TrueNAS Mini inaccessible after software update

ZFS Noob

Contributor
Joined
Nov 27, 2013
Messages
129
So, I purchased a Mini in February, and I've had it sitting idle for 45 days or so. It's been a snapshot target for my other Mini, but I was mostly letting it burn in.

Today I figured it'd likely been long enough, so I logged in and went ahead with the software update. I think it was from 13.0-U3 to U4.

Anyway, the download completed, it rebooted as part of the install, and it's sitting there with all the shiny lights on, but it's not accessible via network interface.

I've connected to the IPMI port, and it tells me the web interface is not available. Rebooting it came up with the same status - sitting there, "web access could not be accessed."

It looks to me like the machine has forgotten its IP address, as it doesn't show in an ipconfig from the command line, and I get "no route to host" when I try to ping the network address it's trying to listen on.

Is the next course of action to just reset the IP address from IPMI, or is there something more complex that might have broken that I'm not thinking about?

Edited to add: I had both 10G ports set up in a LAGG failover configuration. Maybe this was non-standard enough that it caused the problem.
 
Last edited:

HoneyBadger

actually does care
Administrator
Moderator
iXsystems
Joined
Feb 6, 2014
Messages
5,112
I would expect network oddities going from CORE to SCALE because of the change in network interface naming convention, but not for a point-release upgrade within the family.

As long as you didn't configure anything manually from a command-line, it should have preserved the network configuration through the upgrade. You could connect to IPMI, reboot, and choose a prior boot environment (the one pre-upgrade) from the GRUB bootloader menu to see if that restores access as well.

ifconfig might provide some insight to see if your lagg interface was removed or misconfigured somehow.

You could also open the shell and do /usr/local/bin/freenas-debug -A which will generate a debug dump to /var/tmp/fndebug - then zip it up using tar -cvzf /mnt/yourpoolname/datasetname/debugfile.tar.gz /var/tmp/fndebug - that way you've captured whatever weirdness is presently occurring, and you can restore access before investigating further.

/var/tmp/fndebug/Network/dump.txt is probably the most relevant information, since you're having network issues.
 

ZFS Noob

Contributor
Joined
Nov 27, 2013
Messages
129
Well, it's something about the network condig. Re-created the LAGG and it thought the network interface was back up. Struggling with that now - I've always had a rough time configuring LAGG on these things.

As it is I got it to answer on a single network interface, then went in and configured LAGG and ran the network settings change for 60 seconds....10 minutes ago.

I'll keep creating/configuring/deleting the LAGG interface until I do it right by holding my tongue in just the right position, then it'll work again.

Until update, maybe.
 

ZFS Noob

Contributor
Joined
Nov 27, 2013
Messages
129
Well, after hours of messing with this trying to get the configuration just right...

I unplugged both network ports, and plugged them back in, and *poof* the failover LAGG was talking on the network.

Weird. Leaving this in case someone else runs into this in the future.
 

Constantin

Vampire Pig
Joined
May 19, 2017
Messages
1,829
I found failover LAGG to be an unending source of unhappiness for my c2750d4i based mini XL with its iX-branded but Chelsio-under-the-covers 10GbE card as well. I gave up on failover LAGG.

My current motherboard was configured for a while as a simple bridge (2x 1GbE, 2x 10GbE) but now just runs over one 10GbE fiber interface. Super solid and in a home setting easy enough to fix, should it ever break.
 
Top