Description of problem: ----------------------- After upgrade to RHOS-11 nova-api fails to start: 2017-01-11 13:56:15.043 521505 CRITICAL nova [req-24c8a525-7062-4aa1-8eef-fb0876106bba - - - - -] ConfigFileValueError: Value for option osapi_compute_workers is not valid: Should be greater than or equal to 1 2017-01-11 13:56:15.043 521505 ERROR nova Traceback (most recent call last): 2017-01-11 13:56:15.043 521505 ERROR nova File "/usr/bin/nova-api", line 10, in <module> 2017-01-11 13:56:15.043 521505 ERROR nova sys.exit(main()) 2017-01-11 13:56:15.043 521505 ERROR nova File "/usr/lib/python2.7/site-packages/nova/cmd/api.py", line 59, in main 2017-01-11 13:56:15.043 521505 ERROR nova server = service.WSGIService(api, use_ssl=should_use_ssl) 2017-01-11 13:56:15.043 521505 ERROR nova File "/usr/lib/python2.7/site-packages/nova/service.py", line 303, in __init__ 2017-01-11 13:56:15.043 521505 ERROR nova self.workers = (getattr(CONF, '%s_workers' % wname, None) or 2017-01-11 13:56:15.043 521505 ERROR nova File "/usr/lib/python2.7/site-packages/oslo_config/cfg.py", line 2355, in __getattr__ 2017-01-11 13:56:15.043 521505 ERROR nova return self._get(name) 2017-01-11 13:56:15.043 521505 ERROR nova File "/usr/lib/python2.7/site-packages/oslo_config/cfg.py", line 2774, in _get 2017-01-11 13:56:15.043 521505 ERROR nova value = self._do_get(name, group, namespace) 2017-01-11 13:56:15.043 521505 ERROR nova File "/usr/lib/python2.7/site-packages/oslo_config/cfg.py", line 2817, in _do_get 2017-01-11 13:56:15.043 521505 ERROR nova % (opt.name, str(ve))) 2017-01-11 13:56:15.043 521505 ERROR nova ConfigFileValueError: Value for option osapi_compute_workers is not valid: Should be greater than or equal to 1 2017-01-11 13:56:15.043 521505 ERROR nova Version-Release number of selected component (if applicable): ------------------------------------------------------------- openstack-nova-api-15.0.0-0.20170106144910.45f14c2.el7ost.noarch openstack-nova-conductor-15.0.0-0.20170106144910.45f14c2.el7ost.noarch python-novaclient-6.0.0-1.el7ost.noarch openstack-nova-console-15.0.0-0.20170106144910.45f14c2.el7ost.noarch openstack-nova-common-15.0.0-0.20170106144910.45f14c2.el7ost.noarch openstack-nova-compute-15.0.0-0.20170106144910.45f14c2.el7ost.noarch openstack-nova-novncproxy-15.0.0-0.20170106144910.45f14c2.el7ost.noarch python-nova-15.0.0-0.20170106144910.45f14c2.el7ost.noarch openstack-nova-cert-15.0.0-0.20170106144910.45f14c2.el7ost.noarch openstack-nova-scheduler-15.0.0-0.20170106144910.45f14c2.el7ost.noarch puppet-nova-10.1.0-0.20170105233844.4cb2476.el7ost.noarch Steps to Reproduce: 1. Upgrade from RHOS-9 to RHOS-10 2. Upgrade from RHOS-10 to RHOS-11 3. Check status of openstack-nova-api on overcloud controllers Actual results: --------------- Service fails to start Expected results: ----------------- Service is configured during upgrade
Created attachment 1242666 [details] nova.conf I'm attaching nova.conf from an environment which shows the same issue. Note: I wasn't able to complete the Nova upgrade because it's still WIP - https://review.openstack.org/#/c/405241/ . Maybe we need to change these config options to the new required values during the upgrade steps?
Reassigning to Steven Hardy, as I suspect this will be addressed by the following WIP (see review comments where this bug is mentioned): - Add nova service support for composable upgrades - https://review.openstack.org/#/c/405241/ Related nova conf change: - Backwards incompatible change introduced here: - https://review.openstack.org/#/c/332861/13/nova/conf/service.py Related tripleo bugs: - Can't deploy overcloud with ocata/trunk on rhel - https://bugs.launchpad.net/tripleo/+bug/1633735 - nova's osapi_compute_workers is not valid (must be greater than 1) - https://bugs.launchpad.net/tripleo/+bug/1631133 Thanks Steven! Do let me know if want/need any help on this front. Have a great weekend!
PS. See also: - Nova misconfigured after upgrade to RHOS-11: ConfigFileValueError: Value for option metadata_workers is not valid - https://bugzilla.redhat.com/show_bug.cgi?id=1412203
Hi several patch landed for nova and have landed downstream so I think this bug can be closed, Yurii can you confirm ?
Sorry, please change "several patch(es) landed for nova" to "the patch landed upstream"
I think we can consider this as solved as currently the correct configuration gets applied by puppet during the upgrade workflow.
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. https://access.redhat.com/errata/RHEA-2017:1245