Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
"Ben Hutchings" <[EMAIL PROTECTED]> writes: > Absolutely not. The point of keeping a journal is to preserve the > integrity of the filesystem even in an operation is unexpectedly > interrupted (whether by a crash or a power failure). This should > provide resilience to disconnection, too, but that doesn't mean > disconnection without warning would be safe: the integrity of the > filesystem is not the same as the integrity of every file within it. That's not what we're talking about here, though, if you follow the thread. You expect some corruption when you yank out a disk without proper unmounting, if you are unlucky. You don't expect a kernel panic when you do. Well, at least I don't. When I'm talking about it would be 'fine' to yank out a usbkey formated in FAT or ext2 without unmounting it first, I'm talking about the fact it won't cause a kernel panic (again, the context would have been apparent if you followed the thread). -- John L. Fjellstad web: http://www.fjellstad.org/ Quis custodiet ipsos custodes -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
John L Fjellstad wrote: William Ballard <[EMAIL PROTECTED]> writes: I have an external USB 2.0 2.5" hard drive, shows up as /dev/sda, with an ReiserFS partition. If I yank it with it mounted ReiserJS panics. I have the fstab entry marked as sync, rw. I think it might be because reiserfs is a journaling filesystem. I know I sometimes forget and remove my usbkey (which has FAT) without unmounting, and don't see the problems you see. Even ext2 should be fine. Absolutely not. The point of keeping a journal is to preserve the integrity of the filesystem even in an operation is unexpectedly interrupted (whether by a crash or a power failure). This should provide resilience to disconnection, too, but that doesn't mean disconnection without warning would be safe: the integrity of the filesystem is not the same as the integrity of every file within it. If an ext2 filesystem is not cleanly unmounted then there is no guarantee that it will be consistent or quickly repairable. It must generally be fully scanned and repaired by e2fsck. If the filesystem is not mounted with metadata ordering enabled, even that won't necessarily work. FAT isn't resilient to interruption either. FAT filesystem implementations tend to be extremely conservative in ordering of changes, which may make it relatively safe to disconnect a FAT filesystem without unmounting it. However, there's no way of telling whether a FAT filesystem has been cleanly umounted or not, so filesystem damage can easily go unnoticed for some time, giving a false sense of the safety of this practice. Ben. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
William Ballard <[EMAIL PROTECTED]> writes: > [XP-Focused things] > -Install and run multi-GB map software and games in WinXP > -Keep CD images such as Visual Studio and Map software on it, > so they'll always be available (XP Only) > -Treat it as a "general data store" for Ripped DVDs, Install Programs, > my document store, and transferring large projects; (XP/Linux/Both) > > [Linux-focused things] > -Keep VMWare images on it (used in Linux) > -Keep Partimage images of XP O/S partition on it (used in Linux) > -Keep a music library on it (used in Linux) Looking at your usage, I would probably go with ext2 or xfs. XFS is supposed to be the best filesystem for big files. Then again, I'm not sure you need a journaled filesystem if you only read from the system (but someone can probably correct me on that). -- John L. Fjellstad web: http://www.fjellstad.org/ Quis custodiet ipsos custodes -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Fri, Oct 29, 2004 at 10:38:15AM -0700, John L Fjellstad wrote: > I think it might be because reiserfs is a journaling filesystem. I know > I sometimes forget and remove my usbkey (which has FAT) without > unmounting, and don't see the problems you see. Even ext2 should be > fine. Never tried with a journaled filesystem. Like someone else Switching gears to "what would you do": It's an 80GB hd @ 4200RPM, available after upgrading my laptop to 60GB @ 7200RPM. Would like to do the following with the external HD, in no particular order of priority: [XP-Focused things] -Install and run multi-GB map software and games in WinXP -Keep CD images such as Visual Studio and Map software on it, so they'll always be available (XP Only) -Treat it as a "general data store" for Ripped DVDs, Install Programs, my document store, and transferring large projects; (XP/Linux/Both) [Linux-focused things] -Keep VMWare images on it (used in Linux) -Keep Partimage images of XP O/S partition on it (used in Linux) -Keep a music library on it (used in Linux) Ideally, I'd format the whole 80GB as a giant Fat32 partition, but you can't make it that big. So I settled on 50GB NTFS for XP to write to and Linux to read from, and 30GB Reiserfs for Linux to write to and XP to read from. Thus I get the benefits of small clusters and no wasted space in both. Would I be better off using the 30GB as Ext2; to allow safe yanking; or are the benefits of ReiserFS such greater as to keep it the way it is and hope autofs gets fixed in sid. My inclination is to keep it the way it is and wait for autofs to start working again. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
