Slashdot Mirror
IBM Promises Logical Volume Management For Linux
An Anonymous Coward writes: "I found the following message posted to several Linux mailing lists:
************************* Begin Message **************Well, the message sounded interesting so I checked out their website. I found the specified white paper in the "Documents / White Papers" section. After giving it a quick read, I must say that this isn't like any other LVM I've ever heard of! The use of "plug-in modules" to control what devices the LVM sees, how it partitions those devices, and how it uses those partitions to create a volume is incredibly flexible. I also like the part about the LVM working with the filesystem on a volume so that a volume can be resized without any data loss (and without all of the manual steps that must currently be done on Linux). The paper discusses this LVM as an architecture, and it avoids discussing user interfaces and implementation issues. I wonder if this is some pie in the sky vision or if they have any code to back this up?"Hello! Since IBM has begun to publicly support Linux, many of our customers have started showing an interest in Linux. We have received many requests from our customers asking us to enhance certain areas of Linux (logical volume management in particular) in order to make Linux a more acceptable platform for their IT operations. Furthermore, we have been asked to provide a migration path from existing platforms (both IBM and non-IBM) to Linux. IBM has been moving to satisfy these requests by contributing developers and technology to the Linux Community. This is what drove IBM's decision to release JFS to the Linux Community, and it is driving the decision to release logical volume management technology to the Linux Community.
IBM is releasing one of its most advanced architectures for a Logical Volume Management System. This architecture is quite interesting as it completely integrates all disk and volume management into a single, highly extensible, easy to use entity. We hope that the release of this technology will lead to a world class logical volume management system for Linux, one which satisfies the requirements of our customers as well as those of the Linux Community.
The first of several white papers describing the LVMS architecture can be found at the IBM Linux Technology Center website:
Since we would like to have an honest, open discussion about this, I would suggest that all interested parties post their comments to the LVM mailing list (unless someone has a better suggestion!). All comments are welcome!
Thanks!
Ben Rafanello
IBM Linux Technology Center
PS - Information about the LVM mailing list can be found at:
***************** End Message *************************
I wonder what other technologies customers are asking IBM to add to Linux. I'm not especially familiar with AIX - what interesting features does it have?
It'll also be interesting to see how this works with the other Linux LVM (no link - sorry). Will we end up with a similar situation to the current journalled filesystem developments where we go from 0 journalled filesystems to 4 (ext3, reiserfs, XFS and JFS)?
Idiot.
--
Any bets that Linux is _THE_ Enterprise-level OS in 5 years?
when I spoke to Heinz this january at LinuxWorld, and I mentiond that since IBM had just announced the JFS project, that hopefully in the not too distant future they would do the same with their lvm, he said he would like to see that.
as far as duplication of effort, I guess you're right, I mean let's only develop one journaling file system also. Which is it going to be? Choices are good.
"We are not tolerant people. We prefer drastically effective solutions"
From what I konw about the two, the linux lvm is more modular in that it talks to the kernel's virtual filesystem layer. That's better. The IBM LVM is more mature, more functional and from my experience, having used both, easier to use.
IBM is talking about taking their LVM and making it more modular. This is good progress. Take a company that had a good product. They like this product, but there is a Linux model that has some things they like. What do they do? They integrate those features into their product and release the source code.
I think what we are going to end up with here is the best of both worlds.
"We are not tolerant people. We prefer drastically effective solutions"
As long as they don't try and give linux the AIX printing system :-)
what's your problem with lpr and postscript? Linux already uses both anyway.
yeah, I think that was John Patrick... proving why he gets let out without handlers, as opposed to say Armitage... who's now the sr exec vp in charge of *nix or something like that. She's made a couple of statements internally that clearly showed she *didn't get it*, and pissed of a number of open source developers that happen to also be IBMers in the process.
I think one of the smartest things that IBM could do would be to release everything they make for Linux under the GPL, and I think they will.
Think about it. Why is IBM doing Linux? Mindshare, pure and simple. It could cost them in the short run to create things without licensing fees (of course, they do get free developers from the hobbyists who want to join in), but, in the long run, they get the mindshare of everyone whose lives they made just a little bit better that didn't have to pay them for it. Having thousands of OS supporters and developers thinking "IBM is really cool" is an intangible benefit, but it could be worth a whole lot in the future.
I'm not sure if it's a corporate tactic, plain common sense, or pure kindness, but I think I can live with it.
Be nice to your friends. If it weren't for them, you'd be a complete stranger.
Wonder if they'll
include smit/smitty or something like that?
I hope not, I really do. smit/smitty are just wrappers (smit is tc/tkl) that exec the real commands anyway, I'd rather have the command lines with better documentation.
Now the new WebSM system would certainly be welcome... especially if they provided the WHOLE thing, instead of just the part for LVMS... that would put linuxconf and all the others into the "also ran" category in no time flat.
Your objections don't even make sense, and it's clear you've never worked with enterprise environments:
LVs complicate system management; in particular, they make it more difficult to figure out what physical devices a file system actually depends on, and they make it much more likely that you make a mistake when setting up disks.
Not really, the LVM is always happy to tell you what maps to what, and this is not problem unless you create a logical root volume in which case you deserve what you get. Disks are too cheap these days for that sort of foolishness, and if you need LVM you can afford it.
LVs break the correspondence between block numbers and head positions. With simple physical mappings, small differences in block numbers usually correspond to small head movements, something file system designs tend to rely on, but with LVs, all bets are off.
What planet have you been living on, Mr. Jetson? The geometry the OS sees has had little or no relation to the physical geometry of the disk since disk controllers became more intelligent than the first PCs several years ago. Where things will actually land physically on modern disks has been totally out of your control for years, which is the reason Sun dropped the ability to twiddle cylinder groups way back in the move from SunOS to Solaris.
If you take advantage of LVs spanning multiple disks, you just multiplied your risk of data loss, because if any one of those disks goes, so does the whole file system.
That's why LVMs and RAID travel in packs - they form a symbiotic realtionship. With disks as cheap as they are, mnay people just use 0+1 (mirroring and striping), which eliminates this concern. If you're cheap and don't mind abysmal performance in degraded mode, you can go with 5.
If you need file systems bigger than a single disk, use RAID.
Have you ever done this? RAID tools that do this are inherently *doing* LVM, even if they're not calling it that and it's not a separate tool. RAID per se typically handles Mirroring, Striping and Parity, the volume manager handles concatenation. These are often merged in a single tool, but concatenation of disks is not strictly speaking a RAID function.
GNU Parted and PartitionMagic already provide you with the ability to resize partitions without a full backup and restore; you don't need LVMs for that.
While decent, these are hardly enterprise-class tools. I also sincerely doubt that either is capable of resizing volumes that span physical disks, since that *does* imply LVM functionality, unless there's an LVM hiding under the covers and lying to these tools. There are people that have a real need for single files that are tens or hundreds of gigabytes, so *not* spanning disks isn't really an option.
One of these days, Linux may get concatenated mounts, which would give you another, very reliable and simple way of having file systems span multiple disks. Adding concatenated mounts would probably not be any harder than hacking in an LVM.
I'll confess to being ignorant of concatenated mounts (although I asume this would allow simply allow concatenation at mount time, but LVMs have significant benefits, and the people who need them have been using them quite successfully for years. There's certainly noting that's going to prevent the need for them in the near term. LVMs make "big computing" possible on small computers. There's just no effective way around that.
LVMs may have thier shortcomings, and they aren't appropriate everywhere. Most people won't need them, and the ones that do can handle the minimal incremental complexity. Further, what are often called LVMs are usually today also RAID and JFS tools. This sort of capabilty is not easily duplicated in other ways or we'd have been doing it already for years.
It seems you really are living in Jetson-land as your ID indicates if you think things like "object based disk storage systems" are going to have large-scale real-world impacts anytime soon. Ah, the naivete of youth...
"The future's good and the present is nothing to sneeze at." - Roblimo's last
Smit was not bad as a text based admin tool (I found the X11 version very cumbersome). But the problem with it was that for many operations, it was, in fact, the only choice you had, because command line equivalents either didn't exist at all or were not well documented.
Another problem with AIX system management was that much of the configuration information went into a big binary database somewhere. That made it nearly impossible to manipulate that information from scripts. Furthermore, if AIX ran out of space while writing the system management database, the system would become unstable or unbootable.
What AIX has going for it is that it's a robust, stable system once it's running (in part, due to the hardware). But I don't think AIX system management is a good model for Linux to aim for. That's why they make all different kinds. If you like AIX or NT, they are widely available; I see no need to turn Linux into either of them.
