Bug 1001304
| Summary: | Remove obsolete ControlGroup option from beah service file | ||
|---|---|---|---|
| Product: | [Retired] Beaker | Reporter: | Bill Peck <bpeck> |
| Component: | beah | Assignee: | Amit Saha <asaha> |
| Status: | CLOSED CURRENTRELEASE | QA Contact: | tools-bugs <tools-bugs> |
| Severity: | unspecified | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 0.14 | CC: | aigao, asaha, atodorov, bpeck, dcallagh, ebaak, kparal, llim, lpoetter, qwan, rmancy, xjia |
| Target Milestone: | 0.14.4 | Keywords: | TestBlocker |
| Target Release: | --- | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | Bug Fix | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2013-12-19 05:09:09 UTC | Type: | Bug |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
|
Description
Bill Peck
2013-08-26 20:54:50 UTC
This sounds like it needs to be filed as a bug against systemd, not against Beaker. Silently breaking every unit file that contains a CGroup definition instead of emitting a warning that the definition is being ignored is extraordinarily user hostile. Then again...
Bill, Amit tried starting beah on the next-to-latest rawhide systemd (systemd-206-6.fc20.x86_64) and couldn't reproduce the report failure. It appeared to start fine for him:
=========================================
# systemctl status beah-srv
beah-srv.service - The Beaker Harness server.
Loaded: loaded (/usr/lib/systemd/system/beah-srv.service; disabled)
Active: active (running) since Mon 2013-08-26 21:21:25 MDT; 7s ago
Main PID: 7373 (beah-srv)
CGroup: /system.slice/beah-srv.service
└─7373 /usr/bin/python /usr/bin/beah-srv
Aug 26 21:21:25 localhost.localdomain systemd[1]: Starting The Beaker Harness server....
Aug 26 21:21:25 localhost.localdomain systemd[1]: Started The Beaker Harness server..
[root@localhost yum.repos.d]#
=========================================
Starting it as shown also created /sys/fs/cgroup/systemd/system.slice/beah-srv.service/ as expected.
What are the failure symptoms you're seeing?
Independent of any direct impact on beah, this systemd change may have flow on effects to the use cases that led to the ControlGroup setting being added in bug 773259 in the first place.
Hi Nick, It looks like there is just a significant delay before systemd tries to launch beah. I am doing this in my home lab so maybe I don't have time services configured properly and chrony is timing out? In any event, this is what I see in the journal on a fresh install of Fedora-20-20130827 fedora x86_64 Aug 27 13:00:17 localhost.localdomain systemd[1]: [/usr/lib/systemd/system/beah-srv.service:4] Failed to add dependency on beah-beaker-backend, ignoring: Invalid argument Aug 27 13:00:17 localhost.localdomain systemd[1]: [/usr/lib/systemd/system/beah-srv.service:4] Failed to add dependency on beah-fwd-backend, ignoring: Invalid argument Aug 27 13:00:17 localhost.localdomain systemd[1]: [/usr/lib/systemd/system/beah-srv.service:9] Unknown lvalue 'ControlGroup' in section 'Service' So I don't think this is a blocker but I do think we will need to remove this option from the services file eventually. Thanks Verification tip: Note that this affects RHEL 7 and Fedora systems only. Messages such as "[/usr/lib/systemd/system/beah-srv.service:9] Unknown lvalue 'ControlGroup' in section 'Service'" should not be present for RHEL 7 in /var/log/messages. |