Hide Forgot
Description of problem: Packstack fails when starting service neutron-ovs-cleanup. Starting the service manually and running packstack again doesn't help. Version-Release number of selected component (if applicable): openstack-packstack-2013.2.1-0.4.dev740.el6.noarch python-neutron-2013.2-0.3.b2.el6.noarch python-neutronclient-2.2.6-2.el6.noarch openstack-neutron-2013.2-0.3.b2.el6.noarch How reproducible: 100% Steps to Reproduce: 1. packstack --allinone Actual results: ERROR : Error during puppet run : err: /Stage[main]/Neutron::Agents::Ovs/Service[ovs-cleanup-service]/ensure: change from stopped to running failed: Could not start Service[ovs-cleanup-service]: Execution of '/sbin/service neutron-ovs-cleanup start' returned 1: at /var/tmp/packstack/8788692e282a40aaa66b8fee30480a69/modules/neutron/manifests/agents/ovs.pp:137 Expected results: openstack deployed Additional info: # service neutron-ovs-cleanup start # echo $? 1 # service neutron-ovs-cleanup stop # service neutron-ovs-cleanup start # echo $? 1 # service neutron-ovs-cleanup restart # echo $? 0 # service --status-all|grep neutron dnsdomainname: Unknown host 6 ACCEPT tcp -- 0.0.0.0/0 0.0.0.0/0 multiport dports 9696 /* 001 neutron incoming */ neutron-dhcp-agent dead but pid file exists neutron-l3-agent dead but pid file exists neutron-lbaas-agent is stopped neutron-metadata-agent dead but pid file exists neutron-openvswitch-agent dead but pid file exists neutron dead but pid file exists from /var/log/messages: Sep 9 12:45:47 172-24-0-4 abrt: detected unhandled Python exception in '/usr/bin/neutron-ovs-cleanup' Sep 9 12:45:47 172-24-0-4 abrtd: New client connected Sep 9 12:45:47 172-24-0-4 abrt-server[30983]: Saved Python crash dump of pid 30978 to /var/spool/abrt/pyhook-2013-09-09-12:45:47-30978 Sep 9 12:45:47 172-24-0-4 abrtd: Directory 'pyhook-2013-09-09-12:45:47-30978' creation detected Sep 9 12:45:47 172-24-0-4 abrtd: Package 'openstack-neutron-openvswitch' isn't signed with proper key Sep 9 12:45:47 172-24-0-4 abrtd: 'post-create' on '/var/spool/abrt/pyhook-2013-09-09-12:45:47-30978' exited with 1 Sep 9 12:45:47 172-24-0-4 abrtd: Corrupted or bad directory '/var/spool/abrt/pyhook-2013-09-09-12:45:47-30978', deleting Sep 9 12:46:05 172-24-0-4 abrt: detected unhandled Python exception in '/usr/bin/neutron-ovs-cleanup' Sep 9 12:46:05 172-24-0-4 abrtd: New client connected Sep 9 12:46:05 172-24-0-4 abrt-server[31010]: Not saving repeating crash in '/usr/bin/neutron-ovs-cleanup'
The image had some packages preinstalled for grizzly, but otherwise it was clean - no OpenStack installation was done on it before.
Currently we don't support upgrade via Packstack and looking at your machine I think that following is not good condition: [root@rdo4-tempest ~]# cat /etc/passwd | egrep "(quantum|neutron)" quantum:x:164:164:OpenStack Quantum Daemons:/var/lib/quantum:/sbin/nologin neutron:x:164:164:OpenStack Quantum Daemons:/var/lib/neutron:/sbin/nologin Please try on the OpenStack free machine and let me know if your issue still persists.
Yes, the problem is probably that I already had some grizzly packages installed. I knew packstack isn't able to upgrade a running openstack instance, but I thought it wouldn't have problems with the packages alone. I'll try it on a clean image ASAP.
The issue is actually in the fedora neutron packages. They do not Require: python-pbr like the el6 havana ones do. Other projects patch around the python-pbr runtime dependency since it is just used for determining the version which can be done at build time. We may do that in neutron as well, but for now I'm taking the easy way out and just rebuilding with the requires, since we already use it for the el6 branches.
Fix synced to openstack-havana/fedora-19 repo