For the last couple of months, one of our pages has been incredibly slow when it's first opened. It seems that when the page gets opened it triggers the query but it gets stuck trying to execute it. We assume this is what's happening because when it works correctly we can see the other queries that get triggered by the first one.
At the moment, our best guess is that is somehow memory related, but even those results have been inconsistent. We've tested on different computers and browsers on different memory loads and most often than not the one scenario that had the least memory consumption was the one that worked. In rare cases, some scenarios where it wasn't working suddenly start to work.
We've had this setup for almost a year before we started experiencing this problem, and it seems to have gotten worse recently. Looking at the recent Retool changes (up to July) there doesn't seem to be any changes that would affect us directly.
For context, these are the steps our page takes when its first opened:
- The First query decides which other query to trigger
- Second query gets triggered
- The second query grabs some values before accessing our firebase database
- Third query gets triggered using values passed from the third
- The third query returns the obtained values to the second query and the second to the first
*Our page gets stuck on step number 2, before accessing the database, so it's not a problem with the amount of data that's being fetched since it doesn't reach that step.
We wanted to see if anyone else has experienced something similar to this or if someone could give us any insights into what might be going on.
Thanks in advance Retool community!
Thanks for surfacing this. We've seen a few other users report intermittent query failures and are looking into a potential source of those issues that is memory related. The timeline you're describing seems to line up with what we've seen on our end. Are you seeing any particular error messaging attached to the first query that's hanging?
Also, do you have a good estimate of how many times you saw this happen and at what times it occurred in the past few days?
The query doesn't throw any error, it just hangs indefinitely.
Can't provide you with specific hours at the moment but my company operates from 9 to 5, which is where we really notice the issues. There have been occasions where we try to access it on our off hours and it still happens.
We tested with different browsers and Chrome is the biggest offender. On one of those tests, Chrome took about two minutes to fully load the page (one of the rare occasions where it didn't get stuck) while Edge took about 30 seconds (which is still longer than usual).
Got it, alright. Is this in your cloud org or a separate, self-hosted instance? In case it's the former, would you mind if we stepped into the app to take a look directly?
Small update: For some reason removing the timeout time from the initial query allows it to trigger the second and even the third one but it still gets stuck on either one of them.
We wouldn't mind if you checked out our app. It isn't self-hosted so let me know how we should proceed.
We had 4 hidden timers that we used to display a live countdown for each item on a table, up to 4 rows. The issue seems to have been caused when the timer values were invalid because, after removing them, the queries started executing successfully and consistently.
I have since added the timers to a dynamic list so that they only populate if there's data for them. Everything seems to be working as expected now.