WebGUI becomes unresponsive when docker hangs


139 posts in this topic Last Reply

Recommended Posts

  • 2 weeks later...
  • Replies 138
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Posted Images

  • 2 weeks later...
  • 2 weeks later...

running rc5 and had this issue this morning. sabnzbd froze when it had a large queue (20gb, which is bigger than my typical). I tried to stop the container. WEbGui is down. Whats the best solution to get back up and running? I read through and everyone cant seem to kill it. hard reset?  :o :'(

Link to post
  • 4 weeks later...

Any more development on this?

I've been running sabnzbd since b14 and only yesterday this happened to me (I'm still on 6.0 final).

I would like to restart and upgrade to the latest 6.0.1, but I was still in the middle of file copy and a couple of preclears. Will do so later tonight when they're done.

Link to post

I've never had a webGui hang, and I don't use the containers listed above, so I'm speaking off the top of my head here, but a couple of observations -

 

* Since JonP's post, all posts have involved sabnzbd

 

* The one specific error reported involves a dependency on TLS, so it may be worthwhile to check with the author of the sabnzbd container as to TLS support.  Perhaps it isn't set up correctly, and that is causing a hang?

 

* WebGui appears to hang, but no recent posts have any clues as to what is actually happening.  Previously, there were clues that it was waiting for something to return/complete, so it wasn't crashed, just stuck waiting for something that should have quickly returned but didn't.

 

* It seems to me to be 2 problems here, sabnzbd is hanging on something, which causes the webGui to hang waiting for it.  So it looks to me as if the sabnzbd author should look into what is causing it to hang, and Tom may want to look into what is causing emhttp to wait, not continue.

Link to post
  • 3 weeks later...

Good day guys,

 

I'm also experiencing the same issue with sabnzdb hanging and freezing webGUI on 6.0.1. For whatever reason, the process hangs and can't be killed through docker commands. Even a kill -9 doesn't work as the process looks defunct:

 

ps -ef | grep sab
nobody    9241  8796  0 Jul30 ?        00:06:19 [sabnzbdplus] <defunct>

 

 

WebGUI stays frozen in attempting to stop the dockers. Only thingn I can do is reboot the server.

 

Any updates on this?

 

[Edit]: Even attempting to reboot the server properly through powerdown proves challenging...

Link to post

I believe I'm having the same issue - I have 6.0.1 - brand new user.

I just installed Docker (and VMs) and configured needo's PlexServer

 

When the server boots, things are great for a while, then after so long - the webGUI will not respond.

I'm not sure what's going on / what's causing it.

 

I found a guide that instructed me to install PhAzE's Plex Plugin as well as the docker container.

 

I also am working on getting a VM up for Windows Server 2012 R2 - on KMS

 

Note: I'm very new to Linux - been a windows guy forever - but I really wanted to learn and work with unRaid - though it had a ton of advantages - just been running into this and it's been super frustrating.

Link to post

unRAID WebGUI keeps locking up, this is pretty much a new build with the latest RC. This has never happened with the prior releases of 6 and I have all the Dockers items such as Plex shutdown as this is all still new to me. The last time it hung all I was trying to stop a VM for Server 2012 and before that all I did was to stop the array. When it crashed stopping the array I waited 12 hours to see if it would recover and  it didn't so the only way to recover was to manually reboot the server. Let me know what you need from me I will be more than happy to get this resolved.

Link to post

Bump.  Same problem here.  I noticed SABNZBD docker was unresponsive, so I issued a stop from the webgui.  Webgui then is unresponsive trying to execute the stop.  I ssh in, and do docker stop or docker kill manually on sabnzbd, and the command just hangs forever.  If I then kill the "/usr/bin/docker stop SABnzbd' process initiated by the webgui, the gui again is responsive.  So it appears to be a docker problem.

 

Basically, my only recourse here is to do a reboot without shutting down array, because I can't take the array down because it hangs on umounting shares, even though "lsof|grep mnt" shows nothing.  Which means then the next time on boot, the array parity needs to be rechecked.  Still running 6.0, and as I recall, this issue only started with this version, but only within last 4 weeks or so. So perhaps it's an sabnzbd issue?

Link to post

I've experienced the same issue multiple times. I've tried narrowing down the dockers to see if one was the culprit, so all are shutdown except for needo's PMS. I turned off sab, couch, etc. My last attempt was not adding anything to the array and just leaving it be for 2 weeks. Unraid was fine until I copied a file onto the share. No issues copying, plex picked up the file, and then the overnight mover caused the hang I believe. This has happened more than once. After the mover, the array was unresponsive via the webGUI.

 

I can telnet into it, but not access the webGUI or shares via SMB. Killing the docker doesn't work, and Powerdown says it will halt and reboot, but nothing. One disk will refuse to unmount. It is the destination disk for the file. I'm not sure if both the mover and dockers are the problem, and plan to test with no dockers running and just adding a file to the array, letting the mover transfer it from the cache drive.

 

I switched to the docker, from the plugins route, as the selling point was if it crashed, the array would stay up. Right now though, I can't kill the docker, and the mover script isn't completing.

 

After a forced reboot (only means of shutting it down), the cache drive still contains the file, as the mover didn't complete. I can manually move the file via "move now", and everything stays working as it should. It only happens when the mover script runs on it's own.

