Jump to content

6.5.2 Can't Stop or Edit Dockers


DZMM

Recommended Posts

For the last couple of weeks, when I edit a running docker I run the risk of the restart failing and then the docker image disappearing - even when I go to restore via CA the previous settings will be gone.  The docker will hang on restarting in the GUI and then eventually just disappear.  Stopping dockers also used to be a problem, but since I moved my docker image from a UD to my cache this seems to have gone away.

 

Edit: stopping dockers still a problem with image on cache - wheel just spins forever.  I've tried moving appdata back to the cache, but that doesn't help either.  Here's an error I just got when trying to stop beets

 

Jul 22 08:38:42 Highlander nginx: 2018/07/22 08:38:42 [error] 8455#8455: *10057 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.30.10, server: , request: "POST /plugins/dynamix.docker.manager/include/Events.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "1d087a25aac48109ee9a15217a105d14c06e02a6.unraid.net", referrer: "https://1d087a25aac48109ee9a15217a105d14c06e02a6.unraid.net/Docker"

 

192.168.30.10 is the IP address of the W10 VM I'm using.

 

Here's the diagnostics from when I tried to change the IP for my running letsencrypt docker, which then disappeared.

 

Help please

highlander-diagnostics-20180722-0708.zip

 

Edit: second logs includes beets error

 

highlander-diagnostics-20180722-0843.zip

Link to comment
3 hours ago, DZMM said:

I go to restore via CA the previous settings will be gone. 

You should update unRaid to 6.5.3.  From the release notes:

  • webgui: bug fix: prevent deleting user template when (letter case) renaming a docker container

If you're changing a name on an app, then 6.5.2 will guarantee that that CA won't be able to find it anymore in previous apps, and there may be other circumstances that can trigger this.

 

As far as the other problem with stopping containers, just to discount it and you disable SSL access and see if the same behaviour exists?

Link to comment
23 minutes ago, Squid said:

You should update unRaid to 6.5.3.  From the release notes:

  • webgui: bug fix: prevent deleting user template when (letter case) renaming a docker container

If you're changing a name on an app, then 6.5.2 will guarantee that that CA won't be able to find it anymore in previous apps, and there may be other circumstances that can trigger this.

 

It's probably the other circumstances as I'm doing 'trivial' things like changing mappings, not changing names.  Unfortunately, I had to rollback to 6.5.2 as 6.5.3 was causing my W10 VMs to constantly lockup and crash.  I posted about my woes with other users, but the discussion went dead.  

 

 

29 minutes ago, Squid said:

As far as the other problem with stopping containers, just to discount it and you disable SSL access and see if the same behaviour exists?

 

I tried disabling ssl and I just stopped nzbget - took forever (mins) to stop then threw up an error message:

 

89469298_FireShotCapture138-Highlander_Dashboard-http___highlander_Dashboard.thumb.png.421b6d468c9b6de05c2c9800d4bc2a40.png

Link to comment

Only clue I have is this error everytime I try to edit a docker:

 

Jul 27 09:53:54 Highlander nginx: 2018/07/27 09:53:54 [error] 8422#8422: *591973 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.30.100, server: , request: "POST /plugins/dynamix.docker.manager/include/Events.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "1d087a25aac48109ee9a15217a105d14c06e02a6.unraid.net", referrer: "https://1d087a25aac48109ee9a15217a105d14c06e02a6.unraid.net/Dashboard"

 

Link to comment

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...