Jira Shared Home (Jira Data Center) error in RHEL (AWS)

jameers December 17, 2017

Hi,

we have installed jira data center in aws. we are followed https://confluence.atlassian.com/adminjiraserver071/installing-jira-data-center-802592197.html all this steps. but shared home is not working for us. we have given all necessary permission to the folders. but when we start jira it is automatically creating files in local home directory(after deleting). if we remove the cluster.properties file, after restarting it is working fine. Can you please help on this. shared home is /data/jira/sharedhome. nfs service also available in this server.

Thanks in advance

1 answer

1 accepted

0 votes
Answer accepted
Andy Heinzer
Atlassian Team
Atlassian Team members are employees working across the company in a wide variety of roles.
December 18, 2017

What path are you using for the shared home directory location?   This path has to be something that is accessible from all the nodes using the same address.  Hence a NFS share that has a format like //hostname/path or \\hostname\path is expected to be used so that the other nodes can reach this location.

Also, don't delete the local home folder on any of the nodes.  Data center nodes can still make use of this local home folder.   I'd also want to make sure that the cluster.properties file has a uniquely different nodeid for each node in question.

jameers December 18, 2017

path is /data/jira/sharedhome and given full permission(777). In cluster.properties file unique id given. if i remove the cluster.properties file it is working fine. if i keep cluster.properties file it showing "the page isn't working" HTTP error 503

Andy Heinzer
Atlassian Team
Atlassian Team members are employees working across the company in a wide variety of roles.
December 19, 2017

Right, but that path by itself is not going to be accessible to other nodes.   You need to have some other way to make sure this path can be accessed by other potential future nodes.  This usually means configuring an NFS share and using a UNC path.

Without the cluster.properties file, then this really becomes just a single node that is not actually using the data center deployment architecture. 

Try to start this again and then take a closer look at the logs Jira created when this happens.  This will be in two locations, $JIRAINSTALL/logs/catalina.out and possibly in the local $JIRAHOME/log/atlassian-jira.log files.   Seeing what log entries are created at this attempted start up of Jira will tell us more about why this is not working.

Suggest an answer

Log in or Sign up to answer