Amuleweb stops at random

[Kayn2106]

No matter what amuleweb stops randomly with the logs: Command ''amuleweb' '--amule-config-file=/home/amule/.aMule/amule.conf''..... has finished with status code '-6'

This can happened once a day as well as 10 times in an hour, I wasn't able to understand what causes this problem; however it seems a long known problem 'cause I found old threads around the web that mention this behavior, so I'm guessing that's an old problem that has never been fixed. Anyway, let's move on. Since aMule keeps working properly, I'm wondering if it could be possible to write a script that checks if amuleweb is stopped and if needed restarts it or maybe add a modification that does this, just like the kad graph fix and the others...

[ngosang]

Since aMule is not maintained this will be a good improvement. Some notes:

• The entypoint starts amuled -c /home/amule/.aMule -o
• amuled reads the amule.conf file, the section [WebServer] Enabled=1 and it starts amuleweb --amule-config-file=/home/amule/.aMule/amule.conf
• If the process amuled crash it's restarted by the entrypoint
• If the process amuleweb crash it's not restarted by amuled

Possible fix:

• Set [WebServer] Enabled=0 so amuled does not start amuleweb
• Start amuleweb from the entrypoint but we can't use the amule.conf file. Inside the Docker container you can run man amuleweb to see more details.
• We should read the amule.conf file using bash and set the equivalent parameters in amuleweb

Improvement:

• Instead of doing a lot of things in the entrypoint, we should use s6 to keep running the processes. This project has a good example => https://github.com/kanboard/kanboard/tree/main/docker

I don't have time to work on this. Any PR is welcome.

[puigru]

Experiencing the exact same issue

[Kayn2106]

Experiencing the exact same issue

Yeah, that's driving me crazy. Unfortunately I don't have the knowledge to solve this so....

[garret]

I have been experiencing that that the webui and also access via gui ui is extremely laggish after update to latest docker image version. I reverted to previous version (from around october 2023) and experience is smooth. I advice to try to revert back to previous image and see if you still experience the same issues. For the moment, I will stay with the older version since it just works.

[Kayn2106]

I have been experiencing that that the webui and also access via gui ui is extremely laggish after update to latest docker image version. I reverted to previous version (from around october 2023) and experience is smooth. I advice to try to revert back to previous image and see if you still experience the same issues. For the moment, I will stay with the older version since it just works.

I've started using this container back in the mid 2023 and the issue was already there and unfortunately no changes in the following releases... However right now I don't know why, maybe due to the recent Unraid upgrade to 6.12.6, I think it's more stable I've notice less webui crash if none. I mean maybe one crash in the last two weeks, when before it was almost sure at list one or two crashes during the first hours. I can live with that; that's not perfect, but also not unacceptable :)

[puigru]

It's a shame to see that ed2k clients / web interfaces are in such disrepair. I used to use mldonkey which seemed to be more stable but was similarly dated and has other issues. It would be nice if someone took the time to make a modern interface using the amule's EC protocol. That way we could have a better ed2k client without requiring a full on reimplementation.

Something like https://github.com/mwarning/p2p-gui

[Kayn2106]

I spoke too soon. It's crashing again like before: four times only in the last hour and I didn't change anything on my NAS...

[puigru]

For anyone still running into this, I've switched back to mldonkey. It's been working like a champ and it arguably has a better UI I'm using this image: https://github.com/Wibol/mldonkey-docker

[garret]

I confirm I have been running the previous version for many months without any issues. So my suggestion is to not update to the latest tag version on docker. Have you tried that?

I know very well mldonkey but I much prefer the interface of the personalized skin of amule coming with this image :)

[ngosang]

I confirm I have been running the previous version for many months without any issues. So my suggestion is to not update to the latest tag version on docker. Have you tried that?

@garret Could you try with the latest release? https://github.com/ngosang/docker-amule/releases/tag/2.3.3-18

[garret]

@ngosang I just updated to it. I will write a new reply to this issue if I start getting the same issue. Have you just updated the image to the latest alpine package version? Thank you in the meantime.

[garret]

Just experienced after 2 days of running the container non stop that the webinterface does not work anymore. So I can only connect via amule client (but everything else work, so the downloading/uploading thing). I just restarted the container again. If it happens again I guess I will come back again to v16.

[ngosang]

@garret You can try the "auto restart" and "fix kad stats" mods = https://github.com/ngosang/docker-amule?tab=readme-ov-file#modifications

[garret]

I have always had those options enabled since I used your image ;) In any case, I have been continuing using the latest release at the moment of writing. The image has always worked on the background 24/7 and only maybe 3 times in total I had the issue where the web interface went down but amule core was still running. In the meanwhile, I also switched from a raspberry pi server to a nixos mini pc. Coincidentally, today I had an issue where all the 32GB of the ram got used and all the 4 processors of my N100 server. Restarting the container did not help. I had to cancel some files from the amule config to solve. I could see the log and known2.met file were above 180MB. After cancelling those files, everything seems again normal. All in all, I am satisfied with the image and will keep it running.