William Ballard <[EMAIL PROTECTED]> writes: > I have an external USB 2.0 2.5" hard drive, shows up as /dev/sda, with > an ReiserFS partition. If I yank it with it mounted ReiserJS panics. > I have the fstab entry marked as sync, rw. I think it might be because reiserfs is a journaling filesystem. I know I sometimes forget and remove my usbkey (which has FAT) without unmounting, and don't see the problems you see. Even ext2 should be fine. Never tried with a journaled filesystem. Like someone else mentioned in this thread, it might be because the journal hasn't been updated yet. Probably is a bug in reiserfs too (it should recover more gracefully). BTW, checking the manual for mount(8), only ext2, ext3 and ufs supports the sync keyword). -- John L. Fjellstad web: http://www.fjellstad.org/ Quis custodiet ipsos custodes -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Thu, 28 Oct 2004, Ron Johnson wrote: > On Thu, 2004-10-28 at 17:21 -0300, Henrique de Moraes Holschuh wrote: > > It has little to do with USB. It has all to do with the hotplug kernel > > interface, though. > > That's interesting. > CONFIG_HOTPLUG=y > # CONFIG_HOTPLUG_PCI is not set > CONFIG_USB_STORAGE_JUMPSHOT=y > > Note how on my system, HOTPLUG is enabled, but HOTPLUG_PCI is not. > Yet, USB hotplugging works like a charm. As it should. What I meant is that *user space* hotplug handles BOTH because the interface to the kernel is the _same_ AFAIK. Not that you need PCI hotplug in the kernel for USB hotplug to work. In fact, you probably don't need it even for CardBus, since there is a legacy interface for that one I think... -- "One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique Holschuh -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Thu, 2004-10-28 at 23:05 -0400, William Ballard wrote: > On Thu, Oct 28, 2004 at 09:12:40PM -0500, Ron Johnson wrote: > > Floppy disk, CD-Rs & flash drives are a big step down from a multi- > > dozen GB-sized HDD. > > He's testing it on HDs, but getting a lot of feedback about how fat and > ReiserFs don't support sync, and if they did support sync it would be > slow as the dickens. There is a suggestion to use a "fastusb " > option which would transparently remount as async for the duration of a > command, and remount sync when done [1]. As long as you don't yank > during an async op, you shouldn't expect data loss. > > Pretty fascinating project -- just started 2 days ago. Look at the > "usbmount" thread on [EMAIL PROTECTED] > > [1] Pretty sure XPSP2 does somethin like this, async/sync hybrid with > more aggressive flushes, when things "settle down." You used to have to > manually stop a mass storage device, bad things would happen if you > didn't, but that's less true now. Interesting. I wonder if the same issues occur on/with firewire drives (that's what I have). -- - Ron Johnson, Jr. Jefferson, LA USA PGP Key ID 8834C06B "Experience hath shewn, that even under the best forms [of government] those entrusted with power have, in time, and by slow operations, perverted it into tyranny." Thomas Jefferson signature.asc Description: This is a digitally signed message part
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Thu, Oct 28, 2004 at 09:12:40PM -0500, Ron Johnson wrote: > Floppy disk, CD-Rs & flash drives are a big step down from a multi- > dozen GB-sized HDD. He's testing it on HDs, but getting a lot of feedback about how fat and ReiserFs don't support sync, and if they did support sync it would be slow as the dickens. There is a suggestion to use a "fastusb " option which would transparently remount as async for the duration of a command, and remount sync when done [1]. As long as you don't yank during an async op, you shouldn't expect data loss. Pretty fascinating project -- just started 2 days ago. Look at the "usbmount" thread on [EMAIL PROTECTED] [1] Pretty sure XPSP2 does somethin like this, async/sync hybrid with more aggressive flushes, when things "settle down." You used to have to manually stop a mass storage device, bad things would happen if you didn't, but that's less true now. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Thu, 2004-10-28 at 20:13 -0400, William Ballard wrote: > On Thu, Oct 28, 2004 at 06:12:18PM -0500, Ron Johnson wrote: > > / and /boot are formatted ReiserFS? > Yah, so I get what I deserve I guess. Jeez, touchy. I was only asking for verification purposes. If / and /boot were "something else", and reiserfs were a module, then *maybe* you could rmmod/insmod it and the system wouldn't have to die. Oh well. (I'm not going to test that either...) > > So a "shutdown -t1 -now -r" on another console TTY just sits there? > > Yes, as does "halt" and "reboot" and "Ctrl+Alt+Delete". Halt and reboot > say they are doing things then nothing happens. Control returns to that > bash prompt, and you can still Alt+F2 to another TTY. Interesting. Good to know, though. [snip] > > You must be thinking of someone else. I went back and checked > > all of my responses on this thread and nowhere does it mention > > "dumb" or "poorly asked". It doesn't even infer them, IMO. > > Okay, I don't want to get into an argument about this either. I said > "how can I yank a mounted partition" and you said "it's stupid to yank a > mounted partition or even think you should be able to" meanwhile another > guy has in the same timeframe written a script called usbmount that lets > you yank a mounted partition! Floppy disk, CD-Rs & flash drives are a big step down from a multi- dozen GB-sized HDD. ReiserFS is also a heck of a lot more complicated than FAT. > But we can move on past that now, I'm not feeling touchy or anything. > What I'm doing now is messing with autofs which has its own problems, > but would probably work for me (it automatically unmounts parts after > disuse for n seconds, That would do the trick > but the current version in sid is broken and never > unmounts *anything* (bug #278076). If it worked. :) >When autofs is working, then I'll > use that and be happy as a tick. -- - Ron Johnson, Jr. Jefferson, LA USA PGP Key ID 8834C06B "The main reason that M$ gets bashed is that they persist in writing bad code, on top of bad code As many have said, there is NO PERFECT OS. The better OS though, IMHO, is the one that will openly deal with issues, both major, and minor. Microsoft still needs a lot of work in this area." http://www.securityfocus.com/columnists/202/comment/24104#MSG signature.asc Description: This is a digitally signed message part
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Thu, Oct 28, 2004 at 06:12:18PM -0500, Ron Johnson wrote: > / and /boot are formatted ReiserFS? Yah, so I get what I deserve I guess. > So a "shutdown -t1 -now -r" on another console TTY just sits there? Yes, as does "halt" and "reboot" and "Ctrl+Alt+Delete". Halt and reboot say they are doing things then nothing happens. Control returns to that bash prompt, and you can still Alt+F2 to another TTY. Ctrl+Alt+Delete says "the system is going down now" then that TTY becomes unresponsive, but you can Alt+F2 to another TTY. > A "cold kill" (aka "hitting the BRS", to us oldsters) should not > damage ReiserFS partitions. That's what journaling is for, after > all. What kernel are you running? I get ReiserFS corruption from time to time. I have always assumed that it was due to rudely killing the system without a clean shutdown -- particuarly if some disk activity was occurring at the time. For instance sometimes I will get corruption under /var/log, since there is a lot of activity there. My "big" partition will get some recoverable corruption from time to time. I always run the latest kernel. It hasn't happened in a while, but I haven't rudely killed the system much lately until this USB HD Reiserfs panic issue. I will just "not do that" as you suggest. I just asked in case anyone knew anything. > You must be thinking of someone else. I went back and checked > all of my responses on this thread and nowhere does it mention > "dumb" or "poorly asked". It doesn't even infer them, IMO. Okay, I don't want to get into an argument about this either. I said "how can I yank a mounted partition" and you said "it's stupid to yank a mounted partition or even think you should be able to" meanwhile another guy has in the same timeframe written a script called usbmount that lets you yank a mounted partition! But we can move on past that now, I'm not feeling touchy or anything. What I'm doing now is messing with autofs which has its own problems, but would probably work for me (it automatically unmounts parts after disuse for n seconds, but the current version in sid is broken and never unmounts *anything* (bug #278076). When autofs is working, then I'll use that and be happy as a tick. Thanks Ron for the dialog. