Hi everyone! :slightly_smiling_face: I read <the ...
# general
Hi everyone! 🙂 I read the FAQ on error-handling (I like that this is the first question!) but I’m still left wondering: Can you define custom deploy-time error-handling logic in Pulumi? For example, if I hit some CloudFront API rate-limit with a
pulumi up
, can I define error handling that says “Wait thirty seconds if you get this error code and retry only once” in a way that I can unit-test with something like
Not currently. I'm sure this idea has come up before but I can't find a github issue for it... feel free to make a new one, especially if you've got a concrete use case in mind.
Please let me know if/how I can be more helpful in my GitHub issue description! ❤️ I really care a lot about this use-case, as I’ve been so disappointed by pretty much every IaC tool that seems like it could never support it. Maybe it’s just a bad idea to begin with! But if so, I really want to learn why.
No it's a fair feature request, I'm really surprised I can't find any other issues about it because I'm sure it's come up before.
I wouldn’t be surprised if it never has! As far as I’ve seen, there hasn’t been much of a discussion about unit-testable and custom deploy-time error handling in the IaC community. It seems like we’ve converged on templates and DSLs where that sort of thing seems really hard to implement. To me, it’s one of those, “Is everyone crazy, or just me?” sorts of things. Hits you right in the impostor syndrome. 😆 But I think Pulumi challenges a lot of traditional thought in that community in a really exciting way!
😆 1