Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*XFS\s+hung\s+on\s+2\.6\.33\.3\s+kernel\s*$/: 19 ]

Total 19 documents matching your query.

1. XFS hung on 2.6.33.3 kernel (score: 1)
Author: Ilia Mirkin <imirkin@xxxxxxxxxxxx>
Date: Sat, 17 Jul 2010 00:01:11 -0400
Hi XFS developers, I ended up getting this on a 2.6.33.3 kernel. I realize this is not the latest, but I've only gotten this once, so it will not be possible to easily determine whether this is gone
/archives/xfs/2010-07/msg00163.html (38,965 bytes)

2. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Sun, 18 Jul 2010 11:20:33 +1000
<sigh> That WARN_ON() is not coincidental at all. IŽll come back to this later. Waiting on IO completion - this is probably getting EAGAIN due to not being able to take the inode ilock.... Waiting on
/archives/xfs/2010-07/msg00164.html (20,811 bytes)

3. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Ilia Mirkin <imirkin@xxxxxxxxxxxx>
Date: Sat, 17 Jul 2010 21:35:33 -0400
Hi Dave, In case this happens again, was there something more useful I could have collected? Should I have grabbed all task states? Thanks for the analysis. I will try to switch off O_DIRECT from mys
/archives/xfs/2010-07/msg00165.html (26,375 bytes)

4. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Sun, 18 Jul 2010 00:57:02 -0400
While it's slightly crazy it's also a pretty easy way for users to shoot themselve in their feet. Unlike the generic filesystems with their simplistic i_mutex locking we have a way to assure this wor
/archives/xfs/2010-07/msg00167.html (8,414 bytes)

5. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Ilia Mirkin <imirkin@xxxxxxxxxxxx>
Date: Sun, 18 Jul 2010 01:28:43 -0400
If by "this" you mean the WARN_ON's, no problem. It should be easy to repro in a non-critical setup, although I haven't tried. If you mean the hang, it will not be so easy to reproduce, as it has onl
/archives/xfs/2010-07/msg00169.html (11,058 bytes)

6. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Ilia Mirkin <imirkin@xxxxxxxxxxxx>
Date: Sun, 18 Jul 2010 16:17:41 -0400
In case you guys are interested, I've also opened a bug at https://bugs.launchpad.net/percona-xtrabackup/+bug/606981. I never quite bothered to _really_ understand all of the details of O_DIRECT vs m
/archives/xfs/2010-07/msg00172.html (12,754 bytes)

7. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Mon, 19 Jul 2010 09:33:20 +1000
Sorry, that wasn't paticularly clear - What I was trying to say is that I'm not really interested in solving all the generic buffered/direct IO coherency issues. I agree that it should not hang, so w
/archives/xfs/2010-07/msg00184.html (9,635 bytes)

8. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Mon, 19 Jul 2010 09:50:36 +1000
All the task states, including the running tasks, is probably a good start. Also, if the kernel you are running has tracing events enabled and has the necessary XFS tracepoints (I can't remember off
/archives/xfs/2010-07/msg00185.html (10,014 bytes)

9. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Wed, 4 Aug 2010 10:47:46 +1000
Ilia, Can you send me the output of this for your kernel that the traces came from: $ gdb <path/to/vmlinux> (gdb) l *( xfs_write+0x2cc) You can run it against the vmlinux file in the kernel build dir
/archives/xfs/2010-08/msg00040.html (8,529 bytes)

10. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Ilia Mirkin <imirkin@xxxxxxxxxxxx>
Date: Tue, 3 Aug 2010 21:15:53 -0400
No problem - BTW, I'm running this on a 2.6.33.3 kernel (same as the one before, although diff hardware). If you want (and are fine with me "destroying" the current state), I can upgrade it to a kern
/archives/xfs/2010-08/msg00041.html (11,363 bytes)

11. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Wed, 4 Aug 2010 14:27:25 +1000
Make sense. Can you run 'l *(xfs_ilock+0x2c)' as well? I just need to confirm which lock it has blocked on. Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx
/archives/xfs/2010-08/msg00045.html (10,635 bytes)

12. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Ilia Mirkin <imirkin@xxxxxxxxxxxx>
Date: Wed, 4 Aug 2010 00:39:08 -0400
(gdb) l *(xfs_ilock+0x2c) 0xffffffff81221001 is in xfs_ilock (fs/xfs/linux-2.6/mrlock.h:48). 43 down_read_nested(&mrp->mr_lock, subclass); 44 } 45 46 static inline void mrupdate_nested(mrlock_t *mrp,
/archives/xfs/2010-08/msg00046.html (12,655 bytes)

13. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Wed, 4 Aug 2010 16:53:47 +1000
OK, that doesn't help - it followed into the inline function rather than telling me which of the two calls in the function it was. I guess I'll need the disassembly output to work it out. Can you sen
/archives/xfs/2010-08/msg00047.html (12,962 bytes)

14. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Ilia Mirkin <imirkin@xxxxxxxxxxxx>
Date: Wed, 4 Aug 2010 11:40:02 -0400
Looks like it's the first one. (gdb) disass xfs_ilock Dump of assembler code for function xfs_ilock: 0xffffffff81220fd5 <xfs_ilock+0>: push %rbp 0xffffffff81220fd6 <xfs_ilock+1>: test $0x1,%sil 0xfff
/archives/xfs/2010-08/msg00061.html (18,784 bytes)

15. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Thu, 5 Aug 2010 15:45:53 +1000
.... Yup, that's what it looks like to me, too. That means it's the XFS_IOLOCK_EXCL it is blocking trying to get, not the XFS_ILOCK_EXCL. Well, that makes sense, and tells me that the lockdep "held l
/archives/xfs/2010-08/msg00065.html (11,790 bytes)

16. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Thu, 5 Aug 2010 16:25:41 +1000
.... [snip test case] Nice work. I think I see the problem, and it's not something that can be easily fixed. The test-tar process is blocked in xfs_ilock_map_shared() trying to get the ip->i_lock in
/archives/xfs/2010-08/msg00066.html (14,480 bytes)

17. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Ilia Mirkin <imirkin@xxxxxxxxxxxx>
Date: Thu, 5 Aug 2010 02:42:38 -0400
Together or independently? OK. My theory [based on no evidence] was that it had to be bigger than RAM size. I'll try less though. I created the file with dd if=/dev/zero of=... bs=1M count=... How wo
/archives/xfs/2010-08/msg00067.html (11,410 bytes)

18. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Sat, 7 Aug 2010 20:08:21 +1000
independently first - then if that does't work, maybe together? For a 1GB file: $ for i in `seq 256000 1 0`; do Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx
/archives/xfs/2010-08/msg00084.html (10,527 bytes)

19. Re: XFS hung on 2.6.33.3 kernel (score: 1)
Author: qiyansun <qiyan.sun@xxxxxxxxxxxx>
Date: Thu, 4 Apr 2013 14:44:35 -0700 (PDT)
Hi, We ran into a similar situation described in this email thread on the 2.6.32.54 kernel using SLES11SP1. Are there any resolutions or patches available for this issue? Thanks, Qiyan -- View this m
/archives/xfs/2013-04/msg00096.html (8,344 bytes)


This search system is powered by Namazu