we saw the outage yesterday with a quick resolution time (nice job SRE's on that)
not sure if this is related, but we just observed some weird behavior in one of our Workflows this morning. let me explain
we have a postgres query which runs for minutes initially (after a table drop/recreate) and then for seconds for every execution after that. in Query Library, runs execute in seconds (as we want), but then the Query Library-associated block in our Workflow runs indefinitely. We tried hitting "Sync" button on the block to cross that off the list -- it didn't change the behavior.
I saw a post from April 10th about Workflows not respecting timeouts, but in our case we aren't trying to increase beyond 120s and instead just restrict to default 10s. my understanding is there are blanket 120s timeouts across Retool with some exceptions. we're fine with 120 for now
So two things we're seeing:
- timeout of 10000ms in the block doesn't seem to do anything
- it runs for over a minute
- when we copy-pasted the block, plumbed the duplicate back in with the rest of the workflow, and deleted the original, everything worked. it executed in a few seconds
thanks!
If you import a Query Library query into a workflow and attempt to set the timeout in the block settings, that value will not be respected. It should always use the value from the Query Library settings