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Thu, 2004-10-28 at 18:20 -0400, William Ballard wrote: > On Thu, Oct 28, 2004 at 01:18:11PM -0500, Ron Johnson wrote: > > The question about PCI hotplug was very honest, since PCI hotplug > > refers to pulling PCI cards out of live systems, and that, to my > > understanding, has nothing to do with USB (which is hotplug by > > definition). > > I actually think the problem lies in ReiserFS: that's the module that > Panics when you yank the drive. The USB subsystem deals with the loss > of it just fine. The system doesn't actually freeze: you get a panic > from the ReiserFS module (which is builtin for me), about being unable > to write to the journal. Meanwhile all other apps continue to work just > fine -- you just cannot halt or reboot the system, it hangs the TTY > then, but you can change to another TTY and keep working. / and /boot are formatted ReiserFS? The warm feeling that I get from ext2 is why / & /boot are ext3 on my system. (/home is reiserfs, though.) > Since this is the status quo, and the system is still responsive after > this ReiserFS panic, is it still possible to cleanly shutdown and reboot > the system in case I accidentally do it? Right now I have to cold kill > the system without a shutdown, which tends to cause further data > corruption on my ReiserFS HD parts. So a "shutdown -t1 -now -r" on another console TTY just sits there? A "cold kill" (aka "hitting the BRS", to us oldsters) should not damage ReiserFS partitions. That's what journaling is for, after all. What kernel are you running? > Do me a favor and if you think it's a dumb question just don't answer > instead of telling me it's a dumb or poorly asked question. Thanks. You must be thinking of someone else. I went back and checked all of my responses on this thread and nowhere does it mention "dumb" or "poorly asked". It doesn't even infer them, IMO. -- - Ron Johnson, Jr. Jefferson, LA USA PGP Key ID 8834C06B "It isn't necessary to be rich and famous to be happy. It's only necessary to be rich." Alan Alda signature.asc Description: This is a digitally signed message part
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Thu, 2004-10-28 at 17:21 -0300, Henrique de Moraes Holschuh wrote: > On Thu, 28 Oct 2004, Ron Johnson wrote: > > On Thu, 2004-10-28 at 12:22 -0400, William Ballard wrote: > > > On Wed, Oct 27, 2004 at 11:23:07AM -0500, Ron Johnson wrote: > > > > > Anything promising using the PCI hotplug facility? > > Cardbus is PCI hotplug, AFAIK. > > > The question about PCI hotplug was very honest, since PCI hotplug > > refers to pulling PCI cards out of live systems, and that, to my > > understanding, has nothing to do with USB (which is hotplug by > > definition). > > It has little to do with USB. It has all to do with the hotplug kernel > interface, though. That's interesting. $ uname -r 2.6.8 $ $ $ zcat /proc/config.gz | grep HOT CONFIG_HOTPLUG=y # CONFIG_HOTPLUG_PCI is not set CONFIG_USB_STORAGE_JUMPSHOT=y Note how on my system, HOTPLUG is enabled, but HOTPLUG_PCI is not. Yet, USB hotplugging works like a charm. -- - Ron Johnson, Jr. Jefferson, LA USA PGP Key ID 8834C06B "What has a tiny brain, a big mouth, and an opinion nobody cares about? You!" from Murphy Brown signature.asc Description: This is a digitally signed message part
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Thu, Oct 28, 2004 at 01:18:11PM -0500, Ron Johnson wrote: > The question about PCI hotplug was very honest, since PCI hotplug > refers to pulling PCI cards out of live systems, and that, to my > understanding, has nothing to do with USB (which is hotplug by > definition). I actually think the problem lies in ReiserFS: that's the module that Panics when you yank the drive. The USB subsystem deals with the loss of it just fine. The system doesn't actually freeze: you get a panic from the ReiserFS module (which is builtin for me), about being unable to write to the journal. Meanwhile all other apps continue to work just fine -- you just cannot halt or reboot the system, it hangs the TTY then, but you can change to another TTY and keep working. Since this is the status quo, and the system is still responsive after this ReiserFS panic, is it still possible to cleanly shutdown and reboot the system in case I accidentally do it? Right now I have to cold kill the system without a shutdown, which tends to cause further data corruption on my ReiserFS HD parts. Do me a favor and if you think it's a dumb question just don't answer instead of telling me it's a dumb or poorly asked question. Thanks. