I've been a Slackware user since '95 or so. i've tried RedHat, SuSE, Debian, LFS... I have to agree with the original poster; Debian is a horrible, horrible distro to show someone as a demonstration of what Linux can do.
Sure it's got 10 bazillion packages but if you're doing anything but server stuff you'll be using unstable, and at that point you're actually worse off than Slackware. Broken packages, wrong deps and unstable 'unstable' servers are just the beginning. Debian users (or at least the packagers) are fanatics about the whole "Gee Enn You Slash" in front of Linux to boot.
So far out of all of the distros I've tried, SuSE seems to take the cake for simplicity and ease of use. Yes, this is from a hardcore Slackware user. I find Slackware to be the epitome of "do what you want," and is the distro of choice if you have a Resident Smart Guy who will be maintaining it, but if you want a distro that just works, SuSE is the way to go, IMO.
I didn't say it didn't have the capability; What I was getting at was that in all my experimenting I have been unsuccessful ingetting Microchip's MPLAB-IDE working with the parallel-port ICE under WINE. Questions to the WINE list and IRC channel went largely unanswered. Win4Lin worked right out of the box and without any hassle whatsoever.
I didn't mean to imply that it was impossible under WINE, merely (very) difficult.
I know that WINE can hit pretty much every Win32 API call; the problem is that it takes so goddamed long to try and get it to work with some apps (Acrobat 5 and 6, IE, MPLAB-IDE, I could go on...) that it's largely just plain old not worth my time. I'm not a programmer (well I am, but not an application programmer) -- I don't have the patience to try and reverse-engineer the crap that Win32 is, and I am really glad that others do, and I support them. But when I need to get something to work, I am not going to spend wads of time doing so; I'll purchase a program that does it.
Hopefully someday I will be able to purchase a copy of WINE that does what I need it to do. I really do think it is the best solution for an MS-tax free computer, but it's not quite ready yet. That is also why I said it seems to be filling in a lot of little niche areas (winex, crossover, etc.) where the focus is getting a smaller set of apps functioning, but having them function 100% correctly. This is a good thing.
It comes with prepatched kernels for RH, SuSE, etc.. That was always one of my biggest gripes with W4L too; use a damned kernel module like VMWare. At the time I couldn't get the kernel patch to work with my (already patched) kernel.
Oh I don't know... better app support? The ability to access the printer port from win32 apps (Microchip MPLAB comes immediately to mind, doesn't work at all under WINE if you want ot use the expensive ICE you bought), Acrobat 5 or 6, IE for VoloView (AutoCAD drawing viewer)... Lots of little things.
WINE seems to be a very useful "meta project" -- take the bits of WINE you need to get the specific windows functionality you need, but you'll never get 100% of it. I'm not sure why.
Transactions are not a panacea and their automatic use needs to be considered and considered carefully and for every application. A sweeping rule of transactions everywhere is madness.
I completely agree with you here. There's a statement about hammers and nails to be said in this situation, but I'm sure you know it already.
With MVCC, readers do not block writers, and writers do not block readers, but, sometimes, a rollback in one transaction MUST force a cascading rollback in downstream transactions. The database stalls as do all of your users.
Agreed, but cascading rollback is not the rule, it is an exception, and careful database design (i.e. not just throwing up a table, constraints and relations helter-skelter) can make a costly rollback operation one of those "one in a million" types of occurances. Especially in a file system, where I simply can't see cascading rollback being an issue.
Of course I assume a serializable isolation level. If you are going to have that, then, what's the point of having ACID? You can't have ACID without that I for Isolated!
What I meant by that is if two queries both write to the table and they do not have any interdependencies, they are not serialized. If one fails, it will not cause the other to block.
So, where could there be contention? Let's see, in Windows, we have the Registry and its relational replacement, we have the amount of disk space on a volume, we have the contents of a particular file folder. What happens to applications that look for disk free in the middle of a copy of a file folder with 50 files? What happens when you are copying a DVD into a folder? Do you lock the folder query while the DVD is writing?
