Hi there, I have an open question for you all, which is a more wide subject than just Pulumi. I have several cloud projects to manage (with Pulumi), and each time I need to change something, deploy a new service or spawn a new stack I end up with the same problem : a bunch of things have changed and I have to spend minutes/hours to figure out what’s wrong, fix it and then only I can do what I wanted to do… until next time. That can be an apiVersion outdated, gcloud that suddenly decided to change a bunch of things, etc So my question is, how are you all coping with that, can’t we have a stable environment for more than a month straight ? Would be useful to share tips here.

IMO the only viable strategy is that you either run

pulumi preview --expect-no-changes

(preferably with

--refresh

) inside a periodic job (eg. Github Action) that pings you (Slack? Email?) when it fails, or do the same but actually apply the stack (and again ping you when it can’t), ensuring that what you think is running and what is actually running are in sync.