Retries for DatabaseSecretEngineStaticRole are too aggressive

[ameyp]

If reconciliation for DatabaseSecretEngineStaticRole fails (because the database is unreachable, on because the password is wrong), it is retried once every 4-5 seconds. I think this retry is too aggressive, and we should perhaps implement a backoff on failures. For example, retry after 5 seconds, then 10, then 20, then 40, all the way up to maybe 5 minutes.

[raffaelespazzoli]

this is weird. I'm using the default retry policy for operator which should be incremental retry period (doubling it every time) up to a max of about 6 minutes. Pretty much like pods do...

[raffaelespazzoli]

could you re-test after #118 is merged?

[ameyp]

Sure, will re-test after that's merged

[raffaelespazzoli]

can we close this?

[ameyp]

Tried with latest commit on main, still happening for static roles.