6.06.02 2.6.15-51-server soft lockup on cpu#0 on shutdown/umount xfs partition on /dev/md0
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
linux (Ubuntu) |
Invalid
|
Undecided
|
Unassigned |
Bug Description
Binary package hint: linux-image-
Have /home on xfs partition on /dev/md0, RAID1 array which is shared via nfs.
Get sporadic soft lockups when trying to shutdown my 6.06.02 machine. BUG: Soft lockup detected on CPU#0!
root@jersey:
Linux jersey 2.6.15-51-server #1 SMP Thu Dec 6 21:37:18 UTC 2007 i686 GNU/Linux
root@jersey:
processor : 0
vendor_id : AuthenticAMD
cpu family : 6
model : 6
model name : AMD Athlon(tm) XP 2000+
stepping : 2
cpu MHz : 1674.468
cache size : 256 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 1
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 mmx fxsr sse syscall mmxext 3dnowext 3dnow
bogomips : 3351.78
root@jersey:
MemTotal: 1035672 kB
MemFree: 331988 kB
Buffers: 9960 kB
Cached: 621828 kB
SwapCached: 0 kB
Active: 73628 kB
Inactive: 565964 kB
HighTotal: 131008 kB
HighFree: 168 kB
LowTotal: 904664 kB
LowFree: 331820 kB
SwapTotal: 979956 kB
SwapFree: 979956 kB
Dirty: 152 kB
Writeback: 0 kB
Mapped: 14332 kB
Slab: 54516 kB
CommitLimit: 1497792 kB
Committed_AS: 20124 kB
PageTables: 480 kB
VmallocTotal: 118776 kB
VmallocUsed: 5136 kB
VmallocChunk: 113164 kB
I can sometimes reproduce if I try to umount /home from the terminal rather than letting the shutdown scripts do it for me, e.g sync; then /etc/init.
However sometimes the machine manages to shutdown on its own.
System appears responsive to Alt+SysReq+..... but doesnt always give the output expected.
Have twice gotten a backtrace out of the machine which looks like (written down from photos);
Pid: 6345, comm: umount
EIP: 0060:[<f8c67217>] CPU: 0
EIP is at xfs_iflush_
EFLAGS: 00000202 Not tainted (2.6.15-51-server)
EAX: f3fd9104 EBX: c1bf7130 ECX: 00000000 EDX: f3fd9104
ESI: c1bf7000 EDI: f79b4de8 EBP: 00000001 DS: 007b ES: 007b
CR0: 8005003b CR2: 08056400 CR3: 1fec72c0 CR4: 000006b0
[<f8c78414>] cfs_umountfs+
[<f8c9573b>] linvfs_
[<f8c7ff22>] xfs_umount+
[<f8c95e6b>] linvfs_
[<c0179a63>] generic_
[<c017a69e>] kill_block_
[<c017991a>] deactivate_
[<c019253f>] sys_umount+
[<c01925b5>] sys_oldumount+
[<c0103313>] sysenter_
Not sure if this is because I tried to umount the fs to help the machine shutdown with Alt+SysReq+U.
Once I managed to get a list of running tasks which showed that only;
xfsbufd
rc
S40umountfs
S40umountfs
umount
were running.
Any more information I can supply to help?
Have been having some success lately with;
sync d/nfs-kernel- server stop
/etc/init.
sync
umount /home
halt
I dont recall getting the crash when shutting down this way, the crucial element (I suspect) being the sync between stopping nfs and unmounting /home