Telegram bot error

@jantinus Your problem (as you describe it - “clog up the available memory”) is sounding exactly the same as what I am having.

This is the log entry I have:

2019-11-23 10:19:04 WARNING (updater) [telegram.vendor.ptb_urllib3.urllib3.connectionpool] Retrying (Retry(total=2, connect=None, read=None, redirect=None)) after connection broken by 'ConnectTimeoutError(<telegram.vendor.ptb_urllib3.urllib3.connectionpool.HTTPSConnectionPool object at 0x7fd1ea93e8d0>, 'Connect timed out. (connect timeout=5.0)')': /bot8624006###############################ad5sJKfD9iI/getUpdates

At times it requires three or more restarts to clear the log…seems like quite a serious problem to me? Any solutions out there?

UPDATE Decided that since this was a telegram Warning that I would err on the side of caution and revoke my existing API Key for a new one…seems to have stopped the warings so who knows?

UPDATE 2 So since revoking the old API Key all Warning messages have stopped.

PS: Telegram remains working fine?

For me the problem has stopped as well. I now have an external (primary DNS) pi-hole installation on my diskstation. The hassio pi-hole is set up as secondary DNS. No more telegram errors.

1 Like

I don’t use PI-Hole and telegram bot is spammming the adguard with a ton of DNS queries - 2 every 9 seconds …

I have the same issue, appeared in some period of time of the day.

I found this thread: Bot fails when deploying with webhooks on gcloud · Issue #1393 · python-telegram-bot/python-telegram-bot · GitHub
And this article: Notes about GAE Google App Engine · python-telegram-bot/python-telegram-bot Wiki · GitHub

about setting env variable:

GAE_USE_SOCKETS_HTTPLIB : 'true'

Does anyone have any idea how this can be set inside HA?

1 Like

any update about is, Did u find anything about those lines

Might or might not be helpful: I just ran into this error after switching WiFi networks on my HA instance. I restarted HA and the error went away.