Looking for disk free is a read operation. Copying a DVD into a folder is a write operation. With MVCC you have two scenarios: The "get disk free" operation and the "write a dozen.VOB files" operation. If the former starts before the latter, the disk space operation will not see the 4G or so of DVD files. If the latter starts before the former, the results can vary. If the write is one transaction, it has not finished yet so the read doesn't see the affected rows, instead seeing the "version" before. If the write is a transaction for every.VOB file, then you'll see the last completed copy and the disk space will return something between x and x+4G or so.
I fail to see any blocking condition. Now if they lock the directory and you're serializing everything, sure you'll block. That's poor DB design, IMO, and not a description of SQL suckage.
You guys that admire SQL Server because of a few dozen tables and some little data sets (by little I mean less than 100,000,000 rows), have no idea what you are talking about. Database based file systems with transactions are madness.
SQL has done a lot of good work for me. You're right, I have nothing near the hundred million row mark (I think the biggest I've worked on is about the hundred thousand row mark) -- regardless, the idea is the same and poor DB design is poor DB design.
I used to use a nt password reset boot disk a few years ago that was based on Linux. I could change admin passwords on NTFS disks just fine. Not exactly running a database on NTFS but it never corrupted any disks.
Um, parent said in all situations. Your password recovery use is hardly all situations. I too have used NTFS write support without issue but I trust the driver author when he says that he forsees problems in some situations and to only use write support when necessary.
An ACID file system would mean that images would probably take 10 days to write, and that, if someone was writing a long file, and a short file, all users of the short file would have to wait until the long file write was complete.
Don't know much about databases, do you... you don't necessarily have to lock the entire DB when running a long-executing query -- only if it's affecting the entire db.
Hell you hardly have to lock the table to perform most transactions, long-executing or not. If the long-write and short-write must be serialized, you're absolutely correct. But if they don't affect each other (or each other's rows, which is probably the case with almost all filesystem transactions) then your statement is bullshit. I doubt very much that MS is dumb enough to lock the metadata table(s) for the entire period of a file update/write.
The mail spool for a 15k-user ISP in southwestern Ontario is running on Slack9 + LVM (Reiserfs). It exports the spool via NFS and the edge servers (SMTP+IMAP4+POP3, virus+spamscan) mount the spool directly over ipsec. No issues. I can grow the filesystem, take snapshots and it all just works. The PostgreSQL database is also on an LVM volume, but I haven't had to do much with it related to LVM yet, as pg_dump works live.
I have a number of other mail spools for businesses around the area (probably a half dozen to 10 or so) -- around 100-300 users each. Same story, backups are very nice when you can just take a snapshot.
I'm currently evaluating Appgen Custom Suite as a replacement for our current Accpac and Misys accounting and inventory/manufacturing infrastructure. LVM is very nice here for growing the db partition and snapshotting means the system is only down for seconds instead of the time it'd take to dump the entire db to tape.
All in all, I am very pleased with LVM. I do all my LVM on software RAID1 or hardware RAID5, but there's been no interaction or badness show up so far. I realize it's possible to do RAID1 with LVM but there's no documentation on how to "un-fail" the volume after a drive fails or how to manage it, so until then md or the hardware tools will have to do.:-)
You think you've got troubles... I'm using RAV for a 15k-user ISP and I need a solution that doesn't involve multiple layers of SMTP servers just so I can present SMTP TLS to the world.
And what, praytell, is wrong with shipping a package (your choice, tarball, rpm, deb, whatever) that includes the modified kernel and module set?
I realize it's not optimal, but just because people haven't taken the time beforehand to see if alternate HZ values work well (and perhaps coding a little to return scheduling to normal) doesn't mean it's not possible. Ok, WinXP does this better out of the box. That's fine. You've got the source, go fix it yourself and contribute the changes back to the community so Linux will do it out of the box as well instead of whinging about it.
You can bet your sweet bippy that you couldn't have fixed a problem like this in WinXP.
In Canada (unlike the US where these are outlawed) you can buy scanners at your local Radio Shack that will pick up the complete cellphone and police bands.
