504 Gateway timeout for /v2/past_webinars/{webinar_uuid}/polls with thousands of participants

Hi, hoping to get a bit of help with this - is there any way to request maybe a timeout increase for this endpoint to get it to return? We call this endpoint a lot for smaller meetings/webinars and it’s usually ok.

When querying the following url to get polling data for a large webinar, a 504 gateway timeout is returned:

504 Gateway timeout

Which App Type (OAuth / Chatbot / JWT / Webhook)?

Which Endpoint/s?

How To Reproduce (If applicable)
Steps to reproduce the behavior:
Find a webinar with a LOT of participants. In the order of thousands. I don’t know if there were polls actual ran, but better to find one that has. Call the api above, it will gateway timeout.

Screenshots (If applicable)

Additional context

Hey @david.bennington,

Thank you for reaching out to the Zoom Developer Forum. Interesting, thank you for bringing this to our attention. How long does the request run prior to timing out?

In other words, what is the duration of the timeout you’re seeing?


Hi Max,

We had our own timeout at 60s being hit first, when I went to replay manually is what showed the 504, so it must have been at least that.

Interestingly, trying now it’s actually responding - it took 37s to generate/return 1.67MB of poll data, 42K lines of json. I’m going to try again tomorrow during the day, see if I can properly reproduce. We were getting alerts for this failing for 2 days with regular retries.

We have a read timeout, and it was hit while waiting for response headers, so it’s not that it was just a super slow download, we got no response at all.

Hi @david.bennington,

Thanks for confirming these details. Can you let us know if you’re still able to reproduce this today?


Hi Will,

Apologies for the delay, I took some last minute leave and didn’t chase this up, which I’m doing now.

Calls for this webinar are making it through today, albiet it taking (literally) 59.59s to return a response! I suspect that this is variable depending on load & size of the response, perhaps being collated in a non optimal manor every call?

Either way it seems like we should be able to proceed, so feel free to make whatever notes you need and close this out, I’ll re-open or make a new one if it fails agian.

Thanks again,

Hi @david.bennington,

Thanks for getting back to us and for confirming—please do let us know if you see any strange behavior on this again.


I spoke too soon…

Getting gateway timeouts on this again, pretty at least this morning. The response returns after 60s.

After about 20 mins I have managed to get a response. Yes it’s an edge case, but of course we have hit it :smiley:

Also, while trying to re-process this our side, the original archive event is no longer available, but this data is (helpful). But when I query /v2/past_webinars/{{webinarUuid}} in an attempt to get e.g. the topic, any chat text docs etc, I get a 200 OK with an empty response body (not helpful!).

Poll information is available, but without this endpoint things become tricky!

Thanks for following up on this @david.bennington — I just DM’d you for some additional details when you have a chance. Thank you!

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.