LVM only allows resizing of Logical Volumes, not filesystems. Using a filesystem that will allow dynamic resing (JFS, or ReiserFS, or ext2 with a specific kernel patch) will allow current Linux boxes to resize filesytems on the fly. I currently use Linux LVM with ReiserFS and have on many occasions resized filesystems live. Even without the ability to dynamically resize volumes, LVMs allow you to take better advantage of your disk space, however. It allows you to stripe volumes as well as span multiple disks. Solaris and Linux native disk management doesn't really allow for this (although MD will allow you to do both).
it's Caldera not Corel... truely excellent post otherwise.
Officially worded as: "IBM provides how-to and defect support for the four major distributions of the Linux OS: Red Hat Linux, Caldera OpenLinux, TurboLinux, and SuSE Linux."
Most of the Linux community does not fully understand just how robust and featured the AIX platform is. Bear in mind that the features that Linux is JUST NOW beginning to think about have been in AIX for TEN YEARS. Thats like a Millenium in computer years. It will be interesting to see the stunned look on peoples faces when the fully realize just how far Linux has to go to catch up!
Informative my ass. OS/2 was never intended to run across all hardware platforms. IBM barely even wanted it running on x86 servers for fear of it cutting into their midrange business.
They certainly never considered putting their Single User, 286/386-specific, client OS on RS/6000s or AS/400s.
But anyway, like everyone else says, IBM "gets it" now. One thing they never got back in the OS/2 days is that it makes sense to promote PC-based servers. Maybe OS/2 missed the boat, but Linux fills the bill, especially since IBM now is willing to make tons of money of servicing and supporing other people's stuff.
When I hear the word 'innovation', I reach for my pistol.
Oh, gad, let's hope not. SMIT is great for those illiterate in Unix, and a nightmare for the literati. (Although it's "show me the command line of how you'd do this" option can be a good leaning tool for admins.)
I remember way back in the early AIX days discovering the hard way that S*IT didn't keep it's state in the config files, but in the ODB, so any changes you made by hand would be silently undone when S*IT realized the config file didn't match the ODB, forcing you to do nearly everything through S*IT if you wanted the changes to stick. AAAARGH!
"The future's good and the present is nothing to sneeze at." - Roblimo's last
I think that this suspicion is unfair, Aside from the fact that you're basing some of your points on actions of some 15 years ago, keep in mind that IBM is not the only one fighting with Sun over EJB (and Java issues in general) about licensing and other such legal B.S.
/. to see what kind of things IBM has been doing for Linux, Apache and other open source projects to see what their intentions are.
:)
Just do a search here on
Lets be honest here though, they are driven by market demand like any other company. It just so happens that right now, market demand is in line some of the needs of the open source community, so their giving back while making money.
Nothing wrong with that. Don't look a Big Blue gift horse in the mouth.
To answer your question, we have a method for associating drives. In a nutshell, we assign a unique ID to each drive in a machine. We then copy to each drive in the machine the ID of the drive that the system booted from (the BootID). When a drive is moved to a different system, the BootID on the travelling drive will not match the ID of the drive that the new machine actually booted from, thereby allowing the new machine to identify which drive is the travelling drive and resolve any conflicts in favor of its native drive. Of course, this scheme depends upon the how unique the IDs assigned to the drives really are ...
Does it resolve the conflict automatically? Or, more likely, does it ask the manager to pick up a new name?
Heck, I suppose it does both, it first resolve with a temporary name (like conflicting_name_2) and work with it while asking the manager to pick a new name.
Exactly!
Ben Rafanello
IBM Linux Technology Center
As I understand it, Block devices could not be used to simulate "Partition Managers". Currently, implementation of new partitioning schemes requires multiple components in the system to be modified.
2. On top of the block devices you can layer the RAID and LVM "features modules" which are then available transparently to upper layers.
As I understand it, there is no generic architecture or mechanism for doing this under Linux at the moment, and the existing mechanisms which can be used to get around this all have significant limitations or drawbacks.
Integration among those components may need some work but overall this does not seem that different.
When viewed from a sufficient distance, the two systems do not seem all that different. But as you get closer, you find that the devil is in the details ...
Providing a single administration point for the users is certainly something worthwile but that this is hardly something which could not be done with the current system.
True.
Finally, LVM (any flavour) do just provide a framework. A filesystem does not become resizable on its own.
Agreed.
Ben Rafanello
IBM Linux Technology Center
It's probably true, but the advantage of beginning from scratch is taht you are not encumbered by backward compatibility and therefore have more liberty in your design.
"The obvious mathematical breakthrough would be development of an easy way to factor large prime numbers." Bill Gates,
There is already a LVM for linux coming out with kernel version 2.4. There is also a version you can compile into linux as a patch I believe.
It even replicates the exact functionality of LVM for AIX and HPUX, from the opengroup.
I never said I was an expert on IBM's LVM, in fact, I haven't used it since 1994-ish. I have a fair amount of experience doinf these things, though, using half a dozen LVM systems, including Veritas Volume Manager and Sun's Solstice Disk Suite.
The whole point of LVM is to put more than one disk in a VG, so I don't get your comment at all. That *would* explain your problems with JFS, though, since any disk failure is guaranteed to take out your log volume!
If properly configured, all of these things work, and work well. That's why people use them!
"The future's good and the present is nothing to sneeze at." - Roblimo's last
It's always nice when I gigantic coroporation like IBM releases something under GPL. It's not only paving the path to the future, but every time it happens, it makes Microsoft-type companies look worse and worse. IMHO, anyways.
I usually refrain from ad hominem attacks, but in this case, it was warranted - the guy clearly was talking without knowing his subject. It's fair to call him on that, and to point out those factual errors for the rest of the community.
Despite your worries that I'm a latcomer "IT boss", my slashdot ID is about 130,000 lower than yours, and I was a participant here for several years before I bothered to register at the advent of moderation. What *I've* seen Slashdot becoming over that time is a community that is increasingly populated by clueless newbies shooting their mouths off and a signal to noise ratio approaching zero. And yes, sometimes that ticks me off.
My comments weren't intended to be arrogant (sorry if that's what came across), but simply to inject some factual basis into the discussion where little existed previously. The reason I bothered to do this is that I've seen such little things lead to big misconceptions among the community in the past. I would hope that we all recognize that experience is the best teacher and those that have experience doing something can ofer a lot to those of us that haven't.
"The future's good and the present is nothing to sneeze at." - Roblimo's last
They already are working with them. The link they give to visit the LVM mailing list is the identical one to what you gave, and if you look closely at the Linux LVM site you'll find it says its based on the old IBM and HP LVM.
Why don't they just offer to work with the current LVM project?
:wq
I've no exposure with other LVM type products mentioned here but I have used AIX's and it's a thing o' beauty. Even if it's redundant technology, I think Linux benefits from having it available. Our shop is Big Blue all the way (we're currently running Linux on S/390 as a test and have 5 RS/6000 frames as well as an S80 and a HACMP S70) and the only thing that ever causes us a problem is the damned automount process on AIX. Course the OS/390 environment never has a problem...
----
Slán leat agus go n'eirí an bóthar leat
If you use Online JFS 3.3 (requires HP-UX 11) you can supposedly shrink LVs also without much risk of data loss. (I'm not interested in finding out how true their claim is.) Yeah, SAM is horrible for extending volumes (it tries to unmount the volume to extend it). lvextend still reigns supreme. And you're right, it is licensed from Veritas. (Takes quite awhile to wait for an unlock code to arrive after registering. HP says they have to send the license info to Veritas in order to retrieve the code.)
Bureaucracy aside, I love Online JFS. Now if they could only get it to work on system partitions...
If I remember correctly Sun also sells the J2EE as a complete package and refuses to sell only the individual component(s) a customer might require. I'm not sure, but I believe that IBM was not really impressed with this fact either, as it would require vastly more expensive J2EE licences for customers that they would be developing solutions for. IMHO Scott McNealy (sp?) is just a Bill Gates waiting to be given full control of the sandbox. Which is a shame, because they're another end-to-end technology company that develops both excellent hardware and software.
----
Slán leat agus go n'eirí an bóthar leat
"What, has IBM misplaced AIX, OS/390, or OS/400?"
For the purposes of this discussion they may as well have. This is an apples-and-oranges comparison, as I see it. I won't dispute IBM's continued donimance in mid-size to large computing installations. Linux is partly about that, but I am coming at this from a monetarily bigger point of view: the mass PC market. Linux may very well have a chance in that arena someday, where the systems you named likely never will.