Untrue. the NAMPS cell phone bands were blocked in all North American scanners. I have no idea about the other bands, as I've been out of cell hacking for many years now. I could trick the receiver into blocking an alternative set of bands, or I could replace the main processor in the scanner with the identical model from the UK, but the scanner as sold did not allow listening of cell phones.
That is just one scanner, I realize, but all of them are (were) like that.
I just wish (and I've logged feature requests in both SO and OO) that they'd include a "Save Copy As" menu item so that I could rattle off a.doc format without have to go back and run Save As again to get back to OO format.
Both stereo channels have to be sampled at this rate, or you end up with 2 11MHz signals which will sound bad to just about anyone.
OK I know that harmonics can play into the overall sound by beating with each other, but I would be hard-pressed to believe that there is enough energy in the harmonic content of any musical instrument (voice or otherwise) up around 11MHz. Hell I bet there's nothing of significance above 50 or 60kHz.
I wI was evaluating this software just before Oracle bought them. Could you email me so I could ask you some questions about it?
We're currently looking at SuSE Linux OpenExchange server and you just jogged my memory about Steltor.
I dunno, I run my NFS over IPSec and it seems to work just fine. A simple script to block any NFS access that isn't coming in on an ipsec interface and you're all set. rpcinfo and some awk, that's all it takes.
If you'd rather have a 2500 series router over a $500 Linux box, you're on crack. There's no comparison. Those pieces of shit (the 2500s) can't handle more than about a T1's worth of traffic, can't do any kind of CAR or shaping, and can barely handle a few dozen ACLs. And yes, Virginia, you need to use telnet to get at them, because they don't support any kind of encryption.
No thanks. Now if you were talking a 2600-series for a small business or ISP, you're still on drugs, but it's only weed. I'd have to take a good look at what I need if presented with a stock Slackware install vs. a Cisco 2600-series router.
If you put your ISP on a commodity linux box and expect five 9's you need to back off the medication.
While not five 9's, I do run an ISP off of commodity Linux boxes and achieve three 9's (8.77 hours out of the year downtime) -- we're a commercial ISP and frankly, if that's not good enough for you, go buy someone else's service. I can't get three 9's downtime out of my upstream ISP if you count the scheduled downtime (which my three 9's figure does count).
I genuinely tried to install linux for a desktop OS, and ran into more problems with printing than anything else.
I had the same troubles until I settled on CUPS and KDE. Absolutely z.e.r.o. problems. This is with high-end (duplexing, finishing, faxing) mopiers and plain-jane inkjets too. Just beware the WinPrinters.
PHP is the epitome of mixing content and presentation. Sure you can separate it but it just doesn't seem to be aimed at doing so. Mason seems to be better at it -- it at least tries to enforce separation.
No, I don't have a good replacement -- but PHP as a good web content generator is not a good idea IMO.
Re:3 days to a week to compile?
on
Gentoo Reviewed
·
· Score: 1
You're on crack. A total KDE CVS compile (including kdenonbeta and kdeextra modules), using -O3, -mpentium3, --enable-final and prelinking takes no more than about 7 hours. kdelibs and base are both about 90 minutes on a Pentium III/1G with 256M of memory, with kdenetwork being the next largest and everything else taking between 10 and 40 minutes.
How the hell did you learn anything with gentoo other than running emerge and looking at the forums for answers? If you want to learn a distro, use LFS or Slackware; the actual quantity of what you learn with Gentoo is not exactly all that great.
Re:The problem with Gentoo
on
Gentoo Reviewed
·
· Score: 0, Troll
The only time I've had trouble with Slack was at the very beginning... Mind you my beginning with Linux was back in '95.
The only other distro which has made me stand up and take notice has been SuSE; it's almost a totally different Linux than Slack -- built for the corporate environment -- but between setting up SuSE for people who don't want to know the nitty-gritty and then using Slack for myself to keep my edge (well and LFS for my CompactFlash firewalls), the rest of the distros can go to hell; they're either a less featureful version of SuSE or a gone-too-far vesion of Slack.:-)
Slashdot Mirror
I've been a Slackware user since '95 or so. i've tried RedHat, SuSE, Debian, LFS... I have to agree with the original poster; Debian is a horrible, horrible distro to show someone as a demonstration of what Linux can do.
