overcloud: HA: httpd service should not be started on boot by systemd. [root@overcloud-controller-0 heat-admin]# systemctl status httpd httpd.service - Cluster Controlled httpd Loaded: loaded (/usr/lib/systemd/system/httpd.service; enabled)
i just went through rebase hell to make this a downstream change too but after running as sanity check i get this trace [0] on controller. I need to revisit to find out why. for tracking: upstream @ https://review.openstack.org/#/c/206463/ downstream @ https://code.engineering.redhat.com/gerrit/#/c/56898 [0] 6152:Sep 02 07:32:24 overcloud-controller-0.localdomain os-collect-config[1787]: Error: Could not find resource 'Service[httpd]' for relationship from 'File[/etc/openstack-dashboard/local_settings]' on node overcloud-controller-0.localdomain 6153:Sep 02 07:32:24 overcloud-controller-0.localdomain os-collect-config[1787]: Error: Could not find resource 'Service[httpd]' for relationship from 'File[/etc/openstack-dashboard/local_settings]' on node overcloud-controller-0.localdomain 6154:Sep 02 07:32:24 overcloud-controller-0.localdomain os-collect-config[1787]: [2015-09-02 07:32:24,138] (heat-config) [ERROR] Error running /var/lib/heat-config/heat-config-puppet/4b0ad5be-1220-4ae4-9e2d-0489e1f12974.pp. [1]
I spent a while poking at this today. I couldn't get a clean run when I tried to tell the apache module not to manage the service (and leave it up to the existing pacemaker service declaration for httpd later on in the overcloud_controller_pacemaker.pp to start it), i.e. what the current upstream review does. I consistently hit the error in comment 4 above. For that particular error, I traced it down to [1][2]. I think the expectation is that the horizon module will manage apache (via horizon::wsgi::apache). For now best I can offer is to make the service be disabled for system boot (which from the bug title, seems to be the actual request). I also moved the declaration of the pacemaker managed httpd service to much earlier. Updated upstream review @ https://review.openstack.org/#/c/206463/ That at least runs clean for me and ends up with httpd being disabled for boot startup: [root@overcloud-controller-0 heat-admin]# systemctl status httpd httpd.service - Cluster Controlled httpd Loaded: loaded (/usr/lib/systemd/system/httpd.service; disabled) The dashboard also seems fine but didn't poke too much, more testing appreciated: curl -v http://192.0.2.19:80/dashboard < HTTP/1.1 302 FOUND < Date: Fri, 04 Sep 2015 16:23:40 GMT < Server: Apache/2.4.6 (Red Hat Enterprise Linux) < Vary: Accept-Language,Cookie < X-Frame-Options: SAMEORIGIN < Content-Language: en < Location: http://192.0.2.19/dashboard/auth/login/?next=/dashboard/ < Content-Length: 0 < Connection: close < Content-Type: text/html; charset=utf-8 < * Closing connection 0 [1] https://github.com/openstack/puppet-horizon/blob/master/manifests/wsgi/apache.pp#L107 [2]https://github.com/openstack/puppet-horizon/blob/master/manifests/wsgi/apache.pp#L151
*** Bug 1266308 has been marked as a duplicate of this bug. ***
This has a patch downstream and the upstream version merged already; shouldn't we just include this in 7?
openstack-tripleo-heat-templates-0.8.7-12.el7ost.noarch httpd not started on boot.
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://rhn.redhat.com/errata/RHEA-2016-0604.html