alfem/telegram-download-daemon

RPCError in FileRequest

Opened this issue · 13 comments

Hi! I'm getting this: Error: RPCError -503: Timedout (caused by GetFileRequest)
Not sure if this is a local problem caused by my connection or my server but it has been happening for a few days. Is there any way to further investigate ?

Thanks for your work on this marvelous app.

I have this too. It seems to come and go in waves. I think when load on telegram is high it drops more packets and the settings for retry are just too low or something

Same error

I need some more data to debug this error. How many files are you downloading simultaneously? What are you using as your operative system? Are you using docker?

Im using docker, in a raspi 4. It happens only in a few files, it seems to happen only in certain files, but no extension, no certain file, not certain filename....

Hello, i´m having the same error in apparently random files...

maybe i start downloading a 4 rar files pack and 2 download without problems and the other 2 give back error after download part of the file (not always the same ammount).

I´m using docker in a dedicated server.

I have the same error. Raspberry Pi OS operating system on raspberry pi 4 (I don't use docker). Until about a month ago it worked without problems but now in certain files (I didn't find a pattern) it gives the error mentioned in this thread

AFAIK RPC errors show up when something is wrong in Telegram servers. There is very little thing I can do to fix them.

Could you tell me if file download ends successfully after one of this RPCErrors?

AFAIK RPC errors show up when something is wrong in Telegram servers. There is very little thing I can do to fix them.

Could you tell me if file download ends successfully after one of this RPCErrors?

The files end successfully if i resend them to the daemon telegram channel (sometimes only one more try its needed, sometimes give again the RPC error and have to resend 2 or 3 times until it successfully download)... It´s very random in all the steps, really.
Maybe something with the telegram servers (timeouts? workload?.... idk),

I don´t see a pattern here.... sorry...

Could the download be retried (for example 3 times) automatically with a new implementation in the script in case the error occurs?

Could the download be retried (for example 3 times) automatically with a new implementation in the script in case the error occurs?

This works like a charm
https://pypi.org/project/backoff/

Could the download be retried (for example 3 times) automatically with a new implementation in the script in case the error occurs?

This works like a charm https://pypi.org/project/backoff/

Peeeerfect, I implemented this and it works great, thank you very much!!!

For debugging info. When I chose the telegram pro account i no longer encounter this (only expired file references when loading too big a list). I think the throttling of free accounts is something the software doesnt deal with properly

alfem commented

For debugging info. When I chose the telegram pro account

Probably that is the reason I can not reproduce the error :-)