Sure it's got 10 bazillion packages but if you're doing anything but server stuff you'll be using unstable, and at that point you're actually worse off than Slackware. Broken packages, wrong deps and unstable 'unstable' servers are just the beginning. Debian users (or at least the packagers) are fanatics about the whole "Gee Enn You Slash" in front of Linux to boot.
So far out of all of the distros I've tried, SuSE seems to take the cake for simplicity and ease of use. Yes, this is from a hardcore Slackware user. I find Slackware to be the epitome of "do what you want," and is the distro of choice if you have a Resident Smart Guy who will be maintaining it, but if you want a distro that just works, SuSE is the way to go, IMO.
Fix your damned mail/listserver, the KWIUG guys are unable to get mail out!
I didn't say it didn't have the capability; What I was getting at was that in all my experimenting I have been unsuccessful ingetting Microchip's MPLAB-IDE working with the parallel-port ICE under WINE. Questions to the WINE list and IRC channel went largely unanswered. Win4Lin worked right out of the box and without any hassle whatsoever.
I didn't mean to imply that it was impossible under WINE, merely (very) difficult.
I know that WINE can hit pretty much every Win32 API call; the problem is that it takes so goddamed long to try and get it to work with some apps (Acrobat 5 and 6, IE, MPLAB-IDE, I could go on...) that it's largely just plain old not worth my time. I'm not a programmer (well I am, but not an application programmer) -- I don't have the patience to try and reverse-engineer the crap that Win32 is, and I am really glad that others do, and I support them. But when I need to get something to work, I am not going to spend wads of time doing so; I'll purchase a program that does it.
Hopefully someday I will be able to purchase a copy of WINE that does what I need it to do. I really do think it is the best solution for an MS-tax free computer, but it's not quite ready yet. That is also why I said it seems to be filling in a lot of little niche areas (winex, crossover, etc.) where the focus is getting a smaller set of apps functioning, but having them function 100% correctly. This is a good thing.
Ok, colour me impressed; what is involved in getting IE5.5/IE6 working under WINE? I am _very_ interested in this.
It comes with prepatched kernels for RH, SuSE, etc.. That was always one of my biggest gripes with W4L too; use a damned kernel module like VMWare. At the time I couldn't get the kernel patch to work with my (already patched) kernel.
Oh I don't know... better app support? The ability to access the printer port from win32 apps (Microchip MPLAB comes immediately to mind, doesn't work at all under WINE if you want ot use the expensive ICE you bought), Acrobat 5 or 6, IE for VoloView (AutoCAD drawing viewer)... Lots of little things.
WINE seems to be a very useful "meta project" -- take the bits of WINE you need to get the specific windows functionality you need, but you'll never get 100% of it. I'm not sure why.
Transactions are not a panacea and their automatic use needs to be considered and considered carefully and for every application. A sweeping rule of transactions everywhere is madness.
I completely agree with you here. There's a statement about hammers and nails to be said in this situation, but I'm sure you know it already.
With MVCC, readers do not block writers, and writers do not block readers, but, sometimes, a rollback in one transaction MUST force a cascading rollback in downstream transactions. The database stalls as do all of your users.
Agreed, but cascading rollback is not the rule, it is an exception, and careful database design (i.e. not just throwing up a table, constraints and relations helter-skelter) can make a costly rollback operation one of those "one in a million" types of occurances. Especially in a file system, where I simply can't see cascading rollback being an issue.
Of course I assume a serializable isolation level. If you are going to have that, then, what's the point of having ACID? You can't have ACID without that I for Isolated!
What I meant by that is if two queries both write to the table and they do not have any interdependencies, they are not serialized. If one fails, it will not cause the other to block.
So, where could there be contention? Let's see, in Windows, we have the Registry and its relational replacement, we have the amount of disk space on a volume, we have the contents of a particular file folder. What happens to applications that look for disk free in the middle of a copy of a file folder with 50 files? What happens when you are copying a DVD into a folder? Do you lock the folder query while the DVD is writing?