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Thu, 28 Oct 2004, Ron Johnson wrote: > On Thu, 2004-10-28 at 12:22 -0400, William Ballard wrote: > > On Wed, Oct 27, 2004 at 11:23:07AM -0500, Ron Johnson wrote: > > > > Anything promising using the PCI hotplug facility? Cardbus is PCI hotplug, AFAIK. > The question about PCI hotplug was very honest, since PCI hotplug > refers to pulling PCI cards out of live systems, and that, to my > understanding, has nothing to do with USB (which is hotplug by > definition). It has little to do with USB. It has all to do with the hotplug kernel interface, though. -- "One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique Holschuh -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Thu, 2004-10-28 at 12:22 -0400, William Ballard wrote: > On Wed, Oct 27, 2004 at 11:23:07AM -0500, Ron Johnson wrote: > > > Anything promising using the PCI hotplug facility? > > > > Just out of curiosity, but why should *PCI* hotplug have anything > > to do with USB? > > # lspci | grep USB > :00:1d.0 USB Controller: Intel Corp. 82801EB/ER (ICH5/ICH5R) USB > UHCI #1 (rev 02) > :00:1d.1 USB Controller: Intel Corp. 82801EB/ER (ICH5/ICH5R) USB > UHCI #2 (rev 02) > :00:1d.2 USB Controller: Intel Corp. 82801EB/ER (ICH5/ICH5R) USB > UHCI #3 (rev 02) > :00:1d.3 USB Controller: Intel Corp. 82801EB/ER (ICH5/ICH5R) USB > UHCI #4 (rev 02) > :00:1d.7 USB Controller: Intel Corp. 82801EB/ER (ICH5/ICH5R) USB2 > EHCI Controller (rev 02) > > USB and SCSI play very well together. Different components can be used > for purposes very differently from that for which they were originally > intended. > > Could you cut me a break? All your answers have been very > argumentative. I prefer to think of my responses as "probing". The question about PCI hotplug was very honest, since PCI hotplug refers to pulling PCI cards out of live systems, and that, to my understanding, has nothing to do with USB (which is hotplug by definition). Note, though, that the hotplug package "includes support for PCI, Cardbus (PCMCIA), USB and Firewire devices and can automatically configure network interfaces." -- - Ron Johnson, Jr. Jefferson, LA USA PGP Key ID 8834C06B python -c 'print len(str(2**30))' 90309 signature.asc Description: This is a digitally signed message part
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Wed, Oct 27, 2004 at 11:23:07AM -0500, Ron Johnson wrote: > > Anything promising using the PCI hotplug facility? > > Just out of curiosity, but why should *PCI* hotplug have anything > to do with USB? # lspci | grep USB :00:1d.0 USB Controller: Intel Corp. 82801EB/ER (ICH5/ICH5R) USB UHCI #1 (rev 02) :00:1d.1 USB Controller: Intel Corp. 82801EB/ER (ICH5/ICH5R) USB UHCI #2 (rev 02) :00:1d.2 USB Controller: Intel Corp. 82801EB/ER (ICH5/ICH5R) USB UHCI #3 (rev 02) :00:1d.3 USB Controller: Intel Corp. 82801EB/ER (ICH5/ICH5R) USB UHCI #4 (rev 02) :00:1d.7 USB Controller: Intel Corp. 82801EB/ER (ICH5/ICH5R) USB2 EHCI Controller (rev 02) USB and SCSI play very well together. Different components can be used for purposes very differently from that for which they were originally intended. Could you cut me a break? All your answers have been very argumentative. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Wed, Oct 27, 2004 at 06:39:30PM -0400, Peter Hsu wrote: > I think you missed the point. As Ron Johnson pointed out, you need to > use umount before you unplug the device. If you unplug without using > umount, then you should EXPECT your system to puke. For practical > purposes, you should consider time elapsed since you last wrote to the Get off your soapbox. There have been these things called "floppies" for a hell of a long time and they can be yanked without the kernel barfing. If I mark it Sync, I can yank the damn thing. Just look what this guy is doing with the usbmount script. If I yank it at the wrong time, I fully expect to lose data. I do not expect the kernel to panic. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Wed, Oct 27, 2004 at 06:39:30PM -0400, Peter Hsu wrote: > This is not a bug or quirk in Debian. It's a refusal on the part of > the operating system to compensate for failure to follow procedure. No, it's a bug in Linux (as distributed by Debian). I'm using Linux in the sense of "the kernel and closely associated stuff." It is, not acceptable, but perhaps understandable that yanking the drive could cause drive corruption and data loss. For it to freeze the system is simply silly. -- Carl Fink [EMAIL PROTECTED] Jabootu's Minister of Proofreading http://www.jabootu.com -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
