires nothing more than a minor tweak to /etc/profile.

      That depends on the site you're working at. My most recent experience is that most Solaris system admins never bother installing anything but the base OS so when you come onto a Solaris system after using most other UNIXes you feel like you've gone back in time when your only editor was ed (OK, I exaggerate; but only a little). Heck, they rarely even run catman after the installation making looking up information in the manpages next to impossible. Great fun when you're moving from Sol[789] to Sol10 in a multi-vendor shop and you cannot remember the new Solaris commands. svcadm threw me for quite a while until I finally got it burned into my synapses. (And it's so-o-o-o much better than "servicename start|stop".)

      --
      CUR ALLOC 20195.....5804M
  23. ZFS and Dtrace are finding thier way into other OS by Anonymous Coward · · Score: 1, Insightful

    I find quite interesting that the coool features are finding their way into FreeBSD but not linux. Makes you think which users really have more freedom.

  24. If you think... by hummassa · · Score: 0, Flamebait

    openSUSE is a good KDE, then you are a lost cause :-)
    disclaimer: Kubuntu user, for a relatively long time now.

    --
    It's better to be the foot on the boot than the face on the pavement. ~~ tkx Kadin2048
    1. Re:If you think... by chubs730 · · Score: 1

      Kubuntu is a poor KDE desktop compared to most of the other majors. Ubuntu is solid, but Kubuntu isn't quite up to par. OpenSUSE is certainly far superior.

    2. Re:If you think... by HiThere · · Score: 1

      I do prefer KDE over Gnome by a significant margin, but I sure don't like SUSE. Kubuntu is acceptable, but I usually use Debian. (Well, the SUSE info is now a few versions back, so things may be different, I suppose.)

      OTOH, you said OpenSUSE. I can't really comment about that, as I don't know it. I don't really have any direct knowledge of SUSE in any dialect since Novell made their deal with MS. (And I'm not likely to unless I hear that they've released significant version upgrades into GPL3. Otherwise I don't trust their deal with MS. [They revealed parts, and they kept parts hidden, and we don't know WHAT the hidden parts say.])

      --

      I think we've pushed this "anyone can grow up to be president" thing too far.
    3. Re:If you think... by Enderandrew · · Score: 1

      openSUSE 11 is the first version (a fairly recent release BTW) where I really started to like SUSE at all. Package management is pretty fast. I've yet to run into any dependency hell. The new solver works well. Yast is a good tool. The install is very fast. The installer will auto-resize a Windows partition and setup dual-boot (as its recommendation, but you can always change the partition scheme during the installer).

      The KDE, KDE 4 and Gnome desktops are all very solid in openSUSE 11.

      I have no real complaints with it so far, and I finally ditched Gentoo with it. I loved Gentoo for a while, but I think they are hurting from lack of management, direction, and package maintainers.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
  25. Linus is right by Anonymous Coward · · Score: 0

    I am with Linus on this one. Our community should shun Solaris

    1. Re:Linus is right by ciderVisor · · Score: 1

      I am with Linus on this one. Our community should shun Solaris

      Shun the non-believer !!! Shuuuuuuuuuuuuuuuuuuuunnnn !!!

      --
      Squirrel!
  26. Re:I have a problem with this kind of "open source by Anonymous Coward · · Score: 0

    you are a complete and total tool.

  27. One significant difference... by Anonymous Coward · · Score: 3, Interesting

    ... between Solaris and Linux in the Enterprise is how they react to abuse - namely stupid people running ton of stupid memory hog applications.

    Where I work we have Solaris 9 and 10 boxes running literally unattended for 600+ days - they are shared boxes, meaning lot many different applications run on the same OS/FS/Memory/CPUs .

    When a particular app goes haywire and starts (many of them are 64-bit apps) - that particular app just gets a NULL back when there is no longer any memory available. The app can hopefully then calm itself down or release some of its caches etc. but the main point is that the other apps are unaffected and so is the OS.

    I would not even begin to think how Linux could handle this. It has this insane notion of handing out virtually any amount of memory to applications whether or not there is actually that much memory and swap available. So when things get out of control the ugly and stupid OOM killer thinks it knows better which app to kill - depending on your luck you could end up with sshd or some other good behaving app being killed to give memory to this bad app.

    That is scary. Arguably this is all fixable within the applications but ground reality is that App developers are incompetent - at least where I work, they are.

    Plus the newer Solaris releases are close to Linux when it comes to performance. So the only incentive to run Linux is hardware support - if you are on non SPARC hardware that is.

    Linux hopefully some day will have a good memory management subsystem soon - less fragmentation, more predictability, good accounting etc. But till that time Solaris for the stupid "Enterprise" .

