Red Hat Bugzilla – Bug 968362
webadmin: after ovirt-engine restart, if we do not reload the UI before log in, CanDoAction alerts will not appear for the user
Last modified: 2015-09-22 09:09 EDT
Created attachment 754423 [details]
Description of problem:
if we restart ovirt-engine service and log in to the webadmin without reloading the UI, all CanDoAction alerts will not appear to the user.
the alerts will appear in engine log but the user will not see any error in the UI.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. log in to webadmin
2. restart the ovirt-service
3. after the service is up go back to the webadmin and log in without reload of login screen.
4. edit a running vm and try to change the memory while its running.
dialogue will not close but there is no error displayed for the user.
looking at the engine log we can see a CanDoAction is logged.
we should see CanDoAction alerts.
2013-05-29 17:13:59,370 WARN [org.ovirt.engine.core.bll.UpdateVmCommand] (ajp-/127.0.0.1:8702-6) [42f6603e] CanDoAction of action UpdateVm failed. Reasons:VAR__ACTION__UPDATE,
I cannot reproduce this on upstream (i.e. after engine restart, CanDoAction errors appear properly in the web-admin after re-login, even if I haven't reloaded the page in the browser).
I imagine that this is not reproducible in rhevm-3.3 either.
[problem originally reported on sf17.2, i.e. rhevm-3.2]
moving to ON_QA for QE to verify that it is indeed not reproducible in rhevm-3.3. If reproducible - please re-open, and BZ assignee will investigate.
(In reply to Einav Cohen from comment #1)
> I cannot reproduce this on upstream (i.e. after engine restart, CanDoAction
> errors appear properly in the web-admin after re-login, even if I haven't
> reloaded the page in the browser).
> I imagine that this is not reproducible in rhevm-3.3 either.
> [problem originally reported on sf17.2, i.e. rhevm-3.2]
> moving to ON_QA for QE to verify that it is indeed not reproducible in
> rhevm-3.3. If reproducible - please re-open, and BZ assignee will
strike that, I was able to reproduce this in ovirt-engine-3.3 (upstream) - perhaps it matters whether a setup has an apache proxy or not (in development environment, which is without an apache proxy AFAIK, the problem is not reproducible).
@Alex - please install latest rhevm-3.3 in order to try and reproduce that - I think that working in development environment in this case isn't enough.
Just to give full reproduction steps:
1. Login into webadmin. Do an action that will generate a CanDoAction failure, for instance adding a network that already exists. This will show an error message (This step is important, otherwise it won't reproduce).
2. While logged in, restart the ovirt-engine. This might generate an error message popup, but will eventually get you back to the login page without interaction on your part.
3. Log back into the webadmin. Do the same or other action that will generate a CanDoAction failure. Notice the failure in the logs, but no message popping up.
I can reproduce this in both development mode as well as a standard installation. Apache has no influence on this. The fact that you created the failure before restarting the engine is what triggers this. Investigating the cause now.
According to instructions I am able to see CanDoAction actions alerts
And UI gives error message that can be closed
2013-10-02 14:49:13,693 ERROR [org.ovirt.engine.core.utils.ObjectIdentityChecker] (ajp-/127.0.0.1:8702-3) Field memSizeMb can not be updated when status is Up
2013-10-02 14:49:13,694 WARN [org.ovirt.engine.core.utils.ObjectIdentityChecker] (ajp-/127.0.0.1:8702-3) ObjectIdentityChecker.IsUpdateValid:: Not updatable field 'memSizeMb' was updated
2013-10-02 14:49:13,694 WARN [org.ovirt.engine.core.bll.UpdateVmCommand] (ajp-/127.0.0.1:8702-3) CanDoAction of action UpdateVm failed. Reasons:VAR__ACTION__UPDATE,VAR__TYPE__VM,VM_CANNOT_UPDATE_ILLEGAL_FIELD
2013-10-02 14:49:13,694 INFO [org.ovirt.engine.core.bll.UpdateVmCommand] (ajp-/127.0.0.1:8702-3) Lock freed to object EngineLock [exclusiveLocks= key: RHEL64-orig value: VM_NAME
, sharedLocks= ]
Error while executing action:
There was an attempt to change VM values while the VM is not down. Please shut down the VM in order to modify these properties.
Closing as part of 3.4.0