Hello, I recently added `s3compute log manager` to...
# deployment-kubernetes
j
Hello, I recently added
s3compute log manager
to my helm chart and deployed successfully. It works as expected; However, dagit was bumped up to
0.11.3
(from 0.11.2) and a few minutes later my hourly schedule was kicked off but errored out. I didn't make any other changes prior to the above and the version bump but I also couldn't find anything in the release notes except the below, which I don't believe applies in my case (I am using
pipeline_name
but the pipeline exists
Copy code
Schedules and sensors that target a pipeline_name that is not present in the current repository will now error out when the repository is created.
lastest tick Apr 6 .... UTC Failure
d
Hi Jason - the most likely cause here is that your user code deployment is still back on 0.11.2 and you're running into a version incompatibility between the daemon and that deployment
if you update user code to 0.11.3 I'd expect this problem to go away
j
It is on 0.11.2 but I thought minor version differences was okay
d
We try to maintain version compatibility within a single major version when possible but it's not something we currently can guarantee
what we do guarantee is that any breaking changes when you update your full deployment within a major version will be rare (if ever) and clearly noted within the release notes
j
Okay, thanks. I'll update the user code and test that out.... Would you happen to know if there's a way to pin versions?
so
helm upgrade
only makes changes in the values file
I typically just run something like
helm upgrade --install dagster-release dagster/dagster -f helm_values.yaml
r
Just add --version 0.11.2 to the end of the install command to pin the chart version
👍🏾 1
👍🏼 1