    1. Re:One significant difference... by Timothy+Brownawell · · Score: 1

      Where I work we have Solaris 9 and 10 boxes running literally unattended for 600+ days - they are shared boxes, meaning lot many different applications run on the same OS/FS/Memory/CPUs .

      When a particular app goes haywire and starts (many of them are 64-bit apps) - that particular app just gets a NULL back when there is no longer any memory available. The app can hopefully then calm itself down or release some of its caches etc. but the main point is that the other apps are unaffected and so is the OS.

      Or maybe it won't, since we already know that it probably isn't written very well (or maybe you just didn't buy enough ram for your server). And other programs will end up unable to allocate more memory when they need it.

      I would not even begin to think how Linux could handle this. It has this insane notion of handing out virtually any amount of memory to applications whether or not there is actually that much memory and swap available. So when things get out of control the ugly and stupid OOM killer thinks it knows better which app to kill - depending on your luck you could end up with sshd or some other good behaving app being killed to give memory to this bad app.

      You do realize that out-of-memory behavior is configurable, right? The default setting is to use the OOM killer, but it's trivial to tell it to behave "properly".

    2. Re:One significant difference... by JumboMessiah · · Score: 1

      Before you get your panties in a bunch over VM accounting you should read this on one of your linux boxes:

      vi /usr/share/doc/kernel-doc-2.6.18/Documentation/vm/overcommit-accounting

    3. Re:One significant difference... by Anonymous Coward · · Score: 0

      http://www.win.tue.nl/~aeb/linux/lk/lk-9.html

      "9.6 Overcommit and OOM"

    4. Re:One significant difference... by shish · · Score: 2, Informative

      I would not even begin to think how Linux could handle this

      Evidently I am somewhat more adventurous, as I will dare to think. My thought is that linux could handle this by having some sort of kernel setting, called, say, "overcommit_memory", classed in the "sys/vm" part of the proc heirachy. I would think that one could alter the behaviour by echoing a setting into it, eg "echo 2 > /proc/sys/vm/overcommit_memory". Incidentally, I think it could work this way because it actually does work this way :P

      A question though -- with overcommit disabled, things like java or wine (which allocate several hundred megs of ram to play with whether they actually use it or not) tend to start failing; how does solaris deal with this? I wonder if it's even noticed, as the last solaris box I saw had 8GB RAM and 32GB swap, whereas this problem is more apparent on my 64MB linux VMs...

      --
      I mod down anyone who says "I will be modded down for this", regardless of the rest of their comment
    5. Re:One significant difference... by parryFromIndia · · Score: 1

      A question though -- with overcommit disabled, things like java or wine (which allocate several hundred megs of ram to play with whether they actually use it or not) tend to start failing; how does solaris deal with this? I wonder if it's even noticed, as the last solaris box I saw had 8GB RAM and 32GB swap, whereas this problem is more apparent on my 64MB linux VMs...

      For every allocation (malloc/sbrk) Solaris reserves that much VM/swap space first before it will succeed the allocation. There is no notion of overcommiting memory (except for MAP_RESERVE).

      Java - it needs a Gig of "Code Cache" and with overcommit turned off if not enough RAM+SWAP is available and it fails the allocation. So you are right - increasing swap space and tuning overcommit_ratio should deal with that.

      See http://developers.sun.com/solaris/articles/subprocess/subprocess.html for interesting discussion around this topic.

  28. It's neat... by ribo-bailey · · Score: 1

    dtrace and ZFS make me cream my jeans, but they'll probably just get ported to other unix-likes (or similar things like tux3 and kprobes will "get there") Not that I don't like opensolaris. After writing some D scripts, I'd like to use it in a few applications.

  29. Re:I have a problem with this kind of "open source by Anonymous Coward · · Score: 1, Insightful

    All spoken as if by a really naive person. You seem to think that Linux is not in competition with BSD, as if it were above BSD (and OpenSolaris). That's a strange position to take. Do you think that Linux' progress is a result of competition with itself or with Windows alone?

    Both BSD and OpenSolaris are technically equivalent to Linux in many areas and superior in some others. By the way, Linux is also superior to the other two in some areas, too. You would be a fool to discount any of these operating systems, as each has pros and cons associated with different projects and needs.

    Furthermore, if your hypothesis is that OpenSolaris won't be able to compete with Linux because of the "community" aspect, then why would you suggest that it may be able to compete with BSD which also is developed by the "community." That reeks of Linux fanboyism and illogical thinking.

