Description of problem: After we deploy RHOSP13 with director, libvirtd, which is unnecessary in controller nodes, is still running on all controller nodes. ~~~ $ sudo systemctl status libvirtd ● libvirtd.service - Virtualization daemon Loaded: loaded (/usr/lib/systemd/system/libvirtd.service; enabled; vendor preset: enabled) Active: active (running) since Sat 2019-03-23 15:31:51 UTC; 1 day 16h ago Docs: man:libvirtd(8) https://libvirt.org Main PID: 4710 (libvirtd) Tasks: 17 (limit: 32768) Memory: 57.0M CGroup: /system.slice/libvirtd.service └─4710 /usr/sbin/libvirtd Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable. ~~~ While in compute node libvirtd is stopped when t-h-t deploys nova_libvirt container, in the other nodes there are no chance to stop it. As a result, unnecessary libvirtd is still enabled on the other nodes like controller. How reproducible: Always Steps to Reproduce: 1. Deploy overcloud 2. Log in to controller and check the status of libvirtd service Actual results: libvirtd is stopped and disabled Expected results: libvirtd is active and enabled Additional info:
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/RHBA-2019:2624
Hi Rajesh, I checked this in latest RHOSP13z8, but I'm afraid that libvirtd is still enabled on controller nodes. ~~~ [heat-admin@controller-0 ~]$ sudo systemctl status libvirtd ● libvirtd.service - Virtualization daemon Loaded: loaded (/usr/lib/systemd/system/libvirtd.service; enabled; vendor preset: enabled) Active: active (running) since Fri 2019-09-06 00:47:24 UTC; 6 days ago Docs: man:libvirtd(8) https://libvirt.org Main PID: 2292 (libvirtd) Tasks: 17 (limit: 32768) Memory: 8.0K CGroup: /system.slice/libvirtd.service └─2292 /usr/sbin/libvirtd Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable. [heat-admin@controller-0 ~]$ cat /etc/rhosp-release Red Hat OpenStack Platform release 13.0.8 (Queens) ~~~ In addition, I'm checking the upstream rdo ci, but it seems that libvirt is still started on controller nodes as I can see "Starting Virtualization daemon..." in journal example: http://logs.rdoproject.org/82/677482/1/openstack-check/tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001/157668d/logs/overcloud-controller-0/var/log/ Can I ask your look on the above situation? Thank you, Takashi
Hi Thanks for pointing out the issue, it looks like the libvirtd service is still enabled because the script added in [1] doesn't have proper permissions. I have submitted another patch [2] to have proper permission on script. Also you can see in journal output there is no message "Starting Virtualization daemon" on controller nodes [3]. [1] https://review.opendev.org/#/c/668860/ [2] https://review.opendev.org/#/c/682036/ [3] https://logs.rdoproject.org/36/682036/1/openstack-check/tripleo-ci-centos-7-ovb-3ctlr_1comp-featureset001/0511d0d/logs/overcloud-controller-0/var/log/journal.txt.gz