Looking for disk free is a read operation. Copying a DVD into a folder is a write operation. With MVCC you have two scenarios: The "get disk free" operation and the "write a dozen .VOB files" operation. If the former starts before the latter, the disk space operation will not see the 4G or so of DVD files. If the latter starts before the former, the results can vary. If the write is one transaction, it has not finished yet so the read doesn't see the affected rows, instead seeing the "version" before. If the write is a transaction for every .VOB file, then you'll see the last completed copy and the disk space will return something between x and x+4G or so.
I fail to see any blocking condition. Now if they lock the directory and you're serializing everything, sure you'll block. That's poor DB design, IMO, and not a description of SQL suckage.
You guys that admire SQL Server because of a few dozen tables and some little data sets (by little I mean less than 100,000,000 rows), have no idea what you are talking about. Database based file systems with transactions are madness.
SQL has done a lot of good work for me. You're right, I have nothing near the hundred million row mark (I think the biggest I've worked on is about the hundred thousand row mark) -- regardless, the idea is the same and poor DB design is poor DB design.
I used to use a nt password reset boot disk a few years ago that was based on Linux. I could change admin passwords on NTFS disks just fine. Not exactly running a database on NTFS but it never corrupted any disks.
Um, parent said in all situations. Your password recovery use is hardly all situations. I too have used NTFS write support without issue but I trust the driver author when he says that he forsees problems in some situations and to only use write support when necessary.
An ACID file system would mean that images would probably take 10 days to write, and that, if someone was writing a long file, and a short file, all users of the short file would have to wait until the long file write was complete.
Don't know much about databases, do you... you don't necessarily have to lock the entire DB when running a long-executing query -- only if it's affecting the entire db.
Hell you hardly have to lock the table to perform most transactions, long-executing or not. If the long-write and short-write must be serialized, you're absolutely correct. But if they don't affect each other (or each other's rows, which is probably the case with almost all filesystem transactions) then your statement is bullshit. I doubt very much that MS is dumb enough to lock the metadata table(s) for the entire period of a file update/write.
The mail spool for a 15k-user ISP in southwestern Ontario is running on Slack9 + LVM (Reiserfs). It exports the spool via NFS and the edge servers (SMTP+IMAP4+POP3, virus+spamscan) mount the spool directly over ipsec. No issues. I can grow the filesystem, take snapshots and it all just works. The PostgreSQL database is also on an LVM volume, but I haven't had to do much with it related to LVM yet, as pg_dump works live.
I have a number of other mail spools for businesses around the area (probably a half dozen to 10 or so) -- around 100-300 users each. Same story, backups are very nice when you can just take a snapshot.
I'm currently evaluating Appgen Custom Suite as a replacement for our current Accpac and Misys accounting and inventory/manufacturing infrastructure. LVM is very nice here for growing the db partition and snapshotting means the system is only down for seconds instead of the time it'd take to dump the entire db to tape.
All in all, I am very pleased with LVM. I do all my LVM on software RAID1 or hardware RAID5, but there's been no interaction or badness show up so far. I realize it's possible to do RAID1 with LVM but there's no documentation on how to "un-fail" the volume after a drive fails or how to manage it, so until then md or the hardware tools will have to do. :-)
You think you've got troubles... I'm using RAV for a 15k-user ISP and I need a solution that doesn't involve multiple layers of SMTP servers just so I can present SMTP TLS to the world.
And what, praytell, is wrong with shipping a package (your choice, tarball, rpm, deb, whatever) that includes the modified kernel and module set?
I realize it's not optimal, but just because people haven't taken the time beforehand to see if alternate HZ values work well (and perhaps coding a little to return scheduling to normal) doesn't mean it's not possible. Ok, WinXP does this better out of the box. That's fine. You've got the source, go fix it yourself and contribute the changes back to the community so Linux will do it out of the box as well instead of whinging about it.
You can bet your sweet bippy that you couldn't have fixed a problem like this in WinXP.
In Canada (unlike the US where these are outlawed) you can buy scanners at your local Radio Shack that will pick up the complete cellphone and police bands.
