Re: Disk/FS I/O issues in -CURRENT
David O'Brien wrote: > > On Mon, Jun 30, 2003 at 03:30:06PM -0500, Alan L. Cox wrote: > > I've been able to reproduce what I believe is the problem. (In my case, > > I reset my machine and watched the background fsck slowly grind to a > > halt. Foreground fsck is fine.) > > > > The problem actually appears to be in vm_page_alloc(), not > > vm_pageout.c. Look for a commit to resolve this in a few hours. > > Has this been committed. I *so* ran into this yesterday (3 times) with a > 70gb /usr and a June 30th kernel. > Yes. Alan ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Disk/FS I/O issues in -CURRENT
On Mon, Jun 30, 2003 at 03:30:06PM -0500, Alan L. Cox wrote: > I've been able to reproduce what I believe is the problem. (In my case, > I reset my machine and watched the background fsck slowly grind to a > halt. Foreground fsck is fine.) > > The problem actually appears to be in vm_page_alloc(), not > vm_pageout.c. Look for a commit to resolve this in a few hours. Has this been committed. I *so* ran into this yesterday (3 times) with a 70gb /usr and a June 30th kernel. -- -- David ([EMAIL PROTECTED]) ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Disk/FS I/O issues in -CURRENT
Thanks. I'll be keeping my eyes wide open for this one. /Eirik On Mon, 30 Jun 2003 15:30:06 -0500 "Alan L. Cox" <[EMAIL PROTECTED]> wrote: > Peter Holm wrote: > > > > On Mon, Jun 30, 2003 at 03:26:13PM +0200, Eirik Oeverby wrote: > > > Hi, > > > > > > Good to see I'm not the only one. > > > I'm currently going back to a kernel dated 2003.06.27.12.00.00, > > > and I'll test again with that one. > > > > > > > Ok. > > > > I see that alc@ made some recent changes to the vm (vm_pageout.c). > > I don't know if there's any connection to this problem? > > > > I've been able to reproduce what I believe is the problem. (In my > case, I reset my machine and watched the background fsck slowly grind > to a halt. Foreground fsck is fine.) > > The problem actually appears to be in vm_page_alloc(), not > vm_pageout.c. Look for a commit to resolve this in a few hours. > > Regards, > Alan > ___ > [EMAIL PROTECTED] mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-current > To unsubscribe, send any mail to > "[EMAIL PROTECTED]" pgp0.pgp Description: PGP signature
Re: Disk/FS I/O issues in -CURRENT
Peter Holm wrote: > > On Mon, Jun 30, 2003 at 03:26:13PM +0200, Eirik Oeverby wrote: > > Hi, > > > > Good to see I'm not the only one. > > I'm currently going back to a kernel dated 2003.06.27.12.00.00, and I'll test > > again with that one. > > > > Ok. > > I see that alc@ made some recent changes to the vm (vm_pageout.c). > I don't know if there's any connection to this problem? > I've been able to reproduce what I believe is the problem. (In my case, I reset my machine and watched the background fsck slowly grind to a halt. Foreground fsck is fine.) The problem actually appears to be in vm_page_alloc(), not vm_pageout.c. Look for a commit to resolve this in a few hours. Regards, Alan ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Disk/FS I/O issues in -CURRENT
