The process for Jira is killed in Jira software data center.

Question

I am trying to install jira data center in docker. When start the jira process in the second node. the jira process in first node will be killed. and then when I

try to start the jira process in first node. the process in second node will be kill. There is no available error log in all nodes. Is there anybody to kown this problems? Thanks a lot.

taotao · Answer

Thank you all the same! I resolved this problem by myself.  The environment which I built in is docker in mac, then I switch  to the docker in ubuntu, the problem disappered.

It's so wired.

And I encounter a new problem again.

when I use the mysql-ndb as the sql of my data center. the jira service will be killed in a node.  There is also no any error log  in ${HOME}/log/atlassian-jira.log. I will be crazy about this.  By the way, my jira software version is 7.5.0.

Andy Heinzer · Answer

The only thing I can think of that might cause this behavior is if the two nodes do not have unique identifiers in the cluster.properties file per the Installing JIRA Data Center Guide.

Each node needs to have its own unique identifier.  If you have both nodes with the same identifier (which might easy to do if cloning the whole system with docker), then the cluster isn't going to work properly.

I admit this is largely a hunch.  I would expect there to be some sort of log on the system to explain this behavior though.

If this isn't the problem, then I would recommend looking at How to resolve JIRA process being terminated unexpectedly in linux.  I wouldn't expect this to happen in a docker though, unless the host is limited/limiting the amount of memory available to JIRA to start up.  This KB does describe a scenario where the JIRA process can be killed unexpectedly on the system without recording a log entry in JIRA.  If that happens there are other system logs that should record this event as mentioned in that KB.