  30. Re:I have a problem with this kind of "open source by Spy+der+Mann · · Score: 1

    I was not talking about the technical aspects, but about Market Share. It doesn't matter whether your product is superior or not, existing market share is a horrible obstacle to overcome.

    OpenSolaris vs. Linux = Linux vs. Windows

  31. Re:ZFS and Dtrace are finding thier way into other by MobyTurbo · · Score: 1

    I find quite interesting that the coool features are finding their way into FreeBSD but not linux. Makes you think which users really have more freedom.

    Linux's kernel uses the GPL, and CDDL is GPL-incompatible. (But BSD and Apple's OS X open-source license (whatever it's called) compatible.) As a result, BSD and OS X have DTrace, and ZFS is in BSD and is coming to OS X. However, it's not really the fault of the Linux developers that it doesn't have it in the kernel, Sun made yet another open-source license that's incompatible with GPL. (It's pretty hard to make a license incompatible with BSD, the only conditions are, "here's some code, don't sue us".)

  32. Re:Get your head out of the sand by larry+bagina · · Score: 1
    And what if you submit a Linux patch? Somebody has to approve it. Except with Linux, those somebodys have massive egos and reject outsiders (cf - yesterday on slashdot). I don't know if Solaris is any better, but being paid professionals, they're less likely to act like an asshole.

    In either case, you're welcome to take the source code and start your own kernel. Try doing that with MS Shared Source 2008 (tm).

    --
    Do you even lift?

    These aren't the 'roids you're looking for.

  33. Re:Get your head out of the sand by Anonymous Coward · · Score: 0

    t's stupid nonsense like this article and comments that makes business line people want to ignore what the real IT people who have some sense know is true.

    It's angry, antagonistic comments like yours that make the real IT people who have some sense mod you troll. It's clear you have some serious opinions, why are you being such a dick about them?

  34. Gee, sounds familiar. by DaveV1.0 · · Score: 1, Funny

    I did find it frustrating to have to relearn commands that I've been using without thinking for years now (eg ifconfig), and right now I'm not convinced that for me it's worth the mental effort, especially given the relative scarcity of external software available.

    Those sound like the same complaints Windows users have of Linux, but which continually get dismissed by the Linux community as irrelevant.

    "It's not Linux. I have to learn new commands and doesn't run my programs"

    --
    There is no "-1 offended" or "-1 you don't agree with me" mod options for a reason.
    1. Re:Gee, sounds familiar. by BiggerIsBetter · · Score: 1

      "It's not Linux. I have to learn new commands and doesn't run my programs"

      Too true. I wish I had mod points...

      (this comment posted from Ubuntu)

      --
      Forget thrust, drag, lift and weight. Airplanes fly because of money.
  35. Re:Get your head out of the sand by Rhabarber · · Score: 1

    > being paid professionals, they're less
    > likely to act like an asshole.

    That statement is based on what data?
    Could you please provide a reference?

  36. Just add GNU by pedestrian+crossing · · Score: 1

    If you replace a few choice utils with their GNU counterparts, Solaris is just fine to use. You can have the best of both worlds.

    --
    A house divided against itself cannot stand.
  37. Same here by kwabbles · · Score: 1

    Actually - I'm in the middle of a nightmare because I need to get this system out the door and to a client like... YESTERDAY, and for some reason the pkg.opensolaris.org server is having problems and I can't get this thing updated.

    Aside from my grief this morning, my experience with Indiana has been pretty nice. It seems like once things are working, for the most part they stay that way. However, I'd agree with your statement about it being "betaish". The system has some wierd quirks like:

    1. Occasionally when rebooted it cant bring up SMB shares. Reboot again and they come up.
    2. Network Auto Magic... uhhh.... huh? Why? First thing I disabled.
    3. Changing DNS servers in the config and reloading = broken DNS. Reboot and it's fine.

    Anyway, I'm glad Sun has done this. It's always nice to have more "free" alternatives.

    --
    Just disrupt the deflector shield with a tachyon burst.
  38. Its a MacBook Pro by Anonymous Coward · · Score: 0

    its supposed to suck.

    1. Re:Its a MacBook Pro by Anonymous Coward · · Score: 0

      I thought the MacBook was supposed to suck and the MBP was supposed to blow....

  39. Re:What about hardware support? by uassholes · · Score: 2, Informative
    You can check your hardware for solaris compatibility at this site: http://www.sun.com/bigadmin/hcl/

    I have had driver problems with Linux as well. Does it have such a site?

    PS. You can also check the forums at http://opensolaris.org/os/

  40. Pretty good, but will hold out for the next one by motang · · Score: 1

    I have been playing around with OpenSolaris since it came out in VirtualBox and it seems cool. But I have had some issues with it so I shall hold out till the next release which isn't too far away.