Hi, I've found a kernel that works, from the 20th of June. The regression happened somewhere between 2003.06.20.12.00.00 and 2003.06.27.12.00.00 ... Hope that helps. :) /Eirik On Mon, 30 Jun 2003 10:27:08 -0400 Josh Elsasser <[EMAIL PROTECTED]> wrote: > I think I experienced the same bug on my Sony Vaio FX200 with -CURRENT > from Sat Jun 28. I had an unrelated panic, and after rebooting, the > machine locked up after a minute or so during the background fsck. > After rebooting several times, I finally had to boot it single-user > and fsck -y, which did not lock it up. Perhaps creating/using the > filesystem snapshots was triggering the lockup. > > -jre > > On Mon, Jun 30, 2003 at 03:42:58PM +0200, Eirik Oeverby wrote: > > Hi, > > > > Kernel from 27.06 has same behaviour. I would prefer not to have to > > install yet another kernel right now, since I need to get some work > > done. If anyone else has any possible clues as to when this > > regression happened, that would help me (or whoever else would want > > to test by adding date=.mm.dd.hh.mm.ss to their supfile) > > determine what date to pick for testing. > > > > /Eirik > > > > On Mon, 30 Jun 2003 23:32:26 +1000 > > Mark Sergeant <[EMAIL PROTECTED]> wrote: > > > > > I get the same problem on a smp machine and my laptop both running > > > kernels as from today. > > > > > > On Mon, Jun 30, 2003 at 03:26:13PM +0200, Eirik Oeverby wrote: > > > > Hi, > > > > > > > > Good to see I'm not the only one. > > > > I'm currently going back to a kernel dated 2003.06.27.12.00.00, > > > > and I'll test again with that one. > > > > > > > > /Eirik > > > > > > > > On Mon, 30 Jun 2003, Peter Holm wrote: > > > ___ > > > [EMAIL PROTECTED] mailing list > > > http://lists.freebsd.org/mailman/listinfo/freebsd-current > > > To unsubscribe, send any mail to > > > "[EMAIL PROTECTED]" > > > > > ___ > [EMAIL PROTECTED] mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-current > To unsubscribe, send any mail to > "[EMAIL PROTECTED]" pgp0.pgp Description: PGP signature
Re: Disk/FS I/O issues in -CURRENT
I think I experienced the same bug on my Sony Vaio FX200 with -CURRENT from Sat Jun 28. I had an unrelated panic, and after rebooting, the machine locked up after a minute or so during the background fsck. After rebooting several times, I finally had to boot it single-user and fsck -y, which did not lock it up. Perhaps creating/using the filesystem snapshots was triggering the lockup. -jre On Mon, Jun 30, 2003 at 03:42:58PM +0200, Eirik Oeverby wrote: > Hi, > > Kernel from 27.06 has same behaviour. I would prefer not to have to > install yet another kernel right now, since I need to get some work > done. If anyone else has any possible clues as to when this regression > happened, that would help me (or whoever else would want to test by > adding date=.mm.dd.hh.mm.ss to their supfile) determine what date to > pick for testing. > > /Eirik > > On Mon, 30 Jun 2003 23:32:26 +1000 > Mark Sergeant <[EMAIL PROTECTED]> wrote: > > > I get the same problem on a smp machine and my laptop both running > > kernels as from today. > > > > On Mon, Jun 30, 2003 at 03:26:13PM +0200, Eirik Oeverby wrote: > > > Hi, > > > > > > Good to see I'm not the only one. > > > I'm currently going back to a kernel dated 2003.06.27.12.00.00, and > > > I'll test again with that one. > > > > > > /Eirik > > > > > > On Mon, 30 Jun 2003, Peter Holm wrote: > > ___ > > [EMAIL PROTECTED] mailing list > > http://lists.freebsd.org/mailman/listinfo/freebsd-current > > To unsubscribe, send any mail to > > "[EMAIL PROTECTED]" > > ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Disk/FS I/O issues in -CURRENT