Ah, the halcyon days of microchannel! There were several things right about the PS/2, aside from that. The Model 60 tower was the first PC I ever saw that required no tools at all to access and work on. I attribute their failure to (very) high prices, the floppy-based BIOS, and the poor decision not to release MCA to the world's hardware developers. My point is that IBM has the engineering to take over the PC market - the questions have been in marketing, timing, and image. This seems to me a good opportunity to improve on at least one of those factors. I don't think IBM has given up, even after the PS/2 and OS/2 marketplace debacles.
I agree... DAMNED funny!
----
Slán leat agus go n'eirí an bóthar leat
Logical Volumes
Avert the predicament:
Root partition full
One of the things I heard about porting Linux to the S/390 platform was that the development team had to deal with the 16 interrupts available on the PC vs. 65536 (16^4) on S/390. Hmm... I'd love to be able to run OS/2 on our S/390's in single user mode, even tho it would be like driving a Ferrari on a go-kart track!
----
Slán leat agus go n'eirí an bóthar leat
Excellent comment... if you can't live with the abysmal performance of RAID-5 config'd volumes, go with RAID-10! As you said disks are cheap (OK, the SSA DASD that we use is definitely NOT cheap) and AIX LVM supports RAID-10 config'ed vols. I'm the dude actually doing the copying to the new RAID-10 configured spindles at our shop. Provides the Cadillac of data redundancy and speed (especially the SSA connexions). Good ole IBM, gotta love 'em. LOL re: your comment on an LVM root volume!
----
Slán leat agus go n'eirí an bóthar leat
because those illogical volumes are really a bitch to manage. Like herding cats....
-Denor
As long as they don't try and give linux the AIX printing system :-)
A journey of a thousand miles starts with a brutal anal raping at airport security
A good LVM for Linux is here.
There is one fundamental difference. IBM and Sun are competitors. IBM and Open Source are not competitors. The relationship of IBM with Open Source will be very interesting to watch. Remember that symbiosis can be defined as mutual parasitism. I think that somehow it is to IBM's long term advantage for Open Source, especially Linux, to prosper.
Look at what they're doing with Sun and Java. After working with Sun for 4 years, they refuse to ratify Sun's version of EJB, in effect leaving Sun out to dry. IBM was given access to all of Sun's java IP and this is how they repay them?
I'm just a bit leery (sp ?) of IBM cozying up with Linux and Apache. They may think everyone has forgotten how predatory they are but some of us haven't. I think most people see them as a MS foil and as a result consider them to be one of the good guys. Just remember, folks. 15 years ago, IBM was the oppressor of the tech industry. The only reason they aren't as powerful today is because they got too big and unresponsive, which led them to make several bad decisions.
-- You see, there would be these conclusions that you could jump to
Okay, on behalf of all the slashdotters who realize how much IBM is contributing to Open Source I would like to apologize to all IBMers who read the above post.
Besides Linux pure-plays like RedHat and VA Linux no company gets it more than IBM. Their contributions to Apache are impressive, they are releasing laptops with Linux, implementing a better JDK than Sun's for Linux plus they are behind the AlphaWorks site.But instead of thanking them or being grateful for their contributions certain people feel that they should bitch and moan about how IBM will turn on Open Source (how? By stealing GPLed code???). The tiff with Sun you describe is simply that IBM licenses some Java technologies but refused to pay Sun's exorbitant licensing fees for the J2EE brand. Wow, that is so evil, they implement a better version of Java than Sun then refuse to pay Sun for the permission to call it Enterprise Java.
Ingratitude is sad to witness.
I work at IBM/Austin. Rarely I see people in suits, they must be on a business trip or they are prospective employees.
:o)
The software engineers, designers managers I see dress very casualy. Shorts and sandals are common as well as t-shirts. I wear hawaii style polo shirts, ibm t-shirts with chip logos on them, a linux fund t-shirt, t-shirt I got on the Cindarella ballet by Ballet Austin etc. There are two pool tables in our building in the cafeteria area, and there is a salsa/chips/coke/beer 'social meeting' on Fridays. It would be ackward playing pool and drinking beer in a stiff white shirt/suit isn't it!?
The work time is flexible. I am expected to call in if I do not get in by 10a.m. for some reason.
(And I can work from home. While I am writing this I wait for a run to finish.)
But to be honest I have to add that these weeks there is a lot of work to be done (deadlines) and so I did not get to the recently installed pool tables yet.
We develop our software to run on linux as well. (Oh, and some of us thinks it is important to use >1 compilers for development and write portable (endiannes, 32/64 bit 'resistant') code. Beside xlC I normally use gcc with all warnings on and treated as errors, well, ok, since it is not enforced I forget this sometimes...
Matyas
IBM has also announced that they will be porting Linux to the AS/400 - see Midrange Computing for (a little) more detail.
Cheers,
Simon B.
If you moderate me down I shall become more powerful than you can possibly imagine.
(Yes I know readlin is GPL'd, but surely that doesn't stop them writing their own version ?)
I agree with everything, save the commercial thing. They have a few commercials that get me. The main one is the one where the "hackers" have broken into the company payroll server. As a security professional, I thought it was typical marketing fearmongering.
At any rate, IBM is doing some awesome things. They seem to be leading the way in HD technology. I've got a 6.4 gig UDMA drive that I need to upgrade to a bigger one and I'll definitely go with another IBM.
-core
...is still in use on the AS/400.
Cheers,
Simon B.
If you moderate me down I shall become more powerful than you can possibly imagine.
That lasted for all of a couple of years after I signed on with them. Several factors contributed to the decline of OS/2, including the internal infighting that you mention. And of course Microsoft was always happy to give the knife an extra twist at every chance.
I suspect a lot of IBMers are taking some pleasure in the hurts Linux is putting on Microsoft. I'm sure they're enjoying establishing Linux as a good solid leader in the features race, too.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
I hate to disagree (but that won't stop me ;>) but I think that SAM (HP-UX) is a much better admin tool than SMIT. Of course I'm biased as H(ockey)PUX was the first unix with which I worked. SMIT *IS* a nice admin tool and second on my list. Sun's admintool is at the very bottom ;)
"Fighting the underpants gnomes since 1998!" "Bruce Schneier knows the state of schroedinger's cat"
You can compile bash for AIX. Which was always the first thing I did on encountering a new AIX box. Or try zsh, which I believe also features tab completion. If you're really feeling masochistic and have a sufficiently late version of AIX, you could even compile everything you need to run Gnome. I saw a fellow running Enlightenment on an AIX box at IBM once. That was cool.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
Just click on the user info button. If the user # is 7608, you got the right one. ;)
That is the link. Tho it seems that by the time I typed my message a billion other people reported it. Let's see yahoo do that. Suckers.
The message on the other side of this sig is false.
There's already LVM for Linux here:
:-)
http://linux.msede.com/lvm/
Also, and LVM HowTo is available at http://www.linuxdoc.org/HOWTO/LVM-HOWTO.html
I'll be good to see another implementation though. More choices.
I run the LVM on my server with the 2.3.99-pre8 kernel and it seems to work rather well.
Prevent email address forgery. Publish SPF records for y
I'm glad I didn't follow Slashdot SOP and accuse IBM of hubris, misdeeds, or ignorance...
Maybe this will see the EXPERIMENTAL light of day in early 2.4 revs.
Jeremiah
"Flyin' in just a sweet place,
Never been known to fail..."
If IBM is more flexible and courageous (yes, it takes courage to bring something like this project to your boss, knowing that it's going to cost a lot of time and money and you have no real plan to directly recoup those funds), we have Microsoft to thank. IBM's "new attitude" is definitely that of a hungry company, which is the opposite of what they were when Bill & Co. basically cheated them out of the financial windfall that was DOS, and then later out of Windows.
It's fair to ask yourself whether IBM would be bothering with Linux if they had a successful OS that they had developed in-house. Moreover, I don't think it's unfair to wonder whether this isn't a first move at proprietizing an IBM Linux, which might not be a bad thing in itself but this is a huge corporation and I, for one, have a difficult time believing they got that way and stayed that way through altruism. This is a Predator as dangerous as Microsoft, maybe more so since they aren't watched like MS is.
Maybe I am being overly skeptical?
It works fairly well, but I've only used it up to around 15 gigs.
