Measuring "Change Failure Rate"

[Jeren Steffen]

Looking for suggestions on how to measure/track "Change Failure Rate" (deploy failure rate). We use Jira Software, Jira Service Desk, Confluence, Bamboo, Bitbucket, and Octopus.

[Gonchik Tsymzhitov]

Hi! 

Sorry, use case is not clear. Could you clarify, please? 

Cheers,

Gonchik Tsymzhitov

[Phebe Brown]

I just had a request for change failure rate reporting; in this case it would be 

• Change Failure Rate: the % of tickets in a sprint that are Production or Done that are later moved to the Reopened.

[Phebe Brown]

(This is for a given sprint)

[Jeren Steffen]

Sorry for the delayed response, I must have missed the notification of your reply.

We are attempting to follow the 4 kpi's from the book "Accelerate" and are having a difficult time figuring out an accurate (and preferably autonomous) method of tracking Change Failure Rate. This is defined as: The frequency a production deploy results in, and is the direct cause of, any failure requiring immediate repair.

[mmcmorran]

Jeren, did you guys manage to figure out how to monitor this because I am currently in the same boat as you?

[Marcel Risques]

I'd also be interested in learning how to do this, particularly the two "Ops" DevOps metrics (change failure rate & mean time to recovery)

[Karl Hamilton]

I'm also interested in change failure rate relating to, releases, teams and sprints with variable time window etc - on dashboard automatically updated and other Dora measures

[Oleksii Kostiuk]

I was looking into this topic this week. There is no solution available from Atlasian. And Jira alone can not provide it. To be correct this metric should be collected from the CI and Runtime. 

Maybe eventually there will be such function, but for now, i've found different solution. 

https://help.sleuth.io/accelerate-metrics

It has integrations to Bitbucket, Jira and other services, which you might need in order to measure DORA metrics.

I hope this helps.