  41. NexentaStor by chenjeru · · Score: 1

    Nexenta has released an OpenSolaris based storage server platform which we're using as a storage virtualization head. It rocks.

    --
    Even if you're on the right track, you'll get run over if you just sit there. - Will Rogers
  42. Re:I have a problem with this kind of "open source by bigstrat2003 · · Score: 2, Insightful

    Cause anything which isn't GPL isn't open source, amirite?

    --
    "16MB (fuck off, MiB fascists)" - The Mighty Buzzard
  43. Been using it... by PrettyBoy_75 · · Score: 1

    Since June (OS 2008.05) with nary a complaint.

    One of the cooler features that wasn't mentioned in the review is the snapshot ability of the ZFS. With OpenSolaris they have tied it into upgrade process so that a snapshot of old boot environment is created automatically (and placed in the GRUB menu) and you can switch between the two at boot time and manage which one takes priority via beadm pretty seamlessly.

    Give it a whirl...you might like it.

  44. Mixed Feelings by Piranhaa · · Score: 2, Interesting

    OpenSolaris takes quite a bit of time getting used to IMHO coming from FreeBSD, OpenBSD, and many many versions of Linux. I used it because I wanted ZFS, virtulization, and also to try something new.

    I did move back to FreeBSD after about a week or so since I thought OpenSolaris brought unnecessary learning curves for someone new. Things like 'ps' being different than every other distro, network interface setup and modification is annoying, the number of programs that you can compile outside of their package manager are slim, and overall not very friendly (I don't want to use the GUI, ever). However, I have 4gb of ram and ZFS really should only be run under 64-bit FreeBSD. Qemu doesn't seem to run, Xen isn't even an option for virtulization and WINE doesn't work under 64-bit (these are the main reasons I bought 4gb of ram in the first place).

    ZFS has been running flawlessly on FreeBSD for me thus far, and even the maintainer says he's been using it since her ported it over without a hickup. FreeBSD runs version 6 of ZFS, while OpenSolaris currently runs version 11. It IS true, once you go ZFS you don't go back.

    I refuse to run Linux, for personal and limiting reasons, and FreeBSD won't let me virtualize. It seems that in the next few days I'll be biting the bullet and moving back to OpenSolaris. It is very nice that ZFS is seamlessly integrated and snapshots are automatically created when updating the system. This ensures you can easily roll back or boot back into an older install to test different things.
    All in all OpenSolaris HAS some potential, but their licensing is very wack and limiting. If Sun wants their OS to evolve and take on more users in the community, the licence will really need to be changed.

    1. Re:Mixed Feelings by uassholes · · Score: 1

      Former BSDers will be happy to know that UCB versions of a lot of commands can be found in /usr/ucb

    2. Re:Mixed Feelings by uassholes · · Score: 1

      PS. Including ps

    3. Re:Mixed Feelings by Anonymous Coward · · Score: 0

      You should be able to have r/w support on the FreeBSD ZFS using any recent (Open)Solaris version because they have backwards compatibility. You can also have r/w support with FUSE-ZFS.

      Using (Open)Solaris you do get a lot of nice features like virtualization, but the question is whether all your hardware works. Mine didn't, my Areca controller wasn't stable on (Open)Solaris... :-(

  45. Re:Get your head out of the sand by Anonymous Coward · · Score: 0

    Heck, even Itanium beats SPARC these days. Sun needs to get its act together. ROCK is like 2 years late already...

  46. Re:What about hardware support? by Anonymous Coward · · Score: 0

    Yes, Linux has HCL as well.

    But basically, it isn't required. Almost everything just works.

  47. ZFS and GPL by Anonymous Coward · · Score: 0

    Someone likely could easily issue ZFS as patch directly for the kernel as opposed to in FUSE. The problem is that it would be illegal to use it a such because of the license.

    Says who? Can we get someone with a law degree to opine on this?

    If (closed source) NVidia can link into the kernel without issue, I'd like to hear the logic behind why (OSI-approved) ZFS cannot.

    What specifically are the conflicts between the two licenses? Also, why cannot a shim layer (like NVidia?) be used if there are actual conflicts?

    Anyone?

    1. Re:ZFS and GPL by Timothy+Brownawell · · Score: 1

      If (closed source) NVidia can link into the kernel without issue, I'd like to hear the logic behind why (OSI-approved) ZFS cannot.

      NVidia doesn't mind when you link their code to the kernel. ZFS is GPL3, which says you can only link to GPL3 or things that can be treated as GPL3.