Hi, Kernel from 27.06 has same behaviour. I would prefer not to have to install yet another kernel right now, since I need to get some work done. If anyone else has any possible clues as to when this regression happened, that would help me (or whoever else would want to test by adding date=.mm.dd.hh.mm.ss to their supfile) determine what date to pick for testing. /Eirik On Mon, 30 Jun 2003 23:32:26 +1000 Mark Sergeant <[EMAIL PROTECTED]> wrote: > I get the same problem on a smp machine and my laptop both running > kernels as from today. > > On Mon, Jun 30, 2003 at 03:26:13PM +0200, Eirik Oeverby wrote: > > Hi, > > > > Good to see I'm not the only one. > > I'm currently going back to a kernel dated 2003.06.27.12.00.00, and > > I'll test again with that one. > > > > /Eirik > > > > On Mon, 30 Jun 2003, Peter Holm wrote: > ___ > [EMAIL PROTECTED] mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-current > To unsubscribe, send any mail to > "[EMAIL PROTECTED]" pgp0.pgp Description: PGP signature
Re: Disk/FS I/O issues in -CURRENT
I get the same problem on a smp machine and my laptop both running kernels as from today. On Mon, Jun 30, 2003 at 03:26:13PM +0200, Eirik Oeverby wrote: > Hi, > > Good to see I'm not the only one. > I'm currently going back to a kernel dated 2003.06.27.12.00.00, and I'll test > again with that one. > > /Eirik > > On Mon, 30 Jun 2003, Peter Holm wrote: ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Disk/FS I/O issues in -CURRENT
Hi, Good to see I'm not the only one. I'm currently going back to a kernel dated 2003.06.27.12.00.00, and I'll test again with that one. /Eirik On Mon, 30 Jun 2003, Peter Holm wrote: > On Mon, Jun 30, 2003 at 01:42:15PM +0200, Eirik Oeverby wrote: > > Hi folks, > > > > I am having some very weird problems on my laptop, > > I can repeat the problem (noticed with savecore) on a > kernel from Jun 30 05:23 UTC: > > current# df -h . > FilesystemSize Used Avail Capacity Mounted on > /dev/ad0s1f 8.2G 1.9G 5.6G25%/usr > current# dd if=/dev/zero of=100mb bs=1024 count=102400 > load: 4.04 cmd: dd 25063 [running] 0.33u 28.67s 4% 100k > 97657+0 records in > 97657+0 records out > 10768 bytes transferred in 48.549837 secs (2059755 bytes/sec) > > db> ps > pid proc addruid ppid pgrp flag stat wmesgwchan cmd > 25063 c1e5d3c8 cd19d0000 25060 25063 0004002 [RUNQ] dd > 8 c197ed3c ccc9e0000 0 0 204 [CPU 0] pagedaemon > > db> t 8 > siointr1(c0b6c800,0,c051fcc5,693,c867abf0) at siointr1+0xd5 > siointr(c0b6c800) at siointr+0x35 > Xfastintr4() at Xfastintr4+0x63 > --- interrupt, eip = 0xc0377480, esp = 0xc867abdc, ebp = 0xc867abf0 --- > strncmp(c051e785,c051df68,123,0,477b) at strncmp > witness_unlock(c059c280,8,c051e785,35c,1) at witness_unlock+0x5a > _mtx_unlock_flags(c059c280,0,c051e785,35c,c0ba69ec) at _mtx_unlock_flags+0x80 > vm_pageout_scan(0,0,c051e785,5dd,1f4) at vm_pageout_scan+0x40c > vm_pageout(0,c867ad48,c0505bfc,312,0) at vm_pageout+0x2ce > fork_exit(c04688d0,0,c867ad48) at fork_exit+0xc0 > fork_trampoline() at fork_trampoline+0x1a > --- trap 0x1, eip = 0, esp = 0xc867ad7c, ebp = 0 --- > db> > > ___ > [EMAIL PROTECTED] mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-current > To unsubscribe, send any mail to "[EMAIL PROTECTED]" > ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Re: Disk/FS I/O issues in -CURRENT
On Mon, Jun 30, 2003 at 01:42:15PM +0200, Eirik Oeverby wrote: > Hi folks, > > I am having some very weird problems on my laptop, I can repeat the problem (noticed with savecore) on a kernel from Jun 30 05:23 UTC: current# df -h . FilesystemSize Used Avail Capacity Mounted on /dev/ad0s1f 8.2G 1.9G 5.6G25%/usr current# dd if=/dev/zero of=100mb bs=1024 count=102400 load: 4.04 cmd: dd 25063 [running] 0.33u 28.67s 4% 100k 97657+0 records in 97657+0 records out 10768 bytes transferred in 48.549837 secs (2059755 bytes/sec) db> ps pid proc addruid ppid pgrp flag stat wmesgwchan cmd 25063 c1e5d3c8 cd19d0000 25060 25063 0004002 [RUNQ] dd 8 c197ed3c ccc9e0000 0 0 