Goal:
I want the queries to run as expected and not time out, and if they time out to respect set timeouts not always use 120s.
Steps:
I tried changing postgres parameters and Disable converting queries to prepared statements
None have worked
Details:
Queries timeout randomly. they get started but never finish.
We use sshtunneling for a lotof our setups but never have experienced these issues.
Timeouts set under advanced - timeout are just ignored and we need to wai 120s for it to timeout and restart.
This makes the tool absolutely unsusable.
What type of database/resource are you querying? Is there any middleware on the server that you are querying?
So the issue is you want queries that are not working to time out sooner, so you can make tweaks/restart them but instead they are hanging on the connection and can't be exited early?
Hi Jack,
thanks a lot for answering. We solved it for now by having a 'local' version of retool running. Its not our dream setup but that way we could circumvent the sshtunnel.
we have a postgresDB and there is no middleware.
Yes it would already help if the timeouts would be respected.
We figured out that on startup the DB got spammed with connection requests form retool and limits throughput which we believe lead to the weird behavoiur.
Glad you were able to find a work around in the short term.
Interesting theory about the DB getting spammed with connection requests, that definitely seems possible if the throughput is being limited.
My thoughts on next steps would be to dig deeper into the networking requests to confirm this when a Resource on Retool cloud is set to connect to the DB and from there we can check network traffic to see what else we can find out.