Jira / Confluence cold stand by

[Casey Daniell]

I found this old document on a cold standby server, but had another question or two on the topic. https://community.atlassian.com/t5/Questions/Setting-up-cold-standby-instance/qaq-p/251291

It seems like the full process to keep a cold standby ready to go is:

1. Logship MS SQL DBs for Jira and Confluence, to another server
2. rsync the  HOME directories to a cold standby box (assuming defaults here)
   • Jira -- /var/atlassian/application-data/jira exclude: 
     • db.config
     • analytics-logs
     • caches
     • export
     • import
     • log
     • monitor
     • tmp
   • Confluence -- /var/atlassian/application-data/confluence exclude: 
     • confluence.cfg.xml
     • analytics-logs
     • backups
     • bundled-plugins
     • export
     • index
     • journal
     • logs 
     • plugin-cache
     • recovery
     • temp
     • viewfile
     • webresource-temp
3. rsync the Install directories to a cold standby box (still assuming defaults)
   • Jira -- /opt/atlassian/jira
   • Confluence -- /opt/atlassian/confluence

Then in a DR scenario is simply,

1. Apply any logshipped logs to the MS SQL DB
2. Verify the servernames / DB are correct in 
   • Jira- db.config 
   • Confluence - confluence.cfg.xml
3. Start Jira / Confluence
4. Update DNS 
5. Rebuild indexes on Jira / Confluence 

I realize this is a very high level DR plan, but this looks like a pretty good way to keep a DR box in sync with a PROD server and ready to take over should an entire datacenter fail