On Thu, 2008-09-11 at 18:13 -0400, Daniel Savard wrote:
> Le jeudi 11 septembre 2008 à 07:58 -0500, Dave Kleikamp a écrit :
> > On Wed, 2008-09-10 at 17:36 -0400, Daniel Savard wrote:
> 
> > 
> > Ouch.  I need to get gentoo to unmask a newer version.
> > 
> > Can you emerge a more recent jfsutils and see if it helps?  (I don't use
> > the livecd, but I assume you can update a package.)  That said, I'm not
> > aware of a particular bug that was fixed, but version 1.1.8 is pretty
> > old.
> > 
> 
> Not really, since I am running on the livecd and have not yet installed
> Gentoo on this server. I was about to do this. This is a reinstallation
> since this one crashed a while ago and tried to recover it many times
> running into odd problems which may related to this one.
> 
> > 
> > I'm assuming that the hardware is okay, since mkfs.jfs worked.  

I changed my mind.  This looks like a problem with the hardware or
device driver.  I bet you would have problems if you tried ext3 or some
other filesystem instead.

> If a
> > newer jfsutils doesn't work, could you run fsck.jfs under strace so I
> > can see what it's trying to do?
> > 
> > strace -o strace.out fsck.jfs /dev/rootvg/vicepa
> > 
> > Thanks,
> > Shaggy
> 
> After this error, I cannot even do a fdisk /dev/sda. I then rebooted the
> server and then fscked all the filesystems with -f option including
> the /dev/rootvg/vicepa filesystem. It passed the fsck with a correction
> to the root of the filesystem. Then, retried to do a mkfs.jfs ... as in
> previous case it paused/freezed and finally completed. Then fsck.jfs
> lead to same result. This server is having two processors (CPU). I
> checked the syslog and I found this:
> 
> Sep 11 21:52:03 livecd BUG: soft lockup - CPU#0 stuck for 11s!
> [events/0:9]
> Sep 11 21:52:03 livecd 
> Sep 11 21:52:03 livecd Pid: 9, comm: events/0 Not tainted
> (2.6.24-gentoo-r7 #1)
> Sep 11 21:52:03 livecd EIP: 0060:[<c010c6d3>] EFLAGS: 00000297 CPU: 0
> Sep 11 21:52:03 livecd EIP is at native_smp_call_function_mask
> +0x157/0x179
> Sep 11 21:52:03 livecd EAX: fffff000 EBX: f7c73f2c ECX: 00000020 EDX:
> 000c08fb
> Sep 11 21:52:03 livecd ESI: 00000001 EDI: f7c73f50 EBP: 00000001 ESP:
> f7c73f18
> Sep 11 21:52:03 livecd DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
> Sep 11 21:52:03 livecd CR0: 8005003b CR2: bffdefd0 CR3: 1fddf000 CR4:
> 000002d0
> Sep 11 21:52:03 livecd DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3:
> 00000000
> Sep 11 21:52:03 livecd DR6: ffff0ff0 DR7: 00000400
> Sep 11 21:52:03 livecd [<c0109352>] mce_checkregs+0x0/0x9a
> Sep 11 21:52:03 livecd [<c0109352>] mce_checkregs+0x0/0x9a
> Sep 11 21:52:03 livecd [<c0109352>] mce_checkregs+0x0/0x9a
> Sep 11 21:52:03 livecd [<c010ca56>] smp_call_function+0x25/0x2b
> Sep 11 21:52:03 livecd [<c011e016>] on_each_cpu+0x18/0x27
> Sep 11 21:52:03 livecd [<c0109320>] mce_work_fn+0x0/0x32
> Sep 11 21:52:03 livecd [<c010933b>] mce_work_fn+0x1b/0x32
> Sep 11 21:52:03 livecd [<c01266f9>] run_workqueue+0x74/0xf7
> Sep 11 21:52:03 livecd [<c0126e73>] worker_thread+0x0/0x85
> Sep 11 21:52:03 livecd [<c0126eec>] worker_thread+0x79/0x85
> Sep 11 21:52:03 livecd [<c0129611>] autoremove_wake_function+0x0/0x35
> Sep 11 21:52:03 livecd [<c0129548>] kthread+0x38/0x60
> Sep 11 21:52:03 livecd [<c0129510>] kthread
> +0x0/0x60                           
> Sep 11 21:52:03 livecd [<c0103877>] kernel_thread_helper+0x7/0x10
> Sep 11 21:52:03 livecd =======================

