Request timeout after upgrading cluster to M10

MongoDB Atlas
1

After upgrading from the free tier to M10, my requests made from my website, to a Heroku node.js app (which I also upgraded), to the MongoDB, get no response (from Heroku to MongoDB Atlas). But it works fine if I do the requests locally. What could have gone wrong? It was working before the upgrades, I’ve checked the username, password, and I have 0.0.0.0/0 IP address in the network access. Here is the relevant log:

2022-11-13T19:48:27.138026+00:00 app[web.1]: GET /db/levels/sp/all 200 30001.170 ms - 2
2022-11-13T19:48:27.468379+00:00 heroku[router]: at=error code=H12 desc="Request timeout" method=POST path="/db/levels/sp/favs/0" host=gswitchcdb.herokuapp.com request_id=60a029ce-814a-496f-aba5-fbd4c355fed7 fwd="89.114.110.197" dyno=web.1 connect=0ms service=30000ms status=503 bytes=0 protocol=https
2022-11-13T19:48:27.470260+00:00 app[web.1]: POST /db/levels/sp/favs/0 - - ms - -
2022-11-13T19:48:27.470478+00:00 app[web.1]: error: MongooseServerSelectionError: Server selection timed out after 30000 ms

Any help is much appreciated, thanks in advance.

2

I haven’t done that before but the upgrade might have changed DNS names for your cluster. I suggest checking the connection string first since you already have the allowance from “anywhere”.

3

It seems that the issue was indeed related to the connection string. For some reason, on the Heroku version I left out &w=majority at the end of the connection string, which seemed to work before but not now. After adding it, it works as expected. Thanks.

1 Like
4

I think you should accept your response as the solution because it fits more like to be an answer. Mine was a suggestion, and although it is in the connection string, it is different than what was in my mind.

Cheers :wink:

5

This topic was automatically closed 5 days after the last reply. New replies are no longer allowed.