Can we all please vote on https://jira.atlassian.com/browse/CWD-3946

[Gaven Ray]

Hi Crowd Community,

Can we get together and all vote on https://jira.atlassian.com/browse/CWD-3946 so that we might finally get a Crowd installer?

This is the biggest headache when it comes to upgrading Crowd it is fully manual and takes way more time than it should. Confluence and Jira have installers that can do an upgrade it should be possible and would likely help with https://jira.atlassian.com/browse/CWD-5292 ( a zero downtime upgrade path).

https://jira.atlassian.com/browse/CWD-5729 would also be helped by an installer an LTS version of Crowd.

Thanks,

Gaven

[Craig Castle-Mead]

hey @Gaven Ray 

Given Crowd is important within an environment, it might be worth investing in some scripting/desired state configuration to help with your upgrade tasks. We’ve got Puppet running to manage our OS and up across our whole stack, and a Crowd upgrade takes ~ 60 seconds per node. With Crowd running on multiple nodes, the effective downtime of an upgrade is < 20 seconds (drain your A node, stop the app, download and apply customised settings - stop your B node (outages starts) - start the A node (outage ends one started), take your time on finishing the upgrade on the B node).

For some patch upgrades, we’ve even done zero downtime - just rolling upgrades across the nodes. This is NOT supported and use at your own risk.

If you don’t want to use something like Puppet for DSC, you can still script using tools like https://augeas.net/ to help handle changes to parts of files.

CCM