Re: [openstack-dev] [OpenStack-Infra] tgt restart fails in Cinder startup start: job failed to start

2014-03-12 Thread Sukhdev Kapur
Hi Roey, I made this change and have been running this fix on 4 different servers. I believe this fix works. Things are working very smoothly. I think we need to incorporate this change into devstack scripts or capture it in the documentation so that it saves some grief to the next person.

Re: [openstack-dev] [OpenStack-Infra] tgt restart fails in Cinder startup start: job failed to start

2014-03-11 Thread Roey Chen
Forwarding the answer to the relevant mailing lists: --- Hi, Hope this could help, I've encountered this issue myself not to long ago on Ubuntu 12.04 host, it didn't happen again after messing with the Kernel Semaphore Limits parameters [1]: Adding this [2] line to `/etc/sysctl.conf` seems

Re: [openstack-dev] [OpenStack-Infra] tgt restart fails in Cinder startup start: job failed to start

2014-03-10 Thread Sukhdev Kapur
I see the same issue. This issue has crept in during the latest flurry of check-ins. I started noticing this issue a day or two before the Icehouse Feature Freeze deadline. I tried restarting tgt as well, but, it does not help. However, rebooting the VM helps clear it up. Has anybody else seen

Re: [openstack-dev] [OpenStack-Infra] tgt restart fails in Cinder startup start: job failed to start

2014-03-10 Thread Sukhdev Kapur
Hey Rafael, If it helps any, I have noticed that this issue generally occurs after several runs of tempest tests. I run all smoke tests. I have been always suspicious that some new test got added which is not cleaning up afterwards. Can you point me to which tempest test corresponds to