Rolling deployment proceeds with server 2 even if step for server 1 fails

[user13]

Just came across an issue that we haven’t seen before, using Octopus Deploy v2021.2 (Build 7727).

We have a rolling deployment setup that deploys our web application on to two web servers. This morning when deploying to production one step in the process for the first server failed, causing the remaining steps for that server to not proceed.

However, it did start the deployment for server 2 which in this case took down our site completely. I’m quite sure that this has worked fine before.

All steps in the process are set to “only run when previous child steps succeed”. Feels like a bug to me? Is there anything else I can send to you to help you investigate, please let me know.

/Martin

[paul.calvert]

Hi Martin,

Thanks for getting in touch!

It sounds like you may be running into this issue.

It looks like our engineers have started work on a fix for this, but haven’t yet merged it into a release. If you keep an eye on the issue a comment should be added with the relevant versions once a fix is merged.

Regards,
Paul

[user13]

Hi Paul,

Yes that sounds very much like the same issue, thanks for pointing me to the right ticket in GitHub, will keep an eye on it’s progress.

Thanks!
/Martin