Additional transmission-daemon processes spawned for each blocklist updates about docker-transmission HOT 3 CLOSED

I cannot replicate this :

 s6-svc -t /var/run/s6/services/transmission
Caught SIGTERM signal!
root@ce2f1f6db79d:/# localhost:9091/transmission/rpc/ responded: "success"
[2020-01-02 16:10:32.874] Saved "/config/settings.json" (variant.c:1266)

You say -h we changed that some time ago to -t, I assume that is a typo.

Also you have set S6_KILL_GRACETIME was this a debugging thing ?

from docker-transmission.

You say -h we changed that some time ago to -t, I assume that is a typo.

Actually, I think, this is the culprit. My script contained what I quoted ( -h ), ie. it must have been the old version of the script before commit 787d572.

Thinking it over, /config is an external volume to the container, so, /config/blocklist-update.sh is not inherently updated with the container image. I gave it a try and shut down the transmission container, moved away the blocklist-update.sh script from the config directory, then restarted the container. Lo and behold, there was a new (now: up to date) blocklist-update.sh created in the config directory. So, for me this is fixed now.

Maybe this issue report should be closed, and a new one be opened about how to deal with external files, so that they get updated when their templates in the container are updated.

As for S6_KILL_GRACETIME, yes, I found this one while looking for possible causes. I think I'll leave that now as it is, as I noticed that it usually takes more than 3 sec for transmission to exit on this machine anyway. (And I have the time... :-) )

Thank you!... :-)

from docker-transmission.

Personally I'm still seeing this issue.

Executing:

$ /config/blocklist-update.sh
$ ps
PID   USER     TIME  COMMAND
    1 root      0:00 s6-svscan -t0 /var/run/s6/services
   31 root      0:00 s6-supervise s6-fdholderd
  248 root      0:00 s6-supervise transmission
  249 root      0:00 s6-supervise cron
  251 root      0:00 /usr/sbin/crond -f -S -l 0 -c /etc/crontabs
  267 abc       0:11 /usr/bin/transmission-daemon -g /config -c /watch -f
  277 root      0:00 bash
  308 root      0:00 bash ./run
  311 abc       0:02 /usr/bin/transmission-daemon -g /config -c /watch -f
  317 root      0:00 ps

I also had /config being very old so I cleared storage and got it fresh. Hence it includes the s6-svc -t /var/run/s6/services/transmission change mentioned in previous comment.

Maybe as an indicator of the reason this could happen is my config file being mounted as :ro like so:

volumes:
      - ./transmission-settings.json:/config/settings.json:ro
      - /downloads:/downloads
      - transmission-config:/config

I'll try without the :ro and will let you know the outcome.

EDIT: even by removing the readonly the issue still occurs. Let me know if you what kind of information would be helpful to provide.

EDIT 2: maybe it's password related seeing this in logs:

transmission_1  | Caught SIGTERM signal!
transmission_1  | Unexpected response: <h1>401: Unauthorized</h1>Unauthorized User
transmission_1  | [cont-finish.d] executing container finish scripts...
transmission_1  | [cont-finish.d] done.
transmission_1  | [s6-finish] waiting for services.
transmission_1  | [s6-finish] sending all processes the TERM signal.
transmission_1  | [2020-03-17 00:39:18.808] Reloading settings from "/config" (daemon.c:472)
transmission_1  | [2020-03-17 00:39:18.808] RPC Server Adding address to whitelist: 127.0.0.1 (rpc-server.c:971)
transmission_1  | [2020-03-17 00:39:18.808] RPC Server Serving RPC and Web requests on port 127.0.0.1:9091/transmission/ (rpc-server.c:1213)
transmission_1  | [2020-03-17 00:39:18.808] RPC Server Password required (rpc-server.c:1220)
transmission_1  | [2020-03-17 00:39:18.808] Port Forwarding Stopped (port-forwarding.c:180)
transmission_1  | [2020-03-17 00:39:18.900] Reloading settings from "/config" (daemon.c:472)
transmission_1  | [2020-03-17 00:39:18.900] RPC Server Adding address to whitelist: 127.0.0.1 (rpc-server.c:971)
transmission_1  | [2020-03-17 00:39:18.900] RPC Server Serving RPC and Web requests on port 127.0.0.1:9091/transmission/ (rpc-server.c:1213)
transmission_1  | [2020-03-17 00:39:18.900] RPC Server Password required (rpc-server.c:1220)
transmission_1  | [2020-03-17 00:39:18.900] Couldn't bind port 55032 on 0.0.0.0: Address in use (Is another copy of Transmission already running?) (net.c:379)
transmission_1  | [2020-03-17 00:39:18.901] Couldn't bind port 55032 on ::: Address in use (Is another copy of Transmission already running?) (net.c:379)

EDIT 3: Indeed it is due to the auth aspect. If rpc-authentication-required is set to true. S6 Kill fails leaving one process behind. Any ideas on how to fix this?

from docker-transmission.