      Also, why cannot a shim layer (like NVidia?) be used if there are actual conflicts?

      There's no need, ZFS is open source so the end user can just compile the entire thing. The final linking just has to be done by the end user, since the result can't be distributed.

    2. Re:ZFS and GPL by Enderandrew · · Score: 1

      ZFS is still under the CDDL currently. Sun said they'd released ZFS under the GPLv3 if the Linux kernel went GPLv3.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
    3. Re:ZFS and GPL by Timothy+Brownawell · · Score: 1

      Ah, I thought I'd seen someone say they went ahead and did that anyway... do you know if CDDL has similar linking rules, or something else that would block the module approach from the ZFS side?

    4. Re:ZFS and GPL by Enderandrew · · Score: 1

      No, I don't believe the CDDL has blocking rules, which is why FreeBSD was able to pick up DTrace and ZFS. However, the GPL has blocking rules which is preventing the inclusion of ZFS in the Linux kernel.

      --
      http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
  48. Solaris has packaging changes being intoduced by Anonymous Coward · · Score: 0

    Trying to harden Solaris is a nightmare.

    Use JASS:

    http://www.boran.com/security/sp/Solaris_hardening4.html
    http://www.sun.com/software/security/jass/

    Both BSD and Linux offer the option of only installing the base package

    BSD maybe, but the most popular Linux distros install the kitchen sink and interlink everything (especially RPM-based ones). For Solaris, release 10 allows a base install less than 200 MBs using the SUNWCrnet software cluster:

    http://www.securitydocs.com/library/2644

    only choose the services you want with little to no other packages to depend on.

    Being worked on in OpenSolaris:

    http://opensolaris.org/os/project/pkg/

  49. The AdvFS myth ! by Macka · · Score: 2

    just karma whoring here, but it's important to mention that pretty much everything ZFS has to offer was already available on tru64's advFS: http://advfs.sourceforge.net/ [sourceforge.net]

    As much as I love tru64, I think it really is time to put this myth to bed. AdvFS is a good solid filesystem and cluster aware too, but it's no ZFS. AdvFS doesn't do any form of RAID other than concatenating disks into disk pools (domains) which can then be populated by filesets (AdvFS speak for filesystems) that share the same domain space. Every enterprise implementation of AdvFS always always has AdvFS sitting on top of some form of hardware or software RAID. AdvFS itself doesn't provide any RAID like data protection or redundancy.

    AdvFS doesn't come close to the flexibility and power of ZFS. That's just the plain truth.

    Now if you're really interested in Linux's answer to ZFS, you should keep an eye on the development of Btrfs. I wouldn't expect it to be production ready for a couple of years yet, but when it is it should kick ZFS ass!

  50. md broken? by castrox · · Score: 1

    You mean the fact that md is completely broken

    Okay you had me worried there for a sec.. I run an md rig of 4 disks (raid1 and raid5). I think it works beautifully. Never had any problems in 1.5 years since I installed it. So in what respect is md "completely broken"?

    --
    Fight for your digital freedom, join the EFF *now*: http://www.eff.org/support/
    1. Re:md broken? by outZider · · Score: 2, Informative

      Only when it forgets where drives are, or wipes configuration during upgrades, or the fact that it's slower than software raid setups in FreeBSD or even Mac OS X.

      md on a single box occasionally works. Managing 120 machines with md became a reason to never use md again.

      --
      - oZ
      // i am here.
    2. Re:md broken? by castrox · · Score: 1

      No offense but wiping out configuration during upgrades sounds a lot like a poor admin job. Forgetting drives sounds like a hardware issue.

      Of course, I only run md on one machine and not hundreds of them so I guess I can't relate. I'm quite happy with the performance as well - it's near the theoretical maximum considering a configuration of 4 disks each maxing out at 65 MB/s read.

      --
      Fight for your digital freedom, join the EFF *now*: http://www.eff.org/support/
    3. Re:md broken? by outZider · · Score: 5, Interesting

      No offense taken, but your first line is a typical response from most linux users -- anything that goes wrong is either an admin issue or a hardware issue.

      First, md forgetting drives is not a hardware issue. Linux sees the drive, the serial number of the drive is the same, the hardware does not change, the hardware works. Sometimes, you will boot, and it just loses the configuration, so you reconfigure the array, and wait for it to check everything out. For two hours. At 3am.

      Wiping out configuration during upgrades happened for two consecutive releases of the master distribution. Everything is backed up, but 3/4's of the machines didn't boot properly after md was upgraded. Turned out this was a pretty known issue. No one ever thought that people would want things migrated. Everyone seems to have a few hours to manually move arrays over.

