[Sorta resolved] Server will not shut down

• WebGUI has been hung all day - I clicked to refresh a page this morning, and it still hadn't refreshed when I got home from work this evening.
  
• I do not have access to shares via Windows Explorer
  
• Some of my dockers UIs are responding, some are not
  • CrashPlan has stopped scanning/backing up, though it's attempting to, so it appears that at least some drives have unmounted
    
  • SABnzbd and CouchPotato GUIs are continuing to respond normally
    
  • SickBeard, Libresonic, and COPS GUIs are not responding

  [*]I was able to telnet in and copied /user/var/log/syslog* to /boot  so I'll have them when it comes back up

  [*]From a telnet session, I tried 'powerdown'. I got the message that the 'system is going down NOW', but it didn't.

  [*]From a telnet session, I tried 'poweroff'. it didn't

  [*]Based on this thread: https://lime-technology.com/forum/index.php?topic=31735.0, I executed

  • /usr/local/sbin/powerdown
    
  • The result of this was not a power down, but now the system is refusing to let me log in via telnet
    

I have a monitor & keyboard I can physically plug in to the server if that might make a difference.

I will happily supply the logs once I can get the machine restarted and get them off the flash drive.

Is there anything else I can attempt other than holding down the power button until it shuts off?

• WebGUI has been hung all day - I clicked to refresh a page this morning, and it still hadn't refreshed when I got home from work this evening.
  
• I do not have access to shares via Windows Explorer
  
• Some of my dockers UIs are responding, some are not
  • CrashPlan has stopped scanning/backing up, though it's attempting to, so it appears that at least some drives have unmounted
    
  • SABnzbd and CouchPotato GUIs are continuing to respond normally
    
  • SickBeard, Libresonic, and COPS GUIs are not responding

  [*]I was able to telnet in and copied /user/var/log/syslog* to /boot  so I'll have them when it comes back up

  [*]From a telnet session, I tried 'powerdown'. I got the message that the 'system is going down NOW', but it didn't.

  [*]From a telnet session, I tried 'poweroff'. it didn't

  [*]Based on this thread: https://lime-technology.com/forum/index.php?topic=31735.0, I executed

  • /usr/local/sbin/powerdown
    
  • The result of this was not a power down, but now the system is refusing to let me log in via telnet
    

 

I have a monitor & keyboard I can physically plug in to the server if that might make a difference.

I will happily supply the logs once I can get the machine restarted and get them off the flash drive.

 

Is there anything else I can attempt other than holding down the power button until it shuts off?

Are you running 6.2 or 6.2.1?

If telnet is working, type command 'diagnostics' - that will create a zip file in 'logs' directory on your flash.

oops, need to update that .sig...  I'm running 6.2.1

After posting my initial message, while reading about a couple of other people's travails, my Win Explorer suddenly refreshed with a list of all my shares on the unRAID server. I was able to browse the flash drive for a minute, but, foolishly didn't copy the syslog files to my Win10 box.

I'm getting "Incorrect login" every time I attempt to login via telnet, but I plugged the keyboard into the server, logged in with no problem and am running "diagnostics".  It's taking a while...

As I wait for diagnostics to finish its magic, I notice that the Piwigo docker is still working, as I'm able to browse pictures. That most likely means the mariaDB docker is also still running. Ubooquity is running, but Calibre is not.

Ok, diagnostics has been running for more than 15 minutes now.

Additional tips?

ooh! I can briefly get access to the flash share again, so I've copied syslog & syslog.1 to my desktop and am attaching the .zip.

Unfortunately... syslog.1 is 14MB, and it compresses nicely to 544KB, but that's still over the attachment size limit. After a quick perusal of the log, over 120k lines are mover output from last night, including many, many "out of disk space" errors. I've trimmed the errors out, leaving a comment on where they came from, and am posting up the rest.

Sadly, the monitor at the server indicates "Starting diagnostics collection...". During this brief bout of share access, I can see that it created the log directory, but there weren't yet any files in it.

syslog.zip

As a side note, I'm also having issues with:

• Pre-clear: http://lime-technology.com/forum/index.php?topic=39985.msg506020#msg506020
  
• Sending emails from the server: http://lime-technology.com/forum/index.php?topic=52606.0
  
• A new drive failing: http://lime-technology.com/forum/index.php?topic=52573.0
  

Yeah, it's been a bit of a rough week.  :'(

I'd prefer to not hold down the power button until it shuts off since I'm currently running a drive down and relying on parity to get my data back. Unless, of course, a voice of reason & experience indicates that it should be reasonably safe to do so.

It's the next morning and it's still "starting diagnostics capture..." at the terminal.

Would it be worth attempting to manually shut down the dockers from the command line? If so, what are the commands to do so?

Something has changed at the server, I cannot even make a telnet connection to the server this morning.

I was going to shut down dockers from the console. I tried getting a list of running dockers with docker ps, but got no response to that command, either.

I finally held down the power button until it shut down.

Now to work my way through all the issues. #1 is getting that replacement drive precleared and into the array.