Untrue. the NAMPS cell phone bands were blocked in all North American scanners. I have no idea about the other bands, as I've been out of cell hacking for many years now. I could trick the receiver into blocking an alternative set of bands, or I could replace the main processor in the scanner with the identical model from the UK, but the scanner as sold did not allow listening of cell phones.
That is just one scanner, I realize, but all of them are (were) like that.
I just wish (and I've logged feature requests in both SO and OO) that they'd include a "Save Copy As" menu item so that I could rattle off a .doc format without have to go back and run Save As again to get back to OO format.
Both stereo channels have to be sampled at this rate, or you end up with 2 11MHz signals which will sound bad to just about anyone.
OK I know that harmonics can play into the overall sound by beating with each other, but I would be hard-pressed to believe that there is enough energy in the harmonic content of any musical instrument (voice or otherwise) up around 11MHz. Hell I bet there's nothing of significance above 50 or 60kHz.
Yes, there is a tongue-in-cheek here...
I wI was evaluating this software just before Oracle bought them. Could you email me so I could ask you some questions about it? We're currently looking at SuSE Linux OpenExchange server and you just jogged my memory about Steltor.
I dunno, I run my NFS over IPSec and it seems to work just fine. A simple script to block any NFS access that isn't coming in on an ipsec interface and you're all set. rpcinfo and some awk, that's all it takes.
hahahaha... yes you're absolutely right; I did mean that I can't get three 9's uptime out of my upstream provider, including scheduled downtime.
If you'd rather have a 2500 series router over a $500 Linux box, you're on crack. There's no comparison. Those pieces of shit (the 2500s) can't handle more than about a T1's worth of traffic, can't do any kind of CAR or shaping, and can barely handle a few dozen ACLs. And yes, Virginia, you need to use telnet to get at them, because they don't support any kind of encryption.
No thanks. Now if you were talking a 2600-series for a small business or ISP, you're still on drugs, but it's only weed. I'd have to take a good look at what I need if presented with a stock Slackware install vs. a Cisco 2600-series router.
If you put your ISP on a commodity linux box and expect five 9's you need to back off the medication.
While not five 9's, I do run an ISP off of commodity Linux boxes and achieve three 9's (8.77 hours out of the year downtime) -- we're a commercial ISP and frankly, if that's not good enough for you, go buy someone else's service. I can't get three 9's downtime out of my upstream ISP if you count the scheduled downtime (which my three 9's figure does count).
I genuinely tried to install linux for a desktop OS, and ran into more problems with printing than anything else.
I had the same troubles until I settled on CUPS and KDE. Absolutely z.e.r.o. problems. This is with high-end (duplexing, finishing, faxing) mopiers and plain-jane inkjets too. Just beware the WinPrinters.
PHP? You've gotta be kidding.
PHP is the epitome of mixing content and presentation. Sure you can separate it but it just doesn't seem to be aimed at doing so. Mason seems to be better at it -- it at least tries to enforce separation.
No, I don't have a good replacement -- but PHP as a good web content generator is not a good idea IMO.
You're on crack. A total KDE CVS compile (including kdenonbeta and kdeextra modules), using -O3, -mpentium3, --enable-final and prelinking takes no more than about 7 hours. kdelibs and base are both about 90 minutes on a Pentium III/1G with 256M of memory, with kdenetwork being the next largest and everything else taking between 10 and 40 minutes.
How the hell did you learn anything with gentoo other than running emerge and looking at the forums for answers? If you want to learn a distro, use LFS or Slackware; the actual quantity of what you learn with Gentoo is not exactly all that great.
The only time I've had trouble with Slack was at the very beginning... Mind you my beginning with Linux was back in '95.
:-)
The only other distro which has made me stand up and take notice has been SuSE; it's almost a totally different Linux than Slack -- built for the corporate environment -- but between setting up SuSE for people who don't want to know the nitty-gritty and then using Slack for myself to keep my edge (well and LFS for my CompactFlash firewalls), the rest of the distros can go to hell; they're either a less featureful version of SuSE or a gone-too-far vesion of Slack.