Open Stack

Hi,

I have multinode Openstack Icehouse setup. I am trying to attach a volume
of 100GB through cinder service. But I am not able to do that and it is
failing.

When I checked "/var/log/messages" I got the following error while
attaching the volume from the instance.

Nov 11 07:22:34 network-Blade3 sudo:   cinder : TTY=unknown ; PWD=/ ;
USER=root ; COMMAND=/usr/bin/cinder-rootwrap /etc/cinder/rootwrap.conf
tgt-admin --force --delete
iqn.2010-10.org.openstack:volume-ed123cfc-17f8-4fc6-bad4-24aea78407d2
Nov 11 07:22:34 network-Blade3 sudo:   cinder : TTY=unknown ; PWD=/ ;
USER=root ; COMMAND=/usr/bin/cinder-rootwrap /etc/cinder/rootwrap.conf
tgt-admin --show
*Nov 11 07:22:34 network-Blade3 kernel: [945233.808908] tgtd[14087]:
segfault at 0 ip 00007fbb0b8e5b46 sp 00007fffb2591f48 error 4 in
libc-2.11.3.so <http://libc-2.11.3.so>[7fbb0b7d3000+16e000]*

As far as I have understood, Cinder uses tgtd service as iscsi back end to
expose the volumes to the VMs running on compute nodes. There will be a
target created by tgtd for every volume created by cinder. Compute nodes
will start an iscsi session to this target while attaching the volume and
also it closes the session while detaching.

When I further debug this issue by attaching tgtd service to gdb. Segmentation
fault is raised in following method:

Program received signal SIGSEGV, Segmentation fault.
*0x00007f082caa5b46 in __strcmp_sse42 () from /lib64/libc.so.6*

Can anyone tell me how can I fix this?

BR,
Varun
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20141113/cb5ca4a8/attachment.html>