Description of problem: Version-Release number of selected component (if applicable): openstack-swift-1.7.4-9.el6ost.noarch How reproducible: Always Steps to Reproduce: 1. 2. 3. Actual results: for svc in /etc/init.d/openstack*; do $svc status; done openstack-nova-compute (pid 5899) is running... account-server (pid 3912) is running... account-auditor (pid 5166) is running... account-reaper (pid 5172) is running... account-replicator (pid 3930) is running... container-server (pid 3839) is running... container-auditor (pid 5186) is running... container-replicator (pid 3774) is running... container-updater dead but pid file exists object-server (pid 4007) is running... object-auditor dead but pid file exists object-replicator (pid 4025) is running... object-updater dead but pid file exists Expected results: for svc in /etc/init.d/openstack*; do $svc status; done openstack-nova-compute (pid 5899) is running... openstack-swift-account-server (pid 3912) is running... openstack-swift-account-auditor (pid 5166) is running... openstack-swift-account-reaper (pid 5172) is running... openstack-swift-account-replicator (pid 3930) is running... openstack-swift-container-server (pid 3839) is running... ... Additional info:
Wait a moment, do you want openstack-swift-account return openstack-swift-account or openstack-swift-account-server?
Pete, on the face of it this sounds like an easy fix?
Code-wise it's not hard. But it's unpleasant. Currently out scripts punt status to swift-init, which comes from upstream and has no idea that we renamed all the services. So, I see two possible approaches: 1. Patch swift-init to print openstack-blah to match. I do not see upstream agreeing to this, so it's going to be a patch to carry forever. 2. Reimplement status checking in scripts. It's just reading some PID files and issuing kill -0. It's not entirely future-proof. So frankly I was going to ask Jaroslav if we could just WONTFIX this, unless this is needed for some kind of systemd thingie.
(In reply to comment #3) > Code-wise it's not hard. But it's unpleasant. > > Currently out scripts punt status to swift-init, which comes from upstream > and has no idea that we renamed all the services. So, I see two possible > approaches: > > 1. Patch swift-init to print openstack-blah to match. I do not see > upstream agreeing to this, so it's going to be a patch to carry forever. > > 2. Reimplement status checking in scripts. It's just reading some PID > files and issuing kill -0. It's not entirely future-proof. > > So frankly I was going to ask Jaroslav if we could just WONTFIX this, > unless this is needed for some kind of systemd thingie. Ok, doesn't seem worth the hassle at this point. If we can try upstream to rename, great. If not, I think we can close wontfix.
Created attachment 732960 [details] Candidate 1 Okay, I was utterly wrong about the way our scripts worked. To make little amends, here's the fix for the original problem. Note, however, that the bug report expressed a desire to see status for "openstack-swift-object-server", but this just makes no sense: the executable is called "swift-object-server", while service is "openstack-swift-object". Also, $(basename $0) is not used on purpose: we have the name in meta- comments already, so it's copypasted from there, to ensure once again.
I think I committed it... The branches confuse me. The "git push" said: Total 18 (delta 17), reused 0 (delta 0) remote: *** No rules for rh-grizzly-rhel-6. Happy hacking! To ssh://zaitcev.redhat.com/rpms/openstack-swift 835556a..3ea78a1 rh-grizzly-rhel-6 -> rh-grizzly-rhel-6 Okay? Modified in openstack-swift-1.4.7-12.
Working. for svc in /etc/init.d/openstack-swift*; do echo $svc; $svc status; echo; done /etc/init.d/openstack-swift-account openstack-swift-account (pid 30753) is running... /etc/init.d/openstack-swift-account-auditor openstack-swift-account-auditor (pid 30782) is running... /etc/init.d/openstack-swift-account-reaper openstack-swift-account-reaper (pid 30810) is running... /etc/init.d/openstack-swift-account-replicator openstack-swift-account-replicator (pid 30838) is running... /etc/init.d/openstack-swift-container openstack-swift-container (pid 31036) is running... /etc/init.d/openstack-swift-container-auditor openstack-swift-container-auditor (pid 31065) is running... /etc/init.d/openstack-swift-container-replicator openstack-swift-container-replicator (pid 31093) is running... /etc/init.d/openstack-swift-container-updater openstack-swift-container-updater (pid 31008) is running... /etc/init.d/openstack-swift-object openstack-swift-object (pid 30937) is running... /etc/init.d/openstack-swift-object-auditor openstack-swift-object-auditor (pid 30880) is running... /etc/init.d/openstack-swift-object-replicator openstack-swift-object-replicator (pid 30909) is running... /etc/init.d/openstack-swift-object-updater openstack-swift-object-updater (pid 30966) is running...
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. http://rhn.redhat.com/errata/RHBA-2013-0878.html