Pruned deployment versions are re-used by new deployments
Issue
- Using replace with force starts the deployment from starting deleting the existing ones and without force works as expected.
eg :
a deployconfig with 39 versions, has 5 failed and 5 latest deployments preserved after the pruning is run.
So there are dc version #39 38 37 36 35 34 20 22 17 12, and the pruner deleted the rest.
Now, if we deploy as: oc replace --force -f , it starts deployment #1, rather than starting deployment #40.
Since latest deployment is #1, it fails saying "a newer version of this deployment is already running" and gets cancelled.
However, if we do oc replace -f (without the --force option), it picks up deployment #40 and works as expected.
Environment
- Openshift Container Platform
- 3.5
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.