(SOLVED) Server Inaccessible after Upgrade to 6.9.0

[kdmurray]

I did the upgrade through the web console to Unraid 6.9 and when the reboot completed the system seems inaccessible.

 

 - Web GUI is not found

 - Docker services are not found

 - System does not respond to ping/tracert

 

When I check the local console, I am able to log in to the system, and from the number / names of active processes it seems that everything has started up. Docker is running, docker services appear to be running, ps shows what looks to be a full slate of processes running.

 

I've rebooted in safe mode, no change.

 

When I try to boot to the GUI mode, the boot sequence completes and I end up on a black screen with a cursor blinking. It gets no further. It's still responsive, because I can reboot it (ctrl-alt-del) and it does a 'graceful' shutdown and restart, but the GUI never appears.

 

At boot, it's showing the assigned IP (10.10.10.38; static DHCP allocation on the router), and the IP is assigned to br0 -- which I think is correct. When I try to ping something from the unrade system console, it's sending those pings from 10.10.10.201 which would be a valid IP but isn't being assigned from the DHCP pool.

At this point I'm a bit at a loss for where to even look for a solution, or which logs to check. There don't appear to be any obvious errors in syslog, but I admit that I'm a bit out of my depth.

Edited March 6, 2021 by kdmurray

[JorgeB]

Type "diagnostics" in the console then attach here.

[kdmurray]

jupiter-diagnostics-20210305-1919.zip

 

Attached diagnostics file.

[kdmurray]

Additional information / testing:

 

• If I try to ping an internet IP (eg 8.8.8.8) I get ping responses.
• If I try to ping an internet hostname (eg google.com) I get "Name or service not known"
• If I try to ping a LAN IP (eg. 10.10.10.1) I get: "Destination Host Unreachable" -- this is telling because that IP is the Gateway / DNS for the network
• If I manually edit resolv.conf to point to 8.8.8.8 I get ping responses from Internet hostnames (eg google.com; cheese.com)

 

All other devices on my LAN are still using 10.10.10.1 as their default gateway/DNS without incident. Multiple platforms and over multiple days since I attempted the upgrade on Wednesday night.

 

I'm guessing there's some kind of network customization I must have done to support some Docker or VM that I had running prior to the upgrade. Not sure where to look from the command line though.

[kdmurray]

Update - Things are working, but still not sure what went wrong.

 

I found that the Wireguard interface (wg0) was set to auto-start. By removing that from the /boot/config/wireguard/autostart file everything came back up as normal.

 

I'm still not sure what would have changed as a result of the upgrade, or whether this would have happened on restart regardless -- the upgrade was the first system reboot in more than 7 months, and the first since WireGuard was initially installed around year-end.

 

I'm marking this as resolved, but if anyone has any insights as to why WireGuard would cause this, I'd love to better understand.