Operation timed out after 20003 milliseconds with 0 bytes received using Bus::Chain

[Kyon147]

Hey,

Not sure if you've run into this before, but using job chaining, if the chains are big I am seeing Operation timed out after 20003 milliseconds with 0 bytes received from CURL

Have you run into this before?

[marickvantuil]

I can have a check later this week. Lmk if you find something in the meantime.

[Kyon147]

Thanks @marickvantuil I am looking currently on a separate branch to see what could be a cause.

I am wondering if a big chained job is causing CURL underneath for Google tasks to timeout. As the containers that run Cloud Run for jobs have a 600 second timeout, with nginx and php matching.

[marickvantuil]

@Kyon147 Just for full transparency, which version of the package are you using, and on which platform is the application hosted? (AppEngine/Cloud Run/something else).

I first thought it could be slow because of Cloud Tasks's deduplication feature, but even with 500+ tasks in the queue, it only takes around 500ms to create a task. Could it be some quota thing from Google somewhere?

I've also experimented with job chains, but I run into { "message": "Task size too large", "code": 3, "status": "INVALID_ARGUMENT", "details": [] } errors even before anything times out.

[marickvantuil]

@Kyon147 Did you by any chance find out more about this?