Hello List,
I am using LVM2 v2.02.180 in a clvmd based cluster (three nodes), but sometimes I encountered lvm2 command hang problem.
When the comand is hanged, all lvm2 related commands from each node will be hanged.
For example,
First command was hanged at node 2,
sle12sp4b2-nd2:/ # pvmove -i 5 -v /dev/vdb /dev/vdc
Archiving volume group "cluster-vg2" metadata (seqno 34).
Creating logical volume pvmove1
Moving 2560 extents of logical volume cluster-vg2/test-lv.
sle12sp4b2-nd2:/ # cat /proc/15074/stack
[<ffffffffb4662f4b>] unix_stream_read_generic+0x66b/0x870
[<ffffffffb4663215>] unix_stream_recvmsg+0x45/0x50
[<ffffffffb459ad66>] sock_read_iter+0x86/0xd0
[<ffffffffb4240239>] __vfs_read+0xd9/0x140
[<ffffffffb4240fe7>] vfs_read+0x87/0x130
[<ffffffffb42424a2>] SyS_read+0x42/0x90
[<ffffffffb4003924>] do_syscall_64+0x74/0x150
[<ffffffffb480009a>] entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[<ffffffffffffffff>] 0xffffffffffffffff
Then, I run "lvs command on each node, the command was hanged, like
sle12sp4b2-nd2:/ # lvs
sle12sp4b2-nd2:/ # cat /proc/15553/stack
[<ffffffffc059b638>] dm_consult_userspace+0x1e8/0x490 [dm_log_userspace]
[<ffffffffc059a273>] userspace_do_request.isra.3+0x53/0x140 [dm_log_userspace]
[<ffffffffc059a8f7>] userspace_status+0xa7/0x1c0 [dm_log_userspace]
[<ffffffffc0444259>] mirror_status+0x1a9/0x370 [dm_mirror]
[<ffffffffc020751d>] retrieve_status+0xad/0x1c0 [dm_mod]
[<ffffffffc0208561>] table_status+0x51/0x80 [dm_mod]
[<ffffffffc0208258>] ctl_ioctl+0x1d8/0x450 [dm_mod]
[<ffffffffc02084da>] dm_ctl_ioctl+0xa/0x10 [dm_mod]
[<ffffffffb4256c92>] do_vfs_ioctl+0x92/0x5e0
[<ffffffffb4257254>] SyS_ioctl+0x74/0x80
[<ffffffffb4003924>] do_syscall_64+0x74/0x150
[<ffffffffb480009a>] entry_SYSCALL_64_after_hwframe+0x3d/0xa2
[<ffffffffffffffff>] 0xffffffffffffffff
Since I am trying to upgrade LVM2 from vv2.02.120 to v2.02.180,
I do not know what is the real cause to this problem?
maybe it is related to /etc/lvm/lvm.conf file? since the item raid_region_size = 2048 in the new conf, but in the past conf raid_region_size = 512,
or any other configuration items?
To fix this problem, I have to make the configuration file is consistent on each node, and reboot all the nodes.
Thanks
Gang
_______________________________________________
linux-lvm mailing list
linux-lvm@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/linux-lvm
read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/
