I am using LVM2 on FC4 (2.6.15-1.1831_FC4smp) and snapshots. The
problem seems to be that when a snapshot fills up it craps out. Trying
to extend it ends up hanging the lvextend process and a reboot is
needed. After which though the snapshot is corrupted beyond repair.
While hung, lvextend's stack trace looks like:
lvextend D ffff8100d1b00050 0 14786 7501 18185 (NOTLB)
ffff810105b7fc68 0000000000000086 ffff8101115afc00 ffffffff8016594c
0000000000000000 0000000000000000 ffff81011748cbe8 0000000000000000
0000000000000000 000000000001facf
Call Trace:<ffffffff8016594c>{wait_on_page_writeback_range+203}
<ffffffff8806b694>{:dm_mod:dev_suspend+0} <ffffffff801650c7>{__filemap_fdatawrite_range+87}
<ffffffff8806b694>{:dm_mod:dev_suspend+0} <ffffffff80357567>{__down_write+137}
<ffffffff88067c0f>{:dm_mod:dm_suspend+368} <ffffffff80137e56>{default_wake_function+0}
<ffffffff8035746c>{__down_read+50} <ffffffff8806b694>{:dm_mod:dev_suspend+0}
<ffffffff8806b7b0>{:dm_mod:dev_suspend+284} <ffffffff8806a68c>{:dm_mod:ctl_ioctl+536}
<ffffffff8019c84e>{do_ioctl+94} <ffffffff8019caf2>{vfs_ioctl+649}
<ffffffff8019cb66>{sys_ioctl+91} <ffffffff8010fc60>{tracesys+209}
Trying to look at the volume in this state yields:
# /usr/sbin/lvdisplay /dev/vg0/centos4.2-i386-boulder
[ I have to hit INTR here]
CTRL-c detected: giving up waiting for lock
/var/lock/lvm/V_vg0: flock failed: Interrupted system call
Can't lock vg0: skipping
Is this a known issue (in this kernel or otherwise)? Is it fixed in a
later FC kernel like kernel-2.6.15-1.1833_FC4.x86_64.rpm or
kernel-2.6.15-1.2054_FC5.x86_64.rpm?
Thots?
b.
--
My other computer is your Microsoft Windows server.
Brian J. Murrell
Attachment:
signature.asc
Description: This is a digitally signed message part
_______________________________________________ linux-lvm mailing list linux-lvm@redhat.com https://www.redhat.com/mailman/listinfo/linux-lvm read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