I've got 90 drives and just over 970 gig of data under the AIX's logical volume manager. It is rock solid and a dream when dealing with with that much storage.
My company is pretty evenly split between Sun and IBM (though we're phasing out Sun as we upgrade) right now. I've used the semi-Sun product Veritas to work with the 1.1 TB we have running under Sun. While I'll be the first to conceed that the Veritas product looks pretty and for a novice might be easier to use, I don't feel it is as sturdy a product as AIX's LVM.
The downside to LVM is that it's utilitarian. Whereas Veritas has a pretty GUI that turns drives icons different colors depending on their state, LVM is text-based even when running under the GUI smit interface. That could be one of the reasons I prefer LVM.
Over the years, I've moved more and more stuff off AIX and Solaris and it has made sense. One of the prime reasons I haven't moved more is that Linux didn't have Oracle, a JFS or a good tool for managing large amounts of data. Oracle fell. JFS fell. Now I see LVM on the horizon.
If Linux keeps progressing at this rate, I'm going to have to start putting my data where my mouth is. There is soon to be no more 'if only Linux had {hesitation}, I'd move {service} and cut our costs by {huge figure}'. My {hesitations} are just about gone.
InitZero
They are running one where I work. Pretty neat, no single point of failure machine. The department using it hasn't stressed it at all yet, so we really don't have any numbers to compare it to.
Vermifax
Vermifax
Logout
Ahhh, didn't know that having only worked at one customer site :)
Vermifax
Vermifax
Logout
I'd agree that AIX adds a separate, proprietory binary configuration database (the ODM) to the UNIX flat-file configuration model, and I'd agree that the way in which the ODM is bolted-on to the UNIX model leaves some ugliness at the boundaries.
However, to say "SMIT is the only choice" for configuration is somewhat misleading. One of SMIT's strengths is that *everything* is implemented via commands and scripts, under the covers. Indeed, it's fairly easy to extend SMIT itself to perform custom per-system administration tasks by adding menu / form definitions and the actions needed to implement to the ODM database (which is what SMIT reads to produce the menus, as well as being AIX's authoratitive source of config. data).
Result? Everything that can be done via SMIT can also be done direct from the command line. In AIX, most (but lamentably not all) commands that change flat-file configuration also "behind the scenes update the ODM also. For instance, the user configuration commands (mkuser, rmuser, lsuser etc) not only affect the filesystem directly (modifying /etc/passwd, /etc/group, /etc/security/passwd, home/ etc), but also update the ODM. Likewise for most of the the device configuration commands (certainly the common TCP/IP commands have this dual-mode operation).
--
I'd rather have a bottle in front of me than a frontal lobotomy
In ksh type "set -o vi"
Filename completion:
Press ESC \ after typing the unique portion of the file. (Note it will also go as far as the name is unique)
Scrolling:
Once you press ESC you are in vi command mode. Use j/k to scroll through command history and vi commands to edit the command line
Vermifax
Vermifax
Logout
but I don't know that that's their motivation. They definitely seem to show signs of having made a transition from their old behavior. Not that they aren't a bunch of corporate jackals out to make a quick buck, but all of their actions during the growth of linux seem to demonstrate an understanding of the nature of innovation and market change: the future of network computing is mutliple servers, each with whatever OS/HW config fits the task. Who's company is better prepared for it than IBM? All this linux stuff they do is energy poured into further preparation for that market's radical growth over the next few years.
Now they just need to reorganize some people out of the old development centers and into new ones. Encourage them to think in new ways in a new environment. Maybe up here in Boston, so I can work there. That way you can pay me for all the installing-linux-on-Thinkpads advice I give you people.
-jpowers
-jpowers
nt
Vermifax
Vermifax
Logout
First thing's first: Real SMIT users use 'smitty' to tell smit to run in terminal mode. The X version is indeed cumbersome (though I admit I like the wee running man - he falls over if a command fails :).
SMIT actually does (almost) everything via commands that you can execute yourself, which limits it in some ways. However this is also its advantage: Rather than pressing ENTER to tell smit to execute that command you've just given it, press F6 instead. SMIT will now tell you the actual command line it was about to execute. You can take this command and type it yourself, or, more importantly you can stick it in a script. Linuxconf isn't able to do that in the same way as far as I'm aware, or at best it does it in terms of its UI, which isn't a good way to script things. Have you ever tried to script setting disk space quotas under Linux? Aarrgh!
The ODM (binary database) is indeed something of an issue, and something I'm always fretting about whenever I have to do system recovery on an AIX box. Though it's been scary in the past to note Linux people talking about an NT-style 'registry' (shudder). Probably the main thing we can do there is that if such a thing ever does eventuate, ensure that it is well documented at a central site somewhere, and provide good tools to manipulate it. In particular such tools must be small and stand-alone so that they can be run easily in recovery mode.
Some things about AIX I'm not keen on, for instance its price, its devel environment, and some of its 'non-standard' wrinkles. But I've been working with it for 6 years now and I can say in no uncertain terms that its disk handling (LVM and JFS), and its system adminstration (SMIT) are nothing short of stellar, especially if you have to maintain one of these things remotely.
I don't think I'd like to see SMIT verbatim in Linux, but LinuxConf could learn a lot from it.
As far as IBM and the open source community goes I'm pretty comfortable. IBM got whacked good in the 80s, and have pretty much played ball since. More important though is the fact that they've been slowly reshaping themselves into a total-service oriented organisation for some time now, where they can make their money from the services they provide more than the products. Linux slots into this strategy very well, don't you think?
If you go and look at their JFS subsite you'll see that they've been making very steady progress on it since they started - I've been monitoring it regularly and I think what they've acheived in the short time so far is actually pretty remarkable!
These are often merged in a single tool, but concatenation of disks is not strictly speaking a RAID function.
This is wrong, in the begining, RAID was meant to concatenate disk sizes. Rendundant Array of Inexpensive Disks means that many small disks are cheaper than one big disk.
Then it came mirroring and RAID 5. there are references to this in: seagate and DPT web sites and in the SCSI linux howto.Why is it that people feel they just have to use the N-word (Nazi) when they dislike some organization?
Because they seem capable of making the distinction between what you're talking about and their general organizational behavior over the course of decades before that war, which was evil in a different way. Yes, there was a Nazi Party back in Nietzche's day, and while they didn't slaughter Jews (and my kin the Gypsies), they still displyed the same weird organizational behaviors.
We discuss it because the motivation to enforce conformity seems to be something we Americans as a society can't shake, especially in the corporate sector, and any part of our society that easily compared to theirs needs to be watched very carefully.
As it's often said: "Never forget." It's easier to keep it in mind if you talk it out.
-jpowers
-jpowers
At this point in time, hardware RAIDing and mirroring are pure wins in terms of performance if you just want to manage one bus, but if you're creating a highly available server, one that must stay up, no matter the situation, then you're going to need an LVM approach to the problem.
[And remember, kiddies,
- mirror
then- stripe or RAID
. Reducing redundancy in the quest for high-availability is a fool's game.]-- Riding the Winds of Fires Lit in Ancient Days
If I had a moderator point, you'd get it.
jpowers
AC cause I'm running out of karma!
-jpowers
-jpowers
Being a soon-to-be graduate in the UK (Portsmouth, near IBM's headquarters in the UK) I am seriously interested in working for them. My question is, have you any hint specific to IBM to give people wanting to work for them?
"The obvious mathematical breakthrough would be development of an easy way to factor large prime numbers." Bill Gates,
As the LVM for Linux page points out, the LVM concept was initially developed by IBM and subsequently adopted by other systems, so this is an area that IBM understands probably better than anyone else (except maybe for third parties such as Veritas). Indeed, LVM was one of the main selling points for IBM's AIX when it first appeared in the early 90's, and helped IBM greatly to make inroads against incumbents Sun and HP.
Even though this technology obviously won't make it into the 2.4 release, it will dramatically strengthen Linux's enterprise capabilities when 2.6 ships, particularly when coupled with the journaling file system in that release. Flexible volume management is taken for granted by most commercial users today, and capable LVM functions in Linux will put its storage capabilities on par with almost every other operating system available.
Note, though, that regardless of the design of the LVM itself, there are some tricky issues that need to be resolved in the implementation with an actual file system. For example, it turns out that growing volumes is fairly straightforward, but shrinking them is much more difficult. The paper mentions that specific support is needed from the underlying file system to enable shrinking, so even with this "gift" from IBM, there is still a lot of work to be done to come up with a useable solution.
