[linux-lvm] Fwd: i/o hang while taking snapshot of a thinly provisioned LV in RHEL 6

[Rajesh Joseph]

Hi all,

We have a setup which is using thinly provisioned Logical volume. When we
try to create or delete the snapshot
we are experiencing  I/O hang to this logical volume, which has XFS
filesystem on it. Also system got stuck at
VG deactivation during reboot.

Following is the setup details:
* The thin pool size is 3.09TB, and the problematic
thinlv(rian21_projects_media2) size is 2.57TB.
* This issue started appearing after the snapshot number reached 8 , data%
of the thinlv became 90%
* The PV is made from a LUN of size 3.66TB which is coming from "IBM XIV
Storage System"
* We observed that it is taking around 10 mins to de-activate the LV while
shutting down the node and around 10 mins to activate this LV when the
machine is booting up. Is this a normal behavior?
* The lvm & dmsetup command outputs are attached .
* No storage related errors seen in /var/log/messages.

Is there any known issues with the LVM , that can lead to i/o hang during
creation or deletion of snapshot. ?

LVM package version and kernel version are below.

lvm2-2.02.118-3.el6_7.4.x86_64
2.6.32-573.12.1.el6.x86_64
xfsprogs-3.1.1-16.el6.x86_64

It would be really helpful if you can give pointers to diagnose this issue.

Thanks & Regards,
Rajesh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/linux-lvm/attachments/20160503/9c7f9424/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: lvminfo.tar
Type: application/x-tar
Size: 163840 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/linux-lvm/attachments/20160503/9c7f9424/attachment.tar>