18.04.1 server loses network connectivity, brings down the network

[steve130]

I have a home server running 18.04.1. It is a simple server that I use to run PLEX, Samba, and a few other applications. I don't use the server for DNS or DHCP, that is done by my router. The server does have a static IP address.

About once a day, (not always at the same time though) the server loses network connectivity. When this happens, I am not able to ssh into the server. Additionally, all other devices lose internet access, like they can't communicate with the router. As soon as I unplug the ethernet cable from the server, all the other devices get internet access back.

Based on this, I'm reasonably sure this has to do with the server and its network settings. Has anyone encountered a problem like this? Which logs should I be looking in for the cause? The network settings look fine, and it does work most of the time. The problem is not specific to 18.04 as I had the same problem under 16.04.

Any help troubleshooting this issue would be greatly appreciated.

[darkod]

Post the netplan yaml file content please.

And I would suggest to look at /var/log/syslog immediately after you regain connectivity. Look at the last lines of the log before connectivity was lost and it might show you some info.

It is very strange a failure in the server to bring the whole network down. I was about to suspect the router first, although you claim disconnecting the server network cable brings things to normal.

Usually what ever happens on the server itself, the rest of the LAN should work without problems. Unless you have some really special setup that prevents it...

[steve130]

Thanks for your reply.There is nothing in the logs to indicate an issue. The server never regains connectivity. I have to restart it when the issue occurs. I agree that the LAN should continue on if the server has networking issues. I'm a Ubuntu server amateur and definitely didn't intend to create a special setup, but may have inadvertently.I don't seem to have a netplan yaml. There is nothing in /etc/netplan. When I try to check my network settings I get:

steve@server:~$ sudo netplan ip leases eno1
Traceback (most recent call last):
File "/usr/sbin/netplan", line 23, in netplan.main()
File "/usr/share/netplan/netplan/cli/core.py", line 50, in main self.run_command()
File "/usr/share/netplan/netplan/cli/utils.py", line 130, in run_command self.func()
File "/usr/share/netplan/netplan/cli/commands/ip.py", line 56, in run self.run_command()
File "/usr/share/netplan/netplan/cli/utils.py", line 130, in run_command self.func()
File "/usr/share/netplan/netplan/cli/commands/ip.py", line 75, in run self.run_command()
File "/usr/share/netplan/netplan/cli/utils.py", line 130, in run_command self.func()
File "/usr/share/netplan/netplan/cli/commands/ip.py", line 146, in command_ip_leases
key, value = keyvalue.strip().split('=')
ValueError: not enough values to unpack (expected 2, got 1)

[darkod]

Is this a new clean install of 18.04 or upgrade from 16.04? A new install would have netplan config in a yaml file (I just did one last weekend ).

Can you attach monitor and keyboard to the server when the issue happens to see what you get on the screen?

Try things like login, check logs, ping the router, then 8.8.8.8 and google.com...

[steve130]

It is an upgrade from 16.04. The issue happened in 16.04 too though. I'll dig up a keyboard & monitor to see what is going on when the issue presents itself again.

[steve130]

I believe it is an issue with Bind9, which I don't think I even need. At the same time I lost connectivity across the network, the log shows:

server named[1682]: _default: sending trust-anchor-telemetry query '_ta-4a5c-4f66/NULL'

I don't know exactly what that command is, but I know that 'named' belongs to Bind9. I've disabled it and restarted to see if the server becomes more stable. Any idea what a trust-anchor-telemetry query is?