'physical', not 'illogical'. Well, ah, I guess 'physical' is pretty illogical at that now that you mention it, nevermind :))
try { do() || do_not(); } catch (JediException err) { yoda(err); }
IBM would do well to work with the existing project, or at least to establish a good-faith basis for communucation with the developer community already committed to solutions under public license.
The same could be said of Linus and FreeBSD. I think we're all better off with competition, because Open Source competition benefits both projects.
--
Assuming that they are porting the Aix LVM, it is hardly duplication of effort to provide a version of an enterprise scale LVM on other OS's
Vermifax
Vermifax
Logout
Heh. Dude, I spend too much time hitting reload to bother creating new accounts. :P~~~
Officially, IBM is 'business casual'. OTOH. I worked at IBM Austin in the recent past. The development organization I was in pretty much didn't care about dress, my manager was a regular shorts wearer and tee shirts were common. The hours were extremely flexible as well. One guy I worked with got to work at 5:00AM and worked till 3 or so another got there at 11:00 AM and worked late.
There also were some good things about IBM you won't find at a smaller company. For instance IBM is VERY concerned with ergonomics and you can get all kinds of good stuff (chairs, keyboards, etc..) from the ergonomics guys.
I would have to agree with you. IBM is hardly the evil "Big Brother" figure that Apple threw a sledgehammer at in 1984. They have learnt a lot from their mistakes, and have put out solid product both on the hardware and software side.
No, this ain't your grandpappy's IBM. This isn't even your daddy's IBM. This doesn't even seem to be the IBM that killed OS/2 and made the Aptiva series of crapola desktops. If they continue the route they're on, they could be the company that pushes Linux over the top. They also make damn funny commercials, too.
One question: What happened to the Long Trail motherboard and IBM backing of LinuxPPC? Have they gotten any mobo manufacturer interest in producing that thing? I want one, and have since before Apple killed the Mac Clones! Anyone know? Anyone? Bueller?
Knowledge is power. Knowledge shared is power multiplied.
The real question is 'will they actually develop it and make it work?'. The reason it's a question is that IBM has been releasing code to the community for a little while now and targetting it at Linux, but they have not put sufficient developer time into making it work.
Their JFS (journalling file system) is a good example of this. released quite some time ago, no effort to port it to current versions on the kernel and no apparent discussion about that is going on.
The LVM could be similar. *IF* they had released the LVM last year and *IF* they had supported it with a reasonable amount of programming resources we might care (and I say this as a person who has used and *loved* IBM's LVM on AIX). On the other hand, if they just release the code but don't support the transition and don't make an active effort to participate in community development of it, who cares?
Contrast this to SGI who, for all of their failings as a company, are releasing code *and* supporting its inclusion into the kernel through active participation in the linux-kernel list, web sites and other mailing lists. I'd like to see IBM get on board in the same way.
We realize we're the 800 pound gorilla here, so our manners have to be impeccable. IBM exec (John Patrick ?) at a meeting with Linux developers about a year ago. And they have been. Amazing, just amazing.
1000 SlashDot sigs
On their UK website they say that you have to dress up only if/when you meet with clients, which is normal IMO.
"The obvious mathematical breakthrough would be development of an easy way to factor large prime numbers." Bill Gates,
So just how does 'redundant' figure in that??
Redundancy means that more data is written than necessary in order that the original data can be recovered in the case of an error.
John
John_Chalisque
Ah, the naivete of youth ...
Excellent, this is going to be one of my favourites! /. is becoming a place of mega-enterprise-I-have-all-the-experience-you-have -obviously-no-idea IT bosses. Arrogance par excellence. If you have a different opinion, say so (and I would definitely agree with you here). But please refrain from this kind of classy, elitist and sickening arrogance.
Why can't people just discuss a topic without taking everything personally?
First the easy ones:
I wonder if this is some pie in the sky vision or if they have any code to back this up?"
Yes, we have code. IBM has a prototype of this LVMS on at least one of its platforms that I know of.
As long as it's released under a liscense like the GPL, we don't need to trust them
We are currently releasing the technology to the Linux Community. The current white paper provides a high level view of the basic architecture. As the architecture is discussed with the Linux Community, additional white papers will be produced to explain, in greater detail, various aspects of the architecture. Should things progress to the point where we are actually releasing code for some or all of it, then any code we release will be under the GPL.
Heinz Maulshagen and team have already written an LVM for Linux. This LVM is already in the 2.3.x series of kernels. What is IBM's reasoning for this duplication of effort? Might it not be more effective to assist in the current LVM implementation instead of bringing additional complexity? Or are there advantages in IBM's approach to logical volume management?
Heinz and his team have done an excellent job with their LVM. However, we feel that our design has some advantages, and that we are looking at the problem from a different perspective. Let me explain.
IBM's initial concept of a logical volume management system was not all that different from what we find in the Linux Community today. However, over the years, IBM's concept has evolved based upon input from our customers, as well as usability studies performed on our customer base. Our current concept of a logical volume management system integrates disk partitioning and management, bad block relocation, software RAID, encryption, filesystems, logical volume creation, deletion, and management, etc. into a single, coherent, open ended architecture. All of the components of this architecture communicate with the Logical Volume Manager, who co-ordinates their activities and handles all interactions with the user. The Logical Volume Manager is a single program with three interfaces: a command line interface, a text mode interface, and a graphical user interface. Thus, there is a single place for the user to turn to for any aspect of logical volume management. This makes logical volume management easier for the user to learn and use, and it reduces the chances for user error and data loss.
As an example of IBM's concept, lets examine the case where an encrypted volume is to be shrunk. We will assume that the user has already been through any authentication process which may be required to access the encrypted volume and has been granted full access. The user would start the LVM. The LVM would allow the user to select a volume. Once selected, the user would indicate that the volume should be shrunk. The LVM would examine the volume, find which features are in use on the volume (encryption, in this case), find what filesystem is in use on the volume, and then ask each feature that would be affected by the shrink if that feature can handle having the volume shrunk without data loss. If a feature indicates that it can NOT handle having the volume shrunk, then the LVM informs the user that the volume can NOT be shrunk without data loss. If all of the features indicate that they can handle having the volume shrunk, then the LVM would ask the filesystem on the volume how much the filesystem can be shrunk without data loss. If the filesystem indicates that it can't be shrunk without losing data, the LVM will inform the user appropriately. If the filesystem can be shrunk, it will indicate how much it can be shrunk before data loss begins. The LVM will then present this value to the user and allow the user to specify how much to shrink the volume by, limiting the user's choice to only those values that prevent data loss. Once the user has specified how much the volume is to be shrunk by, the LVM will then notify the filesystem to shrink itself. Once the filesystem is done shrinking itself, the LVM will notify the features on the volume that the volume is about to be shrunk. After this, the LVM will actually shrink the volume. Once the volume has been shrunk, the LVM will notify the features on the volume that the shrink has been completed. The filesystem will also be notified that the shrink has been completed. Once all of this has been completed, then the user will be notified that the volume has been successfully shrunk.
The above example is quick and dirty, and it leaves out quite a bit, but I hope it gives you an idea of what we are trying to accomplish. Basically, we are trying to bring together all of the various aspects of logical volume management into a single, cohesive, seamless entity. Furthermore, we are trying to do this in a way which is as flexible and expandable as possible. Consider the following:
The LVMS architecture uses logical disks. Logical disks are created and controlled by plug-in modules called "Device Managers". Thus, anything that can be made to appear as a logical disk through the use of a plug-in "Device Manager" module can be used by the LVMS. As an example, lets say we have a "Device Manager" plug-in which can access a Storage Area Network (SAN). By just adding this plug-in to an existing system, the LVMS would be able to allocate storage on a SAN, make it appear as a logical disk, and then use any of the LVM capabilities that could be used on a local disk. No code in the LVMS, existing plug-in modules, or LVM utilities needs to be modified or changed.
The LVMS Architecture uses logical partitions. Logical partitions are controlled by plug-in modules called "Partition Managers". If you wanted to access a drive that was partitioned for use with a Mac, you would just need a "Partition Manager" plug-in module that understands the partitioning scheme used by the Mac. No LVMS, existing plug-in module, or LVM utility needs to have its code changed. All that needs to be done is have the Mac "Partition Manager" plug-in added to the system.
