Bug 1367720 - "Host to Maintenance" failed via cockpit if there is other host is registering to engine
Summary: "Host to Maintenance" failed via cockpit if there is other host is registerin...
Keywords:
Status: CLOSED NOTABUG
Alias: None
Product: ovirt-node
Classification: oVirt
Component: UI
Version: 4.0
Hardware: All
OS: Linux
unspecified
high vote
Target Milestone: ---
: ---
Assignee: Marek Libra
QA Contact: jianwu
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2016-08-17 10:07 UTC by jianwu
Modified: 2016-08-19 07:46 UTC (History)
16 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2016-08-19 07:46:30 UTC
oVirt Team: Virt
rule-engine: planning_ack?
rule-engine: devel_ack?
cshao: testing_ack+


Attachments (Terms of Use)
The cokpit's tag status when clicking Host to engine-part1 (57.18 KB, image/png)
2016-08-17 10:07 UTC, jianwu
no flags Details
The cokpit's tag status when clicking Host to engine-part2 (57.64 KB, image/png)
2016-08-17 10:08 UTC, jianwu
no flags Details
The cokpit's tag status when clicking Host to engine-part3 (60.79 KB, image/png)
2016-08-17 10:09 UTC, jianwu
no flags Details
The reproduce status on engine (89.39 KB, image/png)
2016-08-17 10:09 UTC, jianwu
no flags Details
The vm status on engine (59.38 KB, image/png)
2016-08-17 10:10 UTC, jianwu
no flags Details
The related log about this bug (9.41 MB, application/x-gzip)
2016-08-17 10:12 UTC, jianwu
no flags Details
engine.log about cockpit (5.22 MB, text/plain)
2016-08-19 02:06 UTC, jianwu
no flags Details

Description jianwu 2016-08-17 10:07:40 UTC
Created attachment 1191555 [details]
The cokpit's tag status when clicking Host to engine-part1

Description of problem:
1.After registering to rhevm, the host cannot maintenance via cockpit page icon "Host to Maintenance" when there are other cluster's hosts are activating on the engine.

Version-Release number of selected component (if applicable):
redhat-virtualization-host-4.0-20160816.1
imgbased-0.8.4-1.el7ev.noarch
cockpit-ws-0.114-2.el7.x86_64
cockpit-ovirt-dashboard-0.10.6-1.3.6.el7ev.noarch


How reproducible:
100%


Steps to Reproduce:
1. Install RHVH rhvh-4.0-0.20160816.1
2. Login cockpit via "ip:9090"
3. Register to engine, set up vm on rhevm.
4. Enter "Vitualization" page, click "VMs in cluster".
5. Click "Host to Maintenance" below "Technical Preview".
   


Actual results:
After step 5, click "Host to Maintenance" failed to maintenance its host when there are other hosts in the state of activating.


Expected results:
After step 5, click "Host to Maintenance" can maintenance its host successfully.

Additional info:
After step5, click "Host to Maintenance" could work when all other cluster keeps inactive. it may be effected by other clusters.

Comment 1 Red Hat Bugzilla Rules Engine 2016-08-17 10:07:49 UTC
Target release should be placed once a package build is known to fix a issue. Since this bug is not modified, the target version has been reset. Please use target milestone to plan a fix for a oVirt release.

Comment 2 jianwu 2016-08-17 10:08:12 UTC
Created attachment 1191556 [details]
The cokpit's tag status when clicking Host to engine-part2

Comment 3 jianwu 2016-08-17 10:09:00 UTC
Created attachment 1191557 [details]
The cokpit's tag status when clicking Host to engine-part3

Comment 4 jianwu 2016-08-17 10:09:44 UTC
Created attachment 1191558 [details]
The reproduce status on engine

Comment 5 jianwu 2016-08-17 10:10:14 UTC
Created attachment 1191559 [details]
The vm status on engine

Comment 6 jianwu 2016-08-17 10:12:31 UTC
Created attachment 1191560 [details]
The related log about this bug

Comment 7 Tomas Jelinek 2016-08-18 06:46:00 UTC
- please provide also the engine.log 
- if you do the same operation in webadmin (e.g. hosts main tab, pick the host and press maintenance), does it behave differently?

Comment 8 jianwu 2016-08-18 07:29:07 UTC
(In reply to Tomas Jelinek from comment #7)
> - please provide also the engine.log 
> - if you do the same operation in webadmin (e.g. hosts main tab, pick the
> host and press maintenance), does it behave differently?

On the engine, can maintenance the host anytime, and it did not effected by other clusters.

i will reproduce the situation on 20160817 new version, and will provide related log.

Thanks

Comment 9 Tomas Jelinek 2016-08-18 08:41:08 UTC
thank you, putting the needinfo back to mark we are waiting for logs

Comment 10 jianwu 2016-08-19 02:06:42 UTC
Created attachment 1192017 [details]
engine.log about cockpit

Comment 11 jianwu 2016-08-19 02:08:03 UTC
(In reply to Tomas Jelinek from comment #9)
> thank you, putting the needinfo back to mark we are waiting for logs

i have upload engine.log form new version rhevh4.0 about "host to maintenance "via cockpit.

Comment 12 Tomas Jelinek 2016-08-19 07:46:30 UTC
so, looking at the logs I would say it behaves correctly and if it works from the UI than it is a coincidence / timing issue.
In the logs I can see that:

Operation Failed: [Cannot switch Host to Maintenance mode. Host is contending for Storage Pool Manager, wait for operation to complete and retry.]

and than two lines after:
[org.ovirt.engine.core.vdsbroker.vdsbroker.SpmStartVDSCommand] (DefaultQuartzScheduler4) [6356c111] spmStart polling ended: taskId 'ae842052-f1b0-4535-a7a4-62b797b1264f' task status 'finished'

So, if you tried to run the maintenance now it would pass.
What cockpit is doing is just sending a REST call to:
/api/hosts/<host id>/deactivate
with body:
<action/>

So, I would say this is not a bug.

You could simulate the same as cockpit is doing by sending POST to <engine API path>/hosts/<host id>/deactivate with <action/> as body
If it behaves incorrectly, than please open a bug for the REST API. And if that behaves differently than the cockpit ui, than please feel free to reopen.


Note You need to log in before you can comment on or make changes to this bug.