Closed ultrakorne closed 2 weeks ago
Thank you for the reporting @ultrakorne.
Release 1.2.3
includes the fix.
Please check 0838ff0d3729f496a8cba28fcce7dce340577d89 , I have included a log about a possible misconfiguration. Maybe you are hitting the same.
thank for the quick reply, will update to 1.2.3!
i was seeing the error consistently, but my conf is
adapter Tesla.Adapter.Finch, name: __MODULE__, receive_timeout: 40_000
which should be strictly greater than the default 30s timeout
Could you check what is the timeout you get with that config?
Maybe check also what happens with request_timeout: 40_000
instead of receive_timeout: 40_000
.
I don't get any timeout with the finch
config you can see here https://github.com/visciang/telegram/commit/0667e3b28a5bab7f7b354dec299ca66bbe755b41
i updated to 1.2.3, I tried request_timeout
and tried even a longer timeout of 140_000.
and with config :telegram, :get_updates_poll_timeout_s, 20
I am basically running your example Counter
from your telegram example repo, just with a Poller instead of the webhook.
I will try to run the sleepbot example to see if that works for me.
@ultrakorne I've reproduced the issue.
It should be an issue with the tesla finch adapter, options are not passed down to finch from the adapter macro. I'm going to open an issue there.
BTW: if you pass the finc options via config it works, see 3b6f0280231286da906c33ede1eac707da0a22f3. Give it a try!
@visciang thank you, this actually fixes the issue.
Thanks for helping out. I just started learning elixir so my debugging capabilities are very limited
I am not very familiar with the telegram api, so apologies if i get it wrong.
whenever connection goes in timeout with the Poller we get
should'nt we do something like