| Summary: | Allow terminate (force stop) instances if a provider is not accessible | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | [Retired] CloudForms Cloud Engine | Reporter: | Jan Provaznik <jprovazn> | ||||||
| Component: | aeolus-conductor | Assignee: | Jan Provaznik <jprovazn> | ||||||
| Status: | CLOSED CURRENTRELEASE | QA Contact: | pushpesh sharma <psharma> | ||||||
| Severity: | high | Docs Contact: | |||||||
| Priority: | unspecified | ||||||||
| Version: | 1.0.0 | CC: | akarol, bbandari, calfonso, cpelland, dajohnso, deltacloud-maint, hbrock, jlaska, jturner, morazi, ssachdev, whayutin | ||||||
| Target Milestone: | rc | Keywords: | Reopened | ||||||
| Target Release: | --- | ||||||||
| Hardware: | Unspecified | ||||||||
| OS: | Unspecified | ||||||||
| Whiteboard: | |||||||||
| Fixed In Version: | v0.8.0-35 | Doc Type: | Bug Fix | ||||||
| Doc Text: | Story Points: | --- | |||||||
| Clone Of: | Environment: | ||||||||
| Last Closed: | 2012-12-13 19:49:35 UTC | Type: | --- | ||||||
| Regression: | --- | Mount Type: | --- | ||||||
| Documentation: | --- | CRM: | |||||||
| Verified Versions: | Category: | --- | |||||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||||
| Attachments: |
|
||||||||
|
Description
Jan Provaznik
2012-02-08 13:18:01 UTC
if this is *not* implemented I suspect customers will have to remove via the db :( Pushed to conductor repo master branch and 0.9-maint branch a289a6f86b55a5049487ea81277a67e8eb878439 I additionally relaized that this patch doesn't cover single-stop instance from deployments show pretty view, switching back to 'assigned' patch for single-stop sent: https://fedorahosted.org/pipermail/aeolus-devel/2012-February/009194.html commit 93c6773dd9983c81a35ead36854b7cf2344c2919 I am not seeing where this is suppose to show up. Basically I started a instance and then stopped the rhevm jboss server thus making the api inaccessible which causes the instance to transition to a 'vanished' state. Was hoping Jan could provide some more detail on how to verify this. Marking needinfo This dialog is displayed when dc core is not accessible, try to stop deltacloudd service instead of rhevm jboss server. Same thing, instance transitions to 'vanished' state... please advise if this needs to be closed as not a bug now or reopened for further work. You are right, inaccessible instances are now set to 'vanished' state, so the termination dialog almost never appears (because there are no active instances which could be stopped). Summary is on wiki page: https://www.aeolusproject.org/redmine/projects/aeolus/wiki/Instance_and_provider_states IOW, it doesn't make sense to have both: -this instance terminate dialog and setting - set vanished (or other) state for inaccessible instances Solution of this depends on what we expect from "disable provider" action - I tend to think that instances should not be stopped when a provider is disabled - then this dialog can be removed, we will set 'not accessible' or 'unknown' state if the provider is down and it will be consistent. Thoughts? assigning to Pushpesh I tried following steps:- 1.Launched a application in ec2. 2.Stopped the deltacloud-core service and as stated instance goes into vanished state. I am agree with Jan Provaznik that if a provider is not accessible(or disabled) then those instances should not be stopped.Vanished state makes sense for me,but i would suggest a more descriptive message should also be displayed like "Provider is no longer accessible,Please ensure all aeolus-service are running or contact the redhat/cloud provider" Marking this bug to assigned state for now.Please provide your suggestions. moving to 1.0.1 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. http://rhn.redhat.com/errata/RHEA-2012-0583.html Reopening, it appears this issue may remain, and was not resolved in cloudforms 1.0. psharma: Can you indicate whether this problem remains in CloudForms-1.0.1? We believe this has been addressed by the improved launching code. The issue should not be reproducible anymore. If an error message is misleading, please fill in a separate BZ for it (it's probably not directly related to this BZ). Vanished state is set for a instance for which provider is inaccesible.History section of the instance gives sufficient clues about why that has happened. Marking as verified.Verified on:- [root@dhcp201-113 ~]# rpm -qa|grep aeolus aeolus-conductor-doc-0.13.7-1.el6cf.noarch aeolus-all-0.13.7-1.el6cf.noarch rubygem-aeolus-cli-0.7.1-1.el6cf.noarch aeolus-configure-2.8.6-1.el6cf.noarch rubygem-aeolus-image-0.3.0-12.el6.noarch aeolus-conductor-0.13.7-1.el6cf.noarch aeolus-conductor-daemons-0.13.7-1.el6cf.noarch [root@dhcp201-113 ~]# Created attachment 613946 [details]
vanished
Created attachment 613948 [details]
history
|