docker becames unresponsive after some time

[hogmac]

Hi - the container is working great and functional so far. it does a great job when can keep online, the problem I'm facing is that after a few minutes without processing queries or being idle like no one is using it the logs start showing entries like this:

2023-10-13T01:37:49Z ERR EOF service=http

and then, became totally unresponsive and need to restart the container to start working again. any idea why this may happen or do you know any workaround to keep it from losing the connection like this?

[Seji64]

Hm, i have the Container Running on Kubernetes and on "normal" Docker and do Not have any of these issues. Is the Docker deamon logs something?

//EDIT:

Could be related => mosajjal/sniproxy#76

[hogmac]

in fact seems related to https://github.com/mosajjal/sniproxy/issues/76 however from my testing i pulled an older snidust image https://github.com/users/seji64/packages/container/snidust/103503017?tag=sha-40ae16a and that seems to work better as it only had a couple of hiccups in the last 12 hrs.

Marking downstream cloudflare-dns (1.1.1.1:443) as 'up'
Marking downstream cloudflare-dns (1.0.0.1:443) as 'up'
Marking downstream dns.google (8.8.8.8:853) as 'up'
Marking downstream dns.google (8.8.4.4:853) as 'up'
PowerDNS DNSDist Security Update Mandatory: Upgrade now.
2023-10-13T07:10:59Z ERR http: no Host in request URL service=http
2023-10-13T07:15:07Z ERR http: no Host in request URL service=http
PowerDNS DNSDist Security Update Mandatory: Upgrade now.

not sure what has changed in newly images that is causing this issues. As a reference I'm on Docker version 24.0.5, build 24.0.5-0ubuntu1~22.04.1. I will keep testing the image version I pulled to see if becomes more stable at some point.

[Seji64]

please note replacing sniproxy with nginx resulted in a breaking change. nginx is listening on port 8080 (before 80) and 8443 (before 443). Ensure you update your docker configuration accordingly!