My server went offline last night.  The console looked fine but I couldn't ping it.  Restarted and it booted and loaded the array fine but assigned itself a 192.168xxx.xxx address as opposed to me LAN's 10.10.x.x range.

 

Fix common problems gave me "Unassigned Devices Plus not installed", something I haven't seen before and something I can't find a reference too.  Not sure if it's related.

 

I stopped the Docker service to work on the network but now I can't get the docker to start again.  When I try to start it doesn't like the paths to the docker img and appdata directory.  They are on cache.  docker image path drop down shows cache then the docker folder.  The appdata dropdown doesn't show my cache directory.  Cache drive (500Gb SSD) is up, running, I can access it from the server CLI.

 

But back to the network.  I assigned a static IP and it takes it fine.  Last night I was able to ping 1.1.1.1 from the server cli fine but from a network device I couldn't ping the server.  I tried various things like swapping NICs over (i had a disabled NIC in the server) and various other things I don't remember now.

 

I just noticed (the next morning) the when I select static IP, it defaults to a /16 and the default gateway says "link" (no quotes).  Changed it to /24 and made sure the right gateway was entered the server has finally come up and I can reach it from other network devices.

 

I still have issues with the docker service.  It won't start.  It doesn't seem able to find the paths for the docker.img (now deleted as part of last nights fault finding) nor the cache drive appdata directory.  In fact on the drive/share dropdown it doesn't even show cache anymore.

Log files attached.  When the server became inaccessible (around 9pm), the logs from the retart straight after the issue, and the logs and diags from now, server running but docker service not starting.  I need to get the docker running as it's affecting the home Wifi (unifi controller runs as a container).

tdm-diagnostics-20200108-0832.zip syslog-20200107-223953.txt syslog-20200107-222728.txt

Edited January 8, 20206 yr by dalben

Any advice at all?  I just cant start the docker service no matter what I do.  I managed to fire up the unifi controller on the raspberry Pi I use for Pihole so home network connectivity is back, but all the other containers (Kodi etc) are down.

Something weird going on as you're missing the user0 mount point, try starting in safe mode.

I didn't try that but I managed to get it up and going by editing the docker.cfg file and changing "No" to "Yes".  Restarted the server and the docker service is running.

 

If there is something specific you want me to check when starting in Maint mode, let me know, happy to run some tests or captures info.

Just weird /mnt/user0 was missing, if all is working well now now it doesn't matter.