I think you missed the point. As Ron Johnson pointed out, you need to use umount before you unplug the device. If you unplug without using umount, then you should EXPECT your system to puke. For practical purposes, you should consider time elapsed since you last wrote to the device irrelevant. Simply because Windows does a better job of compensating for novice users who fail to unmount a device prior to yanking doesn't mean that's proper procedure. This is not a bug or quirk in Debian. It's a refusal on the part of the operating system to compensate for failure to follow procedure. Peter On Oct 27, 2004, at 3:23 AM, William Ballard wrote: On Tue, Oct 26, 2004 at 11:45:48PM -0500, Ron Johnson wrote: ReiserFS expected it to be there, can't find it, and pukes. A bit (no, a lot) ungraceful, but totally understandable. After all, some process might be trying to write to the device... I kuh-now, I kuh-now. But that's what the SYNC option is before. If I yank it within say 20 seconds of having used it, that's on me. Windows actually deals with this better than it used to. Windows used to have serious problems if you lost a HD casually, but I can just yank it now with no ill effects ("if I haven't used it in, "a while"). Just flush a little more aggressively, that's all. Some people used autofs, it seems. Anything promising using the PCI hotplug facility? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Wed, 2004-10-27 at 03:23 -0400, William Ballard wrote: > On Tue, Oct 26, 2004 at 11:45:48PM -0500, Ron Johnson wrote: > > ReiserFS expected it to be there, can't find it, and pukes. A bit > > (no, a lot) ungraceful, but totally understandable. After all, > > some process might be trying to write to the device... > > I kuh-now, I kuh-now. But that's what the SYNC option is before. > If I yank it within say 20 seconds of having used it, that's on me. > > Windows actually deals with this better than it used to. Windows > used to have serious problems if you lost a HD casually, but I can > just yank it now with no ill effects ("if I haven't used it in, > "a while"). Just flush a little more aggressively, that's all. "sync, rw". Did you lose any data when whined about being inappropriately yanked. > Some people used autofs, it seems. There is that. > Anything promising using the PCI hotplug facility? Just out of curiosity, but why should *PCI* hotplug have anything to do with USB? -- - Ron Johnson, Jr. Jefferson, LA USA PGP Key ID 8834C06B "Go not unto the Usenet for advice, for you will be told both yea and nay (and quite a few things that just have nothing at all to do with the question)." Unknown signature.asc Description: This is a digitally signed message part
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Tue, Oct 26, 2004 at 11:45:48PM -0500, Ron Johnson wrote: > ReiserFS expected it to be there, can't find it, and pukes. A bit > (no, a lot) ungraceful, but totally understandable. After all, > some process might be trying to write to the device... I kuh-now, I kuh-now. But that's what the SYNC option is before. If I yank it within say 20 seconds of having used it, that's on me. Windows actually deals with this better than it used to. Windows used to have serious problems if you lost a HD casually, but I can just yank it now with no ill effects ("if I haven't used it in, "a while"). Just flush a little more aggressively, that's all. Some people used autofs, it seems. Anything promising using the PCI hotplug facility? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
On Tue, 2004-10-26 at 23:22 -0400, William Ballard wrote: > I have an external USB 2.0 2.5" hard drive, shows up as /dev/sda, with > an ReiserFS partition. If I yank it with it mounted ReiserJS panics. > I have the fstab entry marked as sync, rw. > > I don't have PCI Hotplugging installed, but I do a lot of USB > hotplugging. > > Has anyone had any experience with this? I'm not surprised. You're yanking out a *mounted* device. That's what the umount command is for. ReiserFS expected it to be there, can't find it, and pukes. A bit (no, a lot) ungraceful, but totally understandable. After all, some process might be trying to write to the device... -- - Ron Johnson, Jr. Jefferson, LA USA PGP Key ID 8834C06B "I prefer the most unfair peace to the most righteous war." Marcus T Cicero signature.asc Description: This is a digitally signed message part
Yanking a USB Hard Drive/ReiserFS causes Kernel Panic
I have an external USB 2.0 2.5" hard drive, shows up as /dev/sda, with an ReiserFS partition. If I yank it with it mounted ReiserJS panics. I have the fstab entry marked as sync, rw. I don't have PCI Hotplugging installed, but I do a lot of USB hotplugging. Has anyone had any experience with this? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]