      Look, I'm all for great, open, free technology. The problem is, most people don't think about the big picture. LVM and MD are fine for personal machines that don't do much more than serve up files, or play music, or what have you. Technology like ZFS is designed to be bulletproof, documented, and it has to be supported. Not only that, but given the right amount of RAM, ZFS can outperform many off the shelf RAID systems, and give you flexibility in mirroring, snapshots, and drive support that LVM cannot possibly compare to.

      The only reason ZFS hasn't had much news in Linux land is that it 'wasn't invented here' and it isn't GPL. Last I heard, there was a movement underway to reimplement ZFS under the GPL. I would imagine we'll see something in five years or so.

      --
      - oZ
      // i am here.
    4. Re:md broken? by Bryan+Ischo · · Score: 1

      Just admit it. You were wrong when you used the term "completely broken". You were exaggerating just for the sake of adding flourish to your post and this person caught you in it. Just admit it already.

      I believe that md does have problems given your further explanations. But to say that it is "completely broken" when obviously many people use it and the other poster claims success for his setup, is just plain wrong.

    5. Re:md broken? by nxtw · · Score: 1

      Only when it forgets where drives are, or wipes configuration during upgrades, or the fact that it's slower than software raid setups in FreeBSD or even Mac OS X.

      md doesn't forget where drives are: if md is configured properly and there are enough connected devices to start the array, it will work. md arrays shouldn't be configured by describing the individual members; I set up my arrays by specifying the UUID.

      md does not have a configuration tool that modifies the configuration file; any configuration is done by hand or by some other tool. It's more likely that your package manager or distribution's own scripts are deleting the md configuration file, possibly due to your own (in)action while upgrading.

    6. Re:md broken? by outZider · · Score: 1

      Okay, I'll give in. md on Linux is mostly broken, but is almost usable for personal needs, as long as you have the time to take care of it when things go wrong.

      --
      - oZ
      // i am here.
    7. Re:md broken? by mikelieman · · Score: 1

      Why didn't you discover the configuration-wipeout issue during Upgrade Configuration Testing? Not enough test cases to properly simulate the combination of hardware in Production or something?

      --
      Technology -- No Place For Wimps! Grateful Dead and Jerry Garcia Chatroom -- http://www.wemissjerry.org
    8. Re:md broken? by outZider · · Score: 1

      Unfortunately, we only had two machines available for testing, representing the two configurations of hardware we were using. The upgrade went off without a hitch.

      Hooray, surprises.

      --
      - oZ
      // i am here.
    9. Re:md broken? by Wolfrider · · Score: 1

      --FYI, I am a big fan of ZFS; but the Next Big Thing in Linux-land is rumoured to be BTRFS. IIRC, Kerneltrap had a post saying the head coder of BTRFS was aiming to do things better than ZFS. (Either that or it was HAMMERFS for Dragonfly BSD.)

      --
      .
      == WolfriderV6 == I'm willing to admit that *I just might* be wrong... Are you??
    10. Re:md broken? by Anonymous Coward · · Score: 1, Informative

      I'll wholeheartedly agree with you. I run Slackware mostly at home for my SAN. Its an iSCSI backend for my desktops and servers... I'm a software development major; a lost drive is a lost portfolio (because you know the backup DVDs always suffer bit rot by the time you need them!), and during testing I can easily blow through a few VMs and/or both my desktop machines. I need a lot of storage for keeping a nice library of VMs around that I can boot at a moments notice, without saturating my development boxes buses, distro ISOs, and snapshots of all of the above. Naturally, I can't get enough storage or bandwidth and most solutions just don't cut it when I want to run three VMs (snapshotting one, rebooting the other, and reverting to a previous snapshot on the other) while doing an SVN commit during an hourly rsnapshot and initializing a hot swapped eSATA drive.

      OK, I'm not usually that bad, but I could be if everything wouldn't start crawling as soon as the second I/O operation started. Anyways, when I first started building my storage box last summer, I had heard that ZFS was the final word in storage. Period. The school has a SPARC Solaris box that the development students use, and I've never been a huge fan of it, but I had also never heard a file system described as sexy before ZFS. After fighting with OpenSolaris (funny how you get a GUI installer and on first boot, it tells you that your graphics card is unsupported!) for about a full day, setting up ZFS was the most pleasant experience I've ever had with a piece of software. I had a ZRAID pool up and running after about five commands and perhaps a small conf file. I could move storage around like it was water; it flowed however I wanted it to. And it was fast. I mean, FAST. It would take whatever I could throw at it while doing a snapshot and not break a sweat.