The LVMS Architecture employs something we call "Feature Plug-ins". Feature Plug-ins control how logical partitions are combined into logical volumes, such as through drive linking, the various forms of mirroring, or software RAID. They can also be used to filter I/O to a volume, as would be the case for encryption. They can also be used to redirect I/O, as in the case of bad block relocation. Furthermore, Feature Plug-ins can be stacked to produce a volume, so that multiple Feature Plug-ins can be used on a single volume (encryption combined with software RAID and bad block relocation, for example). Every volume in the system can employ different Feature Plug-ins, and volumes which do use the same Feature Plug-ins can have them stacked differently. Thus, the user has an enourmous amount of flexibility. One final point to make here is that, since all of the plug-in modules communicate with the LVM, the LVM can coordinate their activities to ensure that the user doesn't do something which will result in the loss of data (at least without adequate warning).
Given all of the above, I hope everyone understands why we think the LVMS Architecture we are releasing has some advantages over other approaches.
Well, this post has gotten excessively long, and I apologize for that. I will try to address additional questions/comments/concerns later, hopefully with smaller posts!
Ben Rafanello
IBM Linux Technology Center
The LVMS sounds really good from the white paper, but it's a shame they place such stringent requirements on the underlying devices. In particular, they seem to require read/write access and random access, which would leave out CD-ROMs and tapes. Of course, many of the features the LVMS provides wouldn't make sense for such devices (I'd like to see someone put together a RAID from a bunch of tape drives!), but some do. The plug-in architecture could be really useful for these things, too.
So why not provide a capability-based facility, which makes use of the capabilities of a device, without giving up on devices that don't have those capabilities? It would be good to bring *all* mountable devices under one LVMS with one consistent interface, even if it meant that not all devices could use all its features.
(There are other plusses and minuses between the two, of course. IBM's lvm can do mirroring and raid at the lvol level instead, whereas the existing lvm requires you to raid the underlying physical volumes instead. But, you can put data in the existing LVM's lvols without worrying about destroying anything by writing over LVM datastructures as you would with IBM's, since IBM stores system data in the beginning of each lvol--In the current system you can safely make an lvol writable by a user, for instance.)
ksh does let you do that stuff, but it's not active by default. Read the man page and you'll find that you can give it Emacs-style bindings. Amusingly, it also has a vi-style mode :-)
I have been perfecting my regex for IBM products. Before your post, it was this:
As you can see, this regex matches the following IBM products:
However, if what you say about (what I thought was written as) DB/2 is correct, then my regex has to be modified again:
I don't make the rules. I just make fun of them.
Id say ibm is aproaching the situation the right way, i think ibm really seems to underestand the open source community, coool.
6 /0049.html
http://linux.msede.com/lvm/mlist/archive/2000/0
extract of ibm announcment
"Since we would like to have an honest, open discussion about this, I would
suggest that all interested parties post their comments to the LVM mailing
list (unless someone has a better suggestion!). All comments are welcome!"
We like to think of it as "stealing them back."
-jpowers
-jpowers
SMIT (System Management Integration Tool) is the best system adminstration tool of any I've seen, on any OS, UNIX or otherwise.
That "big binary database" to which you referred is called the ODM. It's in /etc/objrepos. You can make modifications to the database using the various ODM commands, such as odmget, odmdelete, and odmadd. There is no reason you could not put these commands into scripts. In fact, many of the AIX system commands do just that. Do not mess with these commands (especially odmdelete and odmadd) unless you know what you're doing.
Sun tried to jack up the prices on IBM from what I heard/remember about the whole fiasco.
As an IBMer and much longer Linux lover I wouldn't be working for the company if I didn't see anything but good intentions. I believe that IBM honestly believes in Linux. Not just that it will do well, but that its the best solution up and coming. I won't tell you to trust them/us. Do what you like. But I say I trust them. Their support for Linux is great for the community if you ask me.
Do they make you wear the stiff white shirt of a specific color of computer lore, or are they like lots and lots of tech companies, that recognize being productive means being comfortable (ie, wearing what you like and working how you like?)
Not that this is not intended as a troll...I'm generally interested. I'll be graduating with a BS in CS in 2 years, and I'm trying to get a head start on job prospecting. As such, I don't want to work for a place that imposes arbitrary rules on work habits and the like.
Three years ago, it certainly did. AIX's LVM
has supported mirroring since V3. What it
didn't support was mirroring and striping
together. You could mirror, and you could stripe,
but you couldn't do *both*. Since 4.3.3 came
in last year, you *can* do both now.
Chris Mattern
I actually enjoy SMIT as a system management tool. It is an ascii menuing system that controls just about all of the system. (Of course you can still type all of the commands on the command line as well) It is fairly user friendly and in my opinion very fast (compaired to gui sys management tools)
Vermifax
Vermifax
Logout
Actually, you can come and go as you please, and there's no dress code. I usually wear jeans and a shirt, but you can wear shorts and a t-shirt if you want. It's actually, by far, the coolest place I've ever worked. And I'm still /in/ school. (Internship, donchaknow.)
-Will the Chill
Creator of RPerl, Scouter, Juggler, Mormon, Perl Monger, Serial Entrepreneur, Aspiring Astrophysicist, Community Organiz
C'est la vie. Learn to live with it. I know because I worked in Tech Support for 3 years.
-o Disclaimer: My employer doesn't even agree with me about C indentation style. o-
Many IBMers expressed relief (on the IBM internal forums) that Linux was beyond the control of IBM, so the company couldn't screw it up the way they did with OS/2. Many of them are still bitter about that. Many of the old OS/2 guard have moved to Linux, happy to have an OS that lets them work the way they want to work.
I think it's cool that they're adding neat features to the OS. Having them contributing cool features why not trying to manage or market the OS seems to be the right way of doing things. Perhaps they should look at modifying their other business units to work in this sort of paradigm, with IBMers only contributing to projects managed by other business units...
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
anyone else remember when IBM and their starched white shirts represented nothing short of the Nazi party? Its refreshing to see someone so well known for their closed doors and Let them eat cake mentality step into a movement like linux. When IBM first released JFS, I kept looking for the 'gotcha' that never came. It seemed like IBM really did give up something of theirs to the Linux community. Now with their LV stuff being announced, it really does give a linux geek hope.
...and the geek shall inherit the earth...
www.linux-skunkworks.com
Isn't it the case that the new 2.4-test1 kernel has the beginnings of LVM?
That statement makes NO sense what-so-ever. I can't believe that it has scored like it has! First of all, Linus is a person, not a company. His personal feelings are his own. Nothing stops Linux developers from helping BSD developers (inverse true too). Second of all, Linux doesn't run on FreeBSD (as LVM does/will with Linux). Third of all, in what shape or form will IBM competing with open source developers help Linux (yes I understand the basic concept of competition). The point being, that I believe the original post was trying to make, we don't want another Sun/Blackdown problem. It is a show of good faith for IBM to enbrace what is currently out there. In the long run, both projects can gain if IBM handles this properly and does NOT try to compete. If IBM's LVM is better than what is already in place, they might gain extra help from the Linux LVM group. Either way they can complement each other. LVM for home users (Linux LVM) and LVM for enterprise users (IBM). Having a common look/feel of the tools, would be very nice!
I'm using it on a Warp Server for e-business ("Aurora") system and it has been a BIG help. Frankly, the OS/2 version is better suited for a workstation than a server, but it is awesome to be able to extend partitions on the fly. JFS lends a huge performance improvement over HPFS as well. HPFS386 is a bit faster, but that's because it's tied directly to the network layer and has the risks associated with that. I can also recover a 9 GB JFS volume in less than 15 seconds as opposed to a few minutes for HPFS. That's recover as in, all files back where they're supposed to be, not FILE0001.CHK. JFS is open-source now, and the same version as I'm using. It's robust and mature, not some 0.0.2 super-pre-alpha version.
Gamingmuseum.com: Give your 3D accelerator a rest.
Now if only Apple would incorporate this into Mac OS X and Darwin. Apple, are you listening?
http://linux.msede.com/lvm/
Check Freshmeat for updates.
IBM would do well to work with the existing project, or at least to establish a good-faith basis for communucation with the developer community already committed to solutions under public license.
Jeremiah
"Flyin' in just a sweet place,
Never been known to fail..."
Hehe. I actually work with IBM's logical volume manager, and I found it funny that this "dublin" character thinks this is what enterprises use. LVM is ok as long as you only have 1 hdd per vg. Removing that drive breaks quorum, and it makes life a living hell to simply put your filesystem back up. A failed drive means no quorum as well. ouch. Under AIX we use raid in hardware, or we use a SSA array. while we could mount the whole ssa array as one 128gb vg, we'd rather not do this, as a failure in the middle makes JFS VERY angry. This is especially true if the missing volume is the pdisk you send your log to. ouch. Stay away from IBM's JFS!
