Hide Forgot
Description of problem: ----------------------- I have created a hosted-engine setup with 3 hosts, with gluster volumes as backend. Hosted-Engine was up and running. I have edited to cluster to check 'gluster service' - to enable gluster capability on the cluster. After saving the change, it took sometime and enabling 'gluster capability' on the cluster failed Version-Release number of selected component (if applicable): -------------------------------------------------------------- vdsm-gluster-4.18.999-978.gitcaed6de.el7.centos.noarch vdsm-4.18.999-978.gitcaed6de.el7.centos.x86_64 ovirt-hosted-engine-setup-2.1.0-0.0.master.20161122155539.git00fff8b.el7.centos.noarch How reproducible: ----------------- Always Steps to Reproduce: ------------------- 1. Install hosted-engine 2. Edit the cluster and check the 'Enable gluster service' Actual results: --------------- 'Enabling gluster service' on the cluster waits for sometime and then it fails Expected results: ------------------ 'Enabling gluster service' on the cluster should be successful
Created attachment 1224227 [details] part of the engine log with errors and tracebacks
2 issues here. The root cause seems to be communication issues between engine and vdsm on the host : 2016-11-25 11:20:44,899Z ERROR [org.ovirt.engine.core.vdsbroker.gluster.GetGlusterHostUUIDVDSCommand] (default task-38) [118de521] Command 'GetGlusterHostUUIDVDSCommand(HostName = host3, VdsIdVDSCommandParametersBase:{runAsync='true', hostId='6cc5b665-363c-41bf-a486-9a6897c0b5eb'})' execution failed: VDSGenericException: VDSNetworkException: Message timeout which can be caused by communication issues Once the VDS command fails, there's also an error in setting the host to non-operational due to failure in getting resource. (Not sure if caused by DI changes)
Are you testing with engine from master?
(In reply to Sahina Bose from comment #3) > Are you testing with engine from master? Hi Sahina, Yes, as mentioned in comment0, I have been testing this with ovirt-master
What's the status of the host? Can you provide complete logs and not only partial ones?
I have reprovisioned the setup and I couldn't get the engine logs. I have tried to install HC setup with the latest ovirt-master ( http://resources.ovirt.org/pub/ovirt-master-snapshot/rpm/el7Server/noarch/ovirt-release-master-4.1.0-0.5.master.20161201000129.gitf370ec3.el7.centos.noarch.rpm ) This time, I am not seeing any such problems, while enabling gluster capability on the cluster. Looks the problem got resolved with the latest ovirt-master. CLOSING this bug as the issue is not happening with the master. Will re-open the bug if the problem re-occurs