Intermittent Performance Issues - Query recieve

[pcgeller]

Hey all,

I'm seeing some intermittent performance drops on some dashboards that ingest a moderate amount of data (50mb on disc, ~70k rows). The query will execute fine on the server but when the data is being read into the app the browser will either crash (with a wait option). Chosing to wait will sometimes load the data but looking at the performance metrics through developer tools shows very long response times for the receive processing - upwards of 6 minutes.

This was happening yesterday and disappeared today. But sounds similar to this - Table performance issue loading a large data on live public link

Any update on root causing the previous ticket or way to know when this might be happening?

[alina.retool]

Hey @pcgeller thanks for flagging this. I'll escalate this to our support team and circle back with an update as soon as I can. Thanks for your patience.

[Tess]

Hi @pcgeller Thank you for checking in on this. We escalated this issue to our on-call engineers and they were able to ship a fix yesterday Your apps should be working as expected now, but let us know if you see any unexpected behavior and we'll be happy to take a look

[pcgeller]

Thanks for the patch. I can confirm apps are back to normal performance.

[dguzzo]

@Tess Would you happen to know if "very long response times for the receive processing" as described by pcgeller above — and what yesterdays' fix addressed — would apply to the "Running queries..." info in the bottom right of an App?

For the past week plus, my team has been experiencing horribly long query times intermittently, all the more confusing because I have those queries configured for timeouts which they seem to be overrunning.

I only have a small sample size of qualitative feedback from our internal users, but they mostly say it's been better today. But if my issue is likely to be separate than the one addressed in this post, I'll need to file a separate ticket.

[Tess]

Hi @dguzzo It sounds like that might be different

Is the browser crashing? That was the main identifier of the issue that our on-call engineers addressed.

What query types are these (API, MySQL, etc)? Is it happening specifically on page load or while interacting with the app?

[dguzzo]

Thanks for the response, @Tess, and sorry for my very slow follow-up.

My browser wasn't crashing; the queries were API; and it happens both on load and during interaction.

I'll plan on filing a new issue when I get a chance.

[blackst]

Did you get anywhere with this, we are experiencing this now. Intermittent connections on our queries. A screen refresh seems to solve it but it does not give confidence in using retool.

[Tess]

Hi @blackst thanks for letting us know! API queries? Did it just start this week? Happy to take a closer look if you give me the ok to step in to your org. Otherwise, it would be great to get more info about the queries - is it contained to one resource? How much data is being returned when the queries work? Are you experiencing a better connection to this resource outside of Retool? Any info about patterns or details you've noticed would be helpful as we research this

We're not currently experiencing an outage, but this is definitely something we want to investigate for you!

[blackst]

I have increased our kubernetes CPU to handle the loading issues. Seems to be stable again , but have more problems with deployments will open a new topic