Lowmag.net
Not that this is not intended as a troll...I'm generally interested. I'll be graduating with a BS in CS in 2 years, and I'm trying to get a head start on job prospecting. As such, I don't want to work for a place that imposes arbitrary rules on work habits and the like.
Not at all !!! I work for IBM in the Santa Teresa and Almaden Labs (divvy my time) in San Jose. Today I'm dressed in a UserFriendly.Org "Evil Genius in Training" t-shirt, shorts and sandals. As long as you're not naked, nobody gives a damn. Nobody gives a damn what time I come in or go either. My manager doesn't even give a damn if I want to take more vacation than I'm "offically entitled to".
The values that are strong in my organization are simple - technical achievement, period. Code talks ! Heck, every time I visit my second-line manager (boss' boss) he is hacking away on his debugger :-)
I strongly recommend IBM for anybody looking for interesting, fun systems work. Feel free to send me your resume especially if you'd like to work on the internals of a database engine.
EBCDIC is still in use on VM/ESA for the S/390. Although VM has been around for a long while, it's still continuing on in development and has new releases periodically. So, IMHO, it seems to fit the criteria for a "modern system."
Ironically, VM is one of the systems that Linux is being ported to, causing a sort of revitalization of VM, continuing its modern-ness.
I just spent most of a day reorganising the partitions on several machines (backup, repartition, restore, repeat etc.). A LVM would have been an enormous help.
As for it's being ready or not; IBM don't usually make this kind of announcent unless they've pretty much gotten the technology down so I'd expect the project to be well on it's way.
[ Blairism is the continuation of Thatcherism by other means. ]
Heinz Maulshagen and team have already written an LVM for Linux. This LVM is already in the 2.3.x series of kernels. What is IBM's reasoning for this duplication of effort? Might it not be more effective to assist in the current LVM implementation instead of bringing additional complexity? Or are there advantages in IBM's approach to logical volume management?
http://linux.msede.com/lvm/
Oh yeah and Linux already has a form of LVM. I forget the link.
Comes with 2.3.x 2.4 or its it in 2.2 back ported? Not sure.
The gist of my filesystem manager is: What if the fs was the registry? No risk of screwing your machine by removing the registry. I hate how redundant some concepts in the fs are. So I've come up w/ a framework.
Coming soon.
The message on the other side of this sig is false.
Hmmm... All I can say is try your best to talk w/ any sort of IBM rep that may be at some school function (job fair, etc.). And make sure you know your stuff. No secrets that I know of. And I'm forbidden (non-disclosure agreements) to tell you the secret IBM handshake, which gets you into any IBM building around the world. Oh, wait. I was thinking about my ID card. Nevermind...
-Will the Chill
Creator of RPerl, Scouter, Juggler, Mormon, Perl Monger, Serial Entrepreneur, Aspiring Astrophysicist, Community Organiz
RAID-10?
Is that taking multiple RAID1+0's and concatenating them? (2 drives go into a single logical drive, and then multiple logical drives are concatenated to create one large logical logical drive?
Sounds groovy. I went from a hardware (DPT) RAID-1+0 to RAID-5 because I needed more space. I don't anticipate a drive dying but I should be able to get another in a couple days if one does happen to yak.
It works fairly well, but I've only used it up to around 15 gigs.
I wish some of this cooperation would result in the release of specs for the older RS/6000 machines. I guess releasing an lvm system is more sexy than documentation on 5 year old machines.
~# mount -t ext2 /dev/sda2 /home
Error: fuzzy dice in foo buffer found.
~# mount -t ext2 /dev/sda2 /home -o fuzzy_dice
Error: fuzzy dice requires car.o module
~# insmod car.o
Unresolved symbols: car_need_oil, car_need_gas
~# dd if=/dev/cash of=/dev/bank bs=1 count=300
0+300 records in
0+300 records out
~# dd if=/dev/bank of=/dev/wallet bs=1 count=50
0+50 records in
0+50 records out
~# insmod car.o io=/dev/wallet
~# mount -t ext2 /dev/sda2 /home -o fuzzy_dice
Succeeded.
~# cd /home
~# ls
KERNEL PANIC!
000:000 FF 00 CC G0 BB LE DE G0 0K
Goddamn illogical drives...
dublin had a nice response to this, but let me add a couple of things:
With the LVM I'm familiar with (AIX), you can tell which physical partitions are being used by a file system. I don't recall ever needing to know the specific disk block number for a file. I've been working with Unix since 1992 & have been a SysAdmin for most of that time.
If you don't want your file system spread all over the disk, in AIX you can set up the logical volume as contiguous space. You can tell it *exactly* which physical partitions to use.
Spanning multiple volumes doesn't necessarily "multiply the risk of data loss" -- if you do it correctly, you'll reduce your risk, since you have multiple copies of the data on different disks.
Apparently, much of the rest of the Unix world didn't think AIX's LVM was "poorly designed" or "half-baked". The Veritas volume manager was based on the same concepts; have you looked at the sales growth Veritas has been having lately?
In our experience (I work for a large telecommunications company) LVMs *very clearly* decrease the complexity of systems management. If the Unix flavor for a system doesn't have a LVM built-in, we *always* install one. Certainly you can cause problems with LVMs with inadequate training, but one can screw up filesystems with base-level commands too, if they're not careful and/or don't know what they're doing.
I would agree that many people are too hard on IBM. I have to admit, that at first I was skeptical as hell about their involvement in the free software/open source world. But I have to say that so far they are doing a much better job of living up to their promises than I ever thought they would have.
I think that if IBM can clean up their act like they have, that maybe there is possibly some hope that Microsoft may someday decide that they don't have to play dirty tricks all the time too. I'm not all that optimistic that will happen any time soon, but I won't say it could never happen.
Unfortunately, for IBM it took a large scale turnover at the top levels of their management to get them to change their status quo. If anything, the replacement of Akers by Gerstner was the key turning point for IBM. Microsoft upper management is still firmly entrenched with Gates and his inner circle. As long as they stay where they are, it is unlikely they will change.
Salesmen dress up Global Services at accounts dress up and those are the only groups I belong/talk to so I can't speak for the others.
Vermifax
Vermifax
Logout
Salesmen dress up
Global Services at accounts dress up
and those are the only groups I belong/talk to so I can't speak for the others
Vermifax
Vermifax
Logout
I thought that one of the big things about XFS was the LVM that sort of went with it, and I was hoping that SGI would port it to linux along with XFS. Is that likely to happen?
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
If they do this, I hope they give Linux the tools to make the LVM usable. Particularly the mksysb command(can be configured to back up your root volume) and especially smit.
Smit is about the best sysadmin tool I've ever used. Its much better than linuxconf IMO. I know the purists out there think "a real sysadmin always uses the command line", however what smit gives you is the command line, in a convenient menu form. Rather than wading through the man page, you basically have a menu that comprises each command line option with a description of what it means. You can also hit F6 at any time to see the command you are about to execute. Which also makes it a great tool to learn Unix.
No, Thursday's out. How about never - is never good for you?
as this site is hosted unders ource/linux/
http://oss.software.ibm.com/developerworks/open
does that mean that we get the source atleast ??
GPL ?? BSD ??
what license could it be ?
IBM promised to bring more and more stuff to Linux..And they are doing it
In this article at ZDNET, they talk about how IBM is releasing their Small Business Pack for Linux. They are trying to port more and more over to Linux. A lot of people don't seem to give them credit. I support this outright...and hope they release more stuff under a more flexible license like the GPL. I think that every corporation can see the light eventually. Here's to hoping.
I work with LVM in an enterprise environment we have 230 vg's for one department with between 4 and 8 hdisks per vg. Lvm runs just fine. Who cares if you can't make quorum on a vg with no usable disks anyway. Replace the disk and restore from a backup. Of course there is always the option of turning quorum off. As for a 128gb vg I can't believe you wouldn't mirror your JFS log on at least to lv's on separate disks.
Vermifax
Vermifax
Logout
Linux Logical Volume Management HOWTO effort
t
Richard Allen
bert hubert
Linux contains a wildly powerful Logical Volume Manager which allows you to combine and mangle your disks to create very
flexible volumes for your filesystems.
News
2000-04-16
Released version 0.0.2 - these version numbers a rather bogus, but every once in a while we feel that we have
progressed significantly, and bump the version number a bit. The version on this page is up to date within 15
minutes of our latest changes.