I'm not sure what to make of these.  I can't tell you whether they are
related to the I/O problems.

> Sep 11 21:52:03 livecd =======================
> Sep 11 21:52:13 livecd 99656630
> Sep 11 21:52:13 livecd sd 4:0:0:0: [sda] Result: hostbyte=DID_ERROR
> driverbyte=DRIVER_OK,SUGGEST_OK
> Sep 11 21:52:13 livecd end_request: I/O error, dev sda, sector 59657142
> Sep 11 21:52:13 livecd sd 4:0:0:0: [sda] Result: hostbyte=DID_ERROR
> driverbyte=DRIVER_OK,SUGGEST_OK
> Sep 11 21:52:13 livecd end_request: I/O error, dev sda, sector 59657270
> Sep 11 21:52:13 livecd sd 4:0:0:0: [sda] Result: hostbyte=DID_ERROR
> driverbyte=DRIVER_OK,SUGGEST_OK
> Sep 11 21:52:13 livecd end_request: I/O error, dev sda, sector 59658678
> Sep 11 21:52:13 livecd sd 4:0:0:0: [sda] Result: hostbyte=DID_ERROR
> driverbyte=DRIVER_OK,SUGGEST_OK
> Sep 11 21:52:13 livecd end_request: I/O error, dev sda, sector
> 59654966        
> Sep 11 21:52:13 livecd sd 4:0:0:0: [sda] Result: hostbyte=DID_ERROR
> driverbyte=DRIVER_OK,SUGGEST_OK
> Sep 11 21:52:13 livecd end_request: I/O error, dev sda, sector 59655222
> Sep 11 21:52:13 livecd sd 4:0:0:0: [sda] Result: hostbyte=DID_ERROR
> driverbyte=DRIVER_OK,SUGGEST_OK
> Sep 11 21:52:13 livecd end_request: I/O error, dev sda, sector 59655350
> Sep 11 21:52:13 livecd sd 4:0:0:0: [sda] Result: hostbyte=DID_ERROR
> driverbyte=DRIVER_OK,SUGGEST_OK
> Sep 11 21:52:13 livecd end_request: I/O error, dev sda, sector 59661750
> Sep 11 21:52:13 livecd sd 4:0:0:0: [sda] Result: hostbyte=DID_ERROR
> driverbyte=DRIVER_OK,SUGGEST_OK
> Sep 11 21:52:13 livecd end_request: I/O error, dev sda, sector 59656118
> Sep 11 21:52:13 livecd sd 4:0:0:0: [sda] Result: hostbyte=DID_ERROR
> driverbyte=DRIVER_OK,SUGGEST_OK
> Sep 11 21:52:13 livecd end_request: I/O error, dev sda, sector 59656502
> ....
> and so on for few pages.

This looks like a hardware problem.

> These are messages produced by the mkfs.jfs operation.
> 
> I am posting this because I am having also a Lenovo T61p laptop with
> Gentoo LVM2 and JFS. I just got the same kind of problem when trying to
> write a file in a very busy filesystem (another process was reading the
> directory entries and the table is pretty large).

Do you see the same kind of error messages in the syslog?

> I am wondering if it is something related to SMP support. The Lenovo
> T61p is a dual core.

SMP is probably more common today than UP on systems running LVM2 or
JFS, so any SMP issues should be found as early as any other problems.

-- 
David Kleikamp
IBM Linux Technology Center


-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Jfs-discussion mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jfs-discussion

Reply via email to