 

I don't think there is a memory issue, as there are no other signs and plenty of free RAM available before this occurs. I have NFS off, and AFP and SMB on. I had none of these issues on 5.0, without a single reboot needed for over 2 years. That was while running sab, couch, sb, and plex via plugins.

 

Anyone have any ideas as to why this is occurring, and does anyone else experience this hanging up after the mover script tries to do its thing?

Link to post

When a container is stuck, you have to kill the main docker process. You can stop the other containers cleanly first if you like.

 

Once you kill the main docker process, the gui should become responsive again. And you can restart docker or reboot server

 

Link to post

I've experienced the same issue multiple times. I've tried narrowing down the dockers to see if one was the culprit, so all are shutdown except for needo's PMS. I turned off sab, couch, etc. My last attempt was not adding anything to the array and just leaving it be for 2 weeks. Unraid was fine until I copied a file onto the share. No issues copying, plex picked up the file, and then the overnight mover caused the hang I believe. This has happened more than once. After the mover, the array was unresponsive via the webGUI.

 

I can telnet into it, but not access the webGUI or shares via SMB. Killing the docker doesn't work, and Powerdown says it will halt and reboot, but nothing. One disk will refuse to unmount. It is the destination disk for the file. I'm not sure if both the mover and dockers are the problem, and plan to test with no dockers running and just adding a file to the array, letting the mover transfer it from the cache drive.

 

I switched to the docker, from the plugins route, as the selling point was if it crashed, the array would stay up. Right now though, I can't kill the docker, and the mover script isn't completing.

 

After a forced reboot (only means of shutting it down), the cache drive still contains the file, as the mover didn't complete. I can manually move the file via "move now", and everything stays working as it should. It only happens when the mover script runs on it's own.

 

I don't think there is a memory issue, as there are no other signs and plenty of free RAM available before this occurs. I have NFS off, and AFP and SMB on. I had none of these issues on 5.0, without a single reboot needed for over 2 years. That was while running sab, couch, sb, and plex via plugins.

 

Anyone have any ideas as to why this is occurring, and does anyone else experience this hanging up after the mover script tries to do its thing?

Next time it happens, from console or telnet type
diagnostics

ignore any errors, then get the diagnostics zip file from your flash and post it.

Link to post

When a container is stuck, you have to kill the main docker process. You can stop the other containers cleanly first if you like.

 

Once you kill the main docker process, the gui should become responsive again. And you can restart docker or reboot server

Unfortunately, I've never been able to reboot cleanly after this issue, because the array does not unmount. It always keeps retrying to unmount the disk share.  I checked that my clients are all powered down and I'm not cd'd into any disk shares (lsof|grep mnt does not return anything). 

Link to post

When a container is stuck, you have to kill the main docker process. You can stop the other containers cleanly first if you like.

 

Once you kill the main docker process, the gui should become responsive again. And you can restart docker or reboot server

Unfortunately, I've never been able to reboot cleanly after this issue, because the array does not unmount. It always keeps retrying to unmount the disk share.  I checked that my clients are all powered down and I'm not cd'd into any disk shares (lsof|grep mnt does not return anything).

 

Some process is keeping the array occupied, perhaps doing ps -ef may help to find which process this may be.

 

Killing the offending process would surrender the array and make the GUI responsive again.

 

Link to post

Same issue, on unRaid 6.1rc6 with Plex Media Server docker and Sabnzbd. Brand new setup, installed a few drives, set up the 2 mentioned dockers, and began transferring about 3TB worth of data. Sometime in the night the transfers from my Windows box failed, loaded up unRaid webui, things seemed fine. Couldnt access the unRaid shares from my windows box, decided to try rebooting unRaid. Stopped array, and things froze when it said it was trying to stop the dockers.

 

Hard rebooted. Have not brought the dockers back up. Will probably just install a windows VM to run my stuff for now. Let me know what I can do in the mean time to help diagnose, if anything.

Link to post

So, I seem to have had some success by doing the following (I have plex, sabnzbd dockers):  Go into your plex server settings, Scheduled Tasks, and adjust your task start and stop times so that there is no overlap with the Unraid mover task time.  For example, when I had all these freezing issues, Plex was running between 2am and 5am, and the mover was set to run sometime around 3am.  I actually changed the mover to run at 5:10am (but you could also adjust the plex times). 

 

This seems to have stopped my sabnzbd docker from becoming unresponsive and having to hard-boot.  So far, been running a couple of weeks with no issues and with heavy download volume.

 

I suspect it's an interaction with Plex's housekeeping and the mover process.  Perhaps this workaround might work for other people too.

Link to post

So, I seem to have had some success by doing the following (I have plex, sabnzbd dockers):  Go into your plex server settings, Scheduled Tasks, and adjust your task start and stop times so that there is no overlap with the Unraid mover task time.  For example, when I had all these freezing issues, Plex was running between 2am and 5am, and the mover was set to run sometime around 3am.  I actually changed the mover to run at 5:10am (but you could also adjust the plex times). 

 

This seems to have stopped my sabnzbd docker from becoming unresponsive and having to hard-boot.  So far, been running a couple of weeks with no issues and with heavy download volume.

 

I suspect it's an interaction with Plex's housekeeping and the mover process.  Perhaps this workaround might work for other people too.

 

Thanks for the update! Had not considered Plex Servers internal house keeping duties. I do wonder what causes the actual hangup though...

Link to post

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.