      In the end, OpenSolaris just couldn't deliver; as great as ZFS was, it just wasn't worth babysitting OpenSolaris. I just couldn't help detesting the user land apps with such lacking command line options and a rather fugly directory structure. What, you want to use bash as root? Hah! You get sh - and a stern warning against changing it.

      So, the box is running Slackware (it's rock solid stable, and I've yet to find software that won't compile on it... OpenSolaris blew up while patching twice in the few weeks I was running it, without allowing me to back out the patches, of course!) with md and multipath. Sometimes md decides to rebuild for no apparent reason, every time I reboot multipath renames my md devices so I have a script that figures which is which and mounts them accordingly in rc.local, and then starts the services that need the data contained within. MD is also very iffy when it comes to hot swapping in drives... it will refuse to acknowledge that the drive I just added is part of the device it degraded until I force it to check the metadata and then reinitialize (or was it rebuild or restart or reload the configuration file that only applies to devices with version 1 metadata, or the repair for version two metadata? It's not like it writes any of this to the disk... which is good if you're using RAID 10, as you have to start them in two steps!) If I wanted to do something for snapshotting my options limited as LVM only works for devices (I can't do my usual trick of dd'ing a 8 GB flat file and laying a file system over it so that I can gzip the sparse file and open it somewhere else or mount it on a loopback). Also, LVM is just slow. No matter how I align my PV stride to my RAID device, it just doesn't seem to fall in sync well.

      Posting AC since I wrote this entire thing just to find out that I modded this article earlier today. Nice. Can we get some kind of reminder before we pen the entire War and Peace in a 2"x2" block, please?

    11. Re:md broken? by cerberusss · · Score: 1

      didn't boot properly after md was upgraded. Turned out this was a pretty known issue. No one ever thought that people would want things migrated

      That's typical Linux for you, however, I'd expect your distribution vendor to take this kind of stuff into account. Which distro do you guys run?

      --
      8 of 13 people found this answer helpful. Did you?
    12. Re:md broken? by jacquesm · · Score: 1

      super post, that was well worth reading. And yes, the 'can't post without losing modpoints' really sucks, I think that should not be done for the whole article but on a per thread basis.

    13. Re:md broken? by gullevek · · Score: 1

      yeah, only when you setup a new box, create an md raid and it can't auto-built on boot because of some obscure reason. Now you need to created the raid via initrdram shit and that is not fun if your boot disk is raid too.

      the only way to circumvent this BS is to forcefully create a raid system with a lower version for its super block.

      If they work, they work, but the daily fear of a some HD error and the disk getting kicked out is not fun.

      I would never use MD on any critical production machine, never ever. (neither would I use LVM ...)

      --
      "Freiheit ist immer auch die Freiheit des Andersdenkenden" - Rosa Luxemburg, 1871 - 1919
    14. Re:md broken? by NateTech · · Score: 1

      Yeah, he meant to say "completely irrelevant". Kinda like a Ford Model T isn't a "completely broken" car.

      Does it do things other cars do? Yes. Well? No.

      Would you ever want one for anything other than a collector's item after driving a Ferrari? No.

      There are filesystems that real pros use and filesystems that crash and can't be recovered without backups.

      And there are some pros who are stuck with the latter or who've never enjoyed real filesystems saving their ass at 3AM, so they just "don't know", but loudly tout that their kiddie filesystems keep up, without having used the real ones.

      --
      +++OK ATH
    15. Re:md broken? by Anonymous Coward · · Score: 0

      Who is using software-RAID these days anyway? Adding a RAID card to your server spec won't run you out of pocket - in fact, it's so cheap it doesn't make sense to remove it.

      On package management, I agree solaris is a bad joke. Maintaining a fully-patched system is a nightmare compared to yum update, apt update.

  51. Try OpenVZ by Macka · · Score: 1

    Zones. Because sometimes full-blown virtualisation is too much like hard work

    On Linux you get the same functionalty from OpenVZ or its commercial cousin Virtuozzo.

  52. Re:What about hardware support? by jedidiah · · Score: 1

    This is what started to kill Solaris in the first place.

    Linux? Oh yeah it will support the hard drives you have and the cdrom you have.

    Solairs? Sorry, you will need to replace all of your hard drives and your cdrom with
                      dramatically more expensive equivalents.

    If you are really dying to run the x86 version of Solaris I am sure there is
    a suitable x86 box that Sun is just dying to sell you. It will also simplify
    things when it comes to support.

    "You mean you wanted to use 3rd party RAM? I'm sorry but that just won't do."