2000-04-06
Released lvm-viewer
2000-04-04
Started writing a visual tool for LVM maintenance. Sneak preview screenshot available
2000-04-02
Richard Allen joins as author. He has extensive knowledge of HP/UX LVM, which closely
resembles the Linux implementation.
2000-04-02
Lots of new content. Added 'one big page' html.
2000-04-01
Initial version online
The HOWTO
This document is a very rough first draft and should not be assumed to be widely useful yet.
Files were last updated at 2000-04-28 03:32 CET (ie, about 1171.1 hours ago).
CVS Changelog, or view changes
SGML
ASCII
HTML
One big page of HTML
dvi
ps, ps.gz
pdf, pdf.gz
HTML tarfile
Bazaar
This HOWTO is intended to be very much a Bazaar style development. If it were to be any more open, bits would fall out. If
you think our wording is bad, or you are able to resolve a FIXME, don't hesitate to mail us.
A CVS tree is available. try this:
$ export CVSROOT=:pserver:anon@outpost.ds9a.nl:/var/cvsroo
$ cvs login
CVS password: [enter 'cvs' (without 's)]
$ cvs co lvm-howto
cvs server: Updating lvm-howto
U lvm-howo/lvm-howto.sgml
The idea is that this HOWTO will be a cooperative effort, much like the Linux kernel itself. For the moment, we will be playing
'Linus', and we soon hope to be joined by Alans, Daves, Ingos etcetera.
IBM's implementation of this sort of protocal really doesn't help the volume management structures used in most networked operations. Contrary to Big Blue's supposed Linux adoption, this is a proprietary schema that will have significant detrimental effect on the way volumes are typically organized.
Volume organization is currently a by-product of several decades worth of research into the best known algorithms used for organizing volumes. This way, it is volumetrically organized by volume, in an orginizational way. Which is the results of, mainly, shelves. Every library I've ever been in has their large collection of volumes on shelves.
A more recent development in the field of volume management is a system whereby other types of volume are managed. This has been proven to work because it does. And therefore it is correct. And working.
Really, it's all about knobs. Old school knobs are where it's at in volume management, especially when the neighbor's bitching at 3:00 AM. These new fangled banks of toggle switches just don't work. I believe this is what the white papers are all about, although I haven't read them. Nor do I intend to. I mean, who's got the time?
White papers, of course, are best manageed with volume. Lots of loud paper shuffling >80Db. And then kept in volumes. On shelves. With a knob.
But why white? Always these days it's the anglo papers that we are reading about. What of the afro-papers? And latino-papers? Are not these papers volumes? Do we not organize them volumetrically? I like marbelized paper. It makes resumes look really corporate.
Thank you.
Shrinking a filesystem is supposed to be possible with the OnLineJFS (extra $) option.
I've tried it twice, and it hasn't worked yet.
Always make a backup before attempting this sort of thing.
Folks, the longstanding strength of all (ok, most) unix distros since day 1 have been the ability to specifically tune your individual system to the task at hand. Unlike so many other OS'es that seem to want to be all things to all people simultaneously ... and then lock you in there. IBM's contributions generally (to unix), and this one specifically (to Linux), only add to the choices of tools available to those of us who build and maintain our systems. Having their LVM available only increases choice. You don't *have* to use it, but it sure is nice to have available to use if you need or want it.
So Quit Bitchin.
I have administered AIX systems, and while there were things I didn't like about it, there were many that I felt were awesome. The JFS/LVM system was one of those. In specific instances it was an ass-saver.
Technology that saves a sysadmins ass will always have value.
From the conclusion:
This architecture offers many advantages over previous designs, particularly in the areas of expandability, flexibility, and maintenance, and represents a new approach to logical volume management.
Whether or not it lives up to that praise will have to be closely examined.
"If one is really a superior person, the fact is likely to leak out without too much assistance" -- John Andrew Holmes
IBM's LVM doesn't support true mirroring. Maybe I'm wrong and it's changed. But three years ago it didn't.
The simple truth is that interstellar distances will not fit into the human imagination
- Douglas Adams
Since Linux is open source, and still has an important constituant market in smaller systems, there is no reason to think that incorporation of LVMs has to be exclusive to the current simple partitioning systems. LVMs should remain an optional thing for people who want or need them.
Support for LVMs, particularly the IBM one, if it is compatible with AIX could be important for Linux acceptance in the large enterprise market for that very reason. For that reason alone, I am happy to see IBM bring this to Linux.
Think about it. IBM sells a *wide* variety of products, from laptops to mainframes and from software to service. It complicates matters for them if they have to be "multilingual" with their platform systems; Linux gives them a way around this by, as suggested in the Halloween documents, treating the OS as basically a generic commodity.
The more popular Linux gets, and the better IBM's equipment works with it, the better their bottom line gets. If Linux gives them a platform to standardize around, then they can sell more hardware and support services and not have to worry as much about certain problems (compatibility, stability, etc) that live at the OS level. This is very much in their interest, but it's also in the interest of consumers as well, if Linux becomes a reliable system component along the lines of, say, a hard drive or a word processor. The consumer gets the benefit or a robust & rewarding system platform that works on most any equipment available; IBM gets the benefit of having their entire product range appeal to customers because it all speaks the lingua franca of the computer world that they are positively contributing to.
Consider EBCDIC. Wasn't that the standard text format on IBM machines back in the day? Is it in use on any modern system? [I'm asking -- not rhetorical...] It seems to me that they had the wisdom to (eventually) switch to the open ASCII standard before, and are promoting the new Linux standard now. I see no reason to be suspicious of this behavior -- everyone stands to benefit, or at least everyone that isn't trying to shill a proproetary in-house OS... :)
DO NOT LEAVE IT IS NOT REAL
I myself wear birks, shorts, etc, whenever the weather dictates.
STFU about slashdot bias.
There are better solutions to the problem LVs are supposed to address:
IBM's LVM was one of the reasons I hated using AIX (they did similarly oddball and nonstandard stuff in some other areas). I consider it a poorly designed facility. While we can't keep people from porting stuff to Linux, I hope Linux distributions will not incorporate that kind of nonsense; Linux configuration and system management needs to get simpler, not more complex.
The drive for systems like LVM is understandable because UNIX and Linux file systems and large scale data management are not perfect. For example, a big ISP that runs out of disk space on some important partition and needs more space quickly has a legitimate problem. But rather than rushing to a half-baked solution like LVMs, let's identify what the problems are we really want to solve and come up with good solutions to them. With upcoming technologies like object-based disk storage systems, there seem to be much more straightforward and reliable answers than LVM.
your second line still groks code? wow... we're convinced ours' head would explode like in that dilbert cartoon. (of course that means he'd have to like, show up once in a while... but that's a different story.)
after working with AIX over the course of the last year, all i have to say is this:
i would rather embrace the spawn of satan than to have to work with AIX anymore!
the problem with aix is that it does everything for you. this is GREAT for routine stuff - adding a new fs, changing ip addresses, etc. but when you want to do something slightly different, or do it a certian way, you're screwed. i've found that its easier to reinstall the OS in some cases than it is to try to make it work.
but, then again, some people think i'm on crack.
*shrug*
Forgive my lack of enthusiasm, but after working with IBM's LVM on AIX (pronounced aches) I don't see this as anything all that great. Now when Veritas jumps on the bandwagon I'll dance in the streets!
Allrighty, everybody. I think that anyone who starts to think "why would we need another LVM?" should /definitely/ go read the LVMS whitepaper. It's not the same thing as the previous LVM or LVM's that may have existed. Go. Read.
/definitely/ something that the Linux community is going to want to jump at.
/has/ changed alot. They're actually pretty damned cool! I would highly recommend anyone looking for a cool Linux-related job to consider IBM. Hey, look at me! I'm getting paid to write Open-Source Software!
Let me tell you one of the neato things about working at IBM in Austin - I got to talk to Ben (yes, Ben Rafanello, the guy who posted the stuff to the linux mailing lists) all about this last night, and he told me he was planning on posting that stuff. So anyway, I read the LVMS whitepaper last night, and it is
For those of you who are wondering about what license it will be released under - well, I can't say for sure. Just remember that IBM GPL'd their JFS, and we continue to work on that here as well. (I'm actually working on the JFS here in Austin.)
And yes, IBM
-Will the Chill
Creator of RPerl, Scouter, Juggler, Mormon, Perl Monger, Serial Entrepreneur, Aspiring Astrophysicist, Community Organiz