Description of problem: Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1.Build and push any simple image (in my case rhel6.2) 2.Create a deployable from this image 3.Launch overview page is shown all paramter matching 4.On the aeolus-host stop deltacloud core service 5.Launch the deployable from it,deployment fails and now in create_failed state(as expected). 6.start the deltacloud-core service again 7.deleted the instance. 8.try to launch instance from this deploable results in "some assemblies will not be launched provider_name provider is not available" is displayed. 9.Any new instance launch from any existing deployment in any catalog results in same error. 10 aeolus-service restart doesnt help either,new deployment created also results in this. Additional info: [root@qeblade29 ~]# rpm -qa|grep aeolus aeolus-configure-2.7.0-0.20120712224728gitcc8302d.fc16.noarch rubygem-aeolus-cli-0.6.0-0.20120712224735gite2a391a.fc16.noarch aeolus-conductor-0.11.0-0.20120713030022git8c99e02.fc16.noarch aeolus-conductor-daemons-0.11.0-0.20120713030022git8c99e02.fc16.noarch aeolus-all-0.11.0-0.20120713030022git8c99e02.fc16.noarch aeolus-conductor-doc-0.11.0-0.20120713030022git8c99e02.fc16.noarch aeolus-conductor-devel-0.11.0-0.20120713030022git8c99e02.fc16.noarch rubygem-aeolus-image-0.5.0-1.fc16.noarch
Created attachment 599757 [details] all logs
Works fine after some time(i tried after 2 hrs,may start wokring earlier) and few restart of deltacloud-core service.Might be some latency issue. Please update if this is expected. As it didnt works atleast 20 mins or so after the situation.
Conductor (dbomatic) checks periodically every 5 minutes availability of providers and realms. If a provider is not accessible, it's marked as unavailable. If the provider becomes accessible again in some of next checks, it marked back as available. Information about unvailability is displayed on providers show page, you can also enforce accessibility check on this page by clicking on "Test connection". So if you stopped dc-core, then it's possible that for next ~5 minutes the provider was not available. Can you please retest this and check: - if info about unavailability is displayed on provider show page - if the provider becomes available again after 5-6 minutes since you start dc-core again -> then instance launch on this provider should work again too
I tried steps mentioned in Comment#5.Yes it seems there is only ~5 min latency now.This is acceptable.Marking as closed as it workes for me now.