204 [CPU 0] pagedaemon db> t 8 siointr1(c0b6c800,0,c051fcc5,693,c867abf0) at siointr1+0xd5 siointr(c0b6c800) at siointr+0x35 Xfastintr4() at Xfastintr4+0x63 --- interrupt, eip = 0xc0377480, esp = 0xc867abdc, ebp = 0xc867abf0 --- strncmp(c051e785,c051df68,123,0,477b) at strncmp witness_unlock(c059c280,8,c051e785,35c,1) at witness_unlock+0x5a _mtx_unlock_flags(c059c280,0,c051e785,35c,c0ba69ec) at _mtx_unlock_flags+0x80 vm_pageout_scan(0,0,c051e785,5dd,1f4) at vm_pageout_scan+0x40c vm_pageout(0,c867ad48,c0505bfc,312,0) at vm_pageout+0x2ce fork_exit(c04688d0,0,c867ad48) at fork_exit+0xc0 fork_trampoline() at fork_trampoline+0x1a --- trap 0x1, eip = 0, esp = 0xc867ad7c, ebp = 0 --- db> ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "[EMAIL PROTECTED]"
Disk/FS I/O issues in -CURRENT
Hi folks, I am having some very weird problems on my laptop, which I at first thought were due to a failing drive. However after a lot of testing and trial and error I've concluded that it must be a software problem - I'm smelling filesystem issues or something. Whenever I transfer large files to or from my harddrive, I will *always* get a lockup after ~480 MB transferred. Example scenario: - I use another machine to FTP to my laptop. - I start downloading a file of about 700 MB, and after ~480 MB it simply stops. The HD led on my laptop goes off, I can switch VTs but nothing else. Even keyboard input is non-functional. - After a few seconds another small burst of data (perhaps a few hundred KB or something) will go through, keyboard buffers will be processed, and then it freezes again with only one option - power off. C-A-D is a no-go. The IP stack is up though, I can ping the machine, and I can still change VTs. - If I pause/cancel the download immediately when it freezes, my laptop will come back to life a few seconds later. I can then start downloading another file, or resume the file I had started - everything seems to be back to normal. But whenever I hit ~480 MB transferred from *one* file, it will freeze again. - Downloading many smaller files in one batch, that sum up to >500 MB, is not a problem. All the above points applies to data transfer in the opposite direction aswell. Another way to reproduce this is doing a 'dd if=/dev/random of=testfile bs=8192', or the other way round - eventually (and after between 450 and 500 MB - a bit hard to tell) it will freeze. However, if I use a partition as the output instead of a file, it will *NOT* encounter any such problems. Transferring files between two partitions on the same HD also seems to work fine - atleast when using Midnight Commander. I first encountered the problem when using unison to sync my /home with another machine, and it got a hickup when it started finding these large files (ISOs and the like) and was scanning them to produce a checksum. Once I moved the large (>450 MB) files out of the way, it worked nicely. When I started copying the files over manually, I experienced the freezes again. Moving the files to another partition before copying it to the external machine did not help the issue. I have ran the Drive Fitness Test from IBM (this is a ThinkPad T21) twice, and it finds no errors whatsoever. And I know for a fact that this problem did *not* occur with 5.1-RELEASE, as it was with that version I copied the files to my disk in the first place. I hope this information makes sense to someone. I suppose this is the punishment for following -CURRENT - but unfortunately 5.1-RELEASE had other issues that I couldn't live with. Best regards, /Eirik pgp0.pgp Description: PGP signature