We’ve been experiencing strange issues for the last couple of weeks with Zoom webhooks.
It seems that sometimes the webhook events of our OAuth connected clients (app: Marvelous) are not coming through. When they disconnect and reconnect via OAuth, webhook events for their events start working again.
The latest occurrence was from a customer, who reconnected their Zoom account on Aug 3rd, and today they experienced the issue again.
The event in question is “meeting.started” and uuid is “8oIEExZgSraXf0Q58cNmgA==”
We need these events to let the students of our customers know that the session has started and they can join the session.
Could you please look into this and let us know why this is happening?
Could we please get some response and assistance?
This is affecting a lot of our customers.
Thanks for reaching out and sorry we missed your post
Are you still having this issue where you are not getting the event meeting.started?
If so, could you please share a new meeting id (no older than 7 days) so I can take a look into this
Yes, our customers are still having issues.
And we get a lot of complaints in our inbox because of it.
Is there any other channel where we can report critical errors? Zoom integration is a very important part of our software and not having any feedback for almost two weeks is hurting our clients.
This issue is hard to reproduce, and we can only wait until the customer writes in and asks why their students can’t join the session (where we rely on the event started webhook, so that we can let students in) or why their recordings are not automatically downloaded (where we rely on the recording completed webhook).
Right now, we don’t have another fresh meeting id for you to look into. Is there something else you could do to check, why the webhook events suddenly stop arriving and reconnect fixes the issue?
Talking to our customer who experienced issues this morning, it seems these two meetings might have also been affected:
And after they’ve reconnected their Zoom account via OAuth, the webhook events for meeting gt/bdNoGT+e7xRtqTaoJqQ== arrived normally.
Thanks for this information @jenniferb
Let me try and use those meeting UUIDs you shared with me to investigate this issue, I am not sure i will get any information because the app was reauthorized by the customer but I will keep you updated.
(ZSEE-102887 internal ticket for reference)
We are also experiencing issues with some webhook events not being delivered (e.g.
We had this same issue a few months ago: Not receiving webhooks for customer but receiving webhooks for our own account
Just this morning, the problem is happening again. Customers are complaining about not getting updated data in our product.
We have also noticed this with our own connection - I created 3 separate Zoom meetings this morning and never received webhooks for any of them. After creating a 4th meeting, I did receive a webhook, so I don’t think this is an issue on our end. We are not seeing any errors in our logs, so as far as we can tell the webhook events are simply not getting delivered to our server. We don’t use Cloudflare or any other layer on top of our AWS system, so there isn’t any possibility of these requests getting blocked at a higher level.
I checked the Zoom status page but the only item is about scheduled maintenance for SMS so that doesn’t seem related. The fact that this issue is happening for us and our customers makes me think it’s more widespread than an issue with a single account.
unfortunately, customers can not wait without reconnecting if the integration stops working. They have scheduled events and their customers are waiting for them to start.
There’s a serious issue in Zoom that stops sending webhooks for some accounts after some time.
It’s been happening for about a month now. Could you maybe have a look if something changed in that part of the code. Or check what are the conditions for a webhook of the connected account to be deleted? I think the answer will be found there.
it is definitely widespread. It has happened to over 100 of our clients, and it also happened I believe before, in April/May.
Can you please share with me those meetings id that you created and you did not receive events for?
As well as the name of your integration so I can take a look into this
Thanks for that information. I totally understand why customers need to reconnect the integration and I am looking into this issue. Will provide you more updates as I get them.
The name of the integration is Swivvel.
The meetings I started were instant meetings, and according to this thread it sounds like I can’t access any record of those, so I’m not sure how to find the IDs.
However I can give a rough timeline (these times are all approximate):
10:15 UTC - Started instant meeting, NO webhook
10:23 UTC - Started instant meeting, YES webhook
12:19 UTC - Started instant meeting, NO webhook
12:22 UTC - Started instant meeting, NO webhook
12:27 UTC - Started instant meeting, YES webhook
I think one of the meeting IDs was
83980014243 because I do see a single
meeting.ended event in our logs for that meeting, but there are no corresponding
participant_left events for that meeting ID.
This just happened again. We already reconnected our OAuth integration earlier today.
I just started an instant meeting with ID 897 4877 7346. We have not received any webhook events for this meeting yet.
Hey @elisa.zoom - unfortunately still seeing widespread issues this morning.
Any prognosis from what you can see on your end? It’d be helpful to have a timeline for resolution. We’re getting this issue across the board, missing webhooks for both meeting starts and meeting ends.
Here’s an example from just now: 88243256790
Thanks so much.
same here, this problem is really putting pressure on our support and annoying our customers.
Can you report any progress so far and the estimated timeline for resolution?
I will send you both private messages @cedric-swivvel and @jenniferb
Thanks. One new piece of data: I just checked the
/webhook_logs API endpoint and I am seeing a bunch of webhook delivery errors, such as this:
"response_body": "sun.security.provider.certpath.SunCertPathBuilderException: unable to find valid certification path to requested target",
We’re seeing this for the meeting IDs that we gave as examples above.
Here’s another thread that seems related: Zoom webhook status -12 - #14 by apochr
Though the stated resolution in that other thread is:
This is resolved for us now. The issue was infact that we weren’t including the intermediate certificate chain. My only guess is that zoom started checking for this last month as it was previously communicating use the cert. This site was helpful in testing our end of the connection: https://whatsmychaincert.com/
However, our cert chain is correct so I don’t think this is the cause for our current issue.
I sent you a private message earlier today, could you please follow up there so you could provided more detailed information