This message was deleted.
# general
s
This message was deleted.
b
can you clarify a bit more what workflow you're using?
w
Doing exactly this, haven't hit too many issues (raw kube and helm). Also happy to help :)
m
Two similar scenarios: 1. Deploying a Helm chart, I changed some values passed into the chart after it had been deployed. When I re-run
pulumi up
it partially fails, b/c it tries to re-create the deployment instead of just applying over top. 2. Using
k8s.yaml.ConfigFile
, I deployed several resources. Then, I changed the provider to target a different namespace, which now causes some failures b/c pulumi tries to re-create Cluster-scoped resources instead of ignoring or just applying overtop. I guess those are a bit more nuanced scenarios than I was originally thinking. 🙂
w
Are you using chart or release for 1?
I'm using Release and that use case works fine for me.
m
Remote Chart for 1
I saw the Release option, but didn’t choose it for some reason…