    --
    A Pirate and a Puritan look the same on a balance sheet.
  53. ulimit by DragonHawk · · Score: 1

    When a particular app goes haywire and starts ... that particular app just gets a NULL back when there is no longer any memory available. ... I would not even begin to think how Linux could handle this

    Ummm... forget about the OOM-tuning stuff the other people are preaching, how about the good old ulimit command? It's only been around since... System V Release 4, according to ulimit(3).

    http://certcities.com/editorial/columns/story.asp?EditorialsID=214

    Maybe Linux's handling of memory limits could be more intelligent (I have no idea), but that's no excuse to ignore basic Unix process accounting.

    --

    dragonhawk@iname.microsoft.com
    I do not like Microsoft. Remove them from my email address.
  54. Linux Gets Completely Fair Scheduler by Macka · · Score: 1

    Ha, just turned up a Slashdot article about this from last year:

    http://linux.slashdot.org/article.pl?sid=07/07/10/2335217&from=rss

  55. Clarification by Enderandrew · · Score: 2, Interesting

    Sun has not yet released ZFS an openSolaris under GPLv3, which is the first step.

    Next, the Linux kernel would need to be GPLv3.

    Linus can relicense all of his contributions to GPLv3, but then the kernel can not include any code currently licensed GPLv2. So actually, every developer who contributed code and maintained copyright on that code would have to be contacted, and all of them would have to agree to relicense the code.

    Unlike many other projects where people contributed under "GPLv2 or later", the Linux kernel is basically all "GPLv2 specifically".

    It would be a logistic nightmare to relicense the Linux kernel, and many developers have stated they would be opposed to it on principle as well.

    Sun could just license ZFS and the openSolaris kernel under GPLv2, but in their eyes it would be effectively giving it away for nothing. The reason they'd consider GPLv3, is to entice the Linux kernel to go the same route, and then both can take from each other.

    When Sun discovered that Linux wasn't likely to go GPLv3, they decided not to either.

    --
    http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
  56. I thought it sucked. by Red+Leader. · · Score: 2, Insightful

    I gave Solaris 10 more than a fair shake a few months ago (with an eye on its ZFS support) when I had a hard drive fail. I worked pretty hard at getting it to run and really didn't get very far. Note: I've been using Debian for almost 10 years now -- so I'm pretty biased.

    From what I remember there was an astroturfed Sun-staff-only developer community, little information available online, slow as hell boot time, ZFS boot partition complications, and a broken KDE (the X server didn't work correctly; I have absolutely ZERO problems, even with 3D here in Linux).

    And when I looked ahead to maintaining the system (the VAST BULK of where overhead is spent) I didn't see anything that looked as sane or easy as Debian. No incremental updates, just some arcane BSD-esque 'port' or .tar.gz package system (excusable for the rare unpackaged Perl module, but unacceptable for the whole damn system). I'm quite admittedly not very knowledgeable about BSD and Unix, but damn those systems seem like a bitch to maintain. And Nexenta simply wasn't there yet.

    Solaris 10: pass.

  57. Re:What about hardware support? by Alpha830RulZ · · Score: 1

    Oh? Try using an Atheros 5007 chipset for wireless.

    --
    I was taught to respect my elders. The trouble is, it's getting harder and harder to find some.
  58. sun is a small company...... by Anonymous Coward · · Score: 0

    when comparing to the linux community.....
    then why they think they can drive......
    noway......
    sun is nothing better than linux......unless they r strong like bsd.....
    if not.....supporting the most popular and latest things is the only way to go......
    thus.....it's sun-setting already....
    yawn~~~~

  59. KDE on solaris by anilg · · Score: 1

    You've not heard of belenix have you?
    www.belenix.org

    --
    http://dilemma.gulecha.org - My philospohical short film.
  60. Re:I have a problem with this kind of "open source by Eighty7 · · Score: 1

    All OSS licenses are equal, but some are more equal than others ...

  61. blastwave is back up by khb · · Score: 1

    As ZaMoose observed; without any clues as to the backstory.

    1. Re:blastwave is back up by khb · · Score: 1

      has dclarke's story.

  62. ZFS of course by onglipo · · Score: 1

    One of my FreeNAS mirror disks broke and switched to an OpenSolaris NAS . Happy with all the additional functionality.

  63. Re:What about hardware support? by Anonymous Coward · · Score: 0

    Why would I? I have a ZD1211B, IWL3945, RT61, RT2500. They all work on Linux. Most of them work on FreeBSD and OSX as well. My Areca RAID controller however doesn't work stable on (Open)Solaris. And buying a new RAID controller (while my current one is top notch, and is in use) is quite a task.