0

I've noticed every since using my laptop with my Aorus RTX 2080Ti gaming box (egpu w ethernet) with either 20.04 or 20.10 I've been encountering NetworkManager crashes.

What happens during a crash is that the wired connection fails (network icon with questionmark in top right) and I'm unable to disconnect from ethernet or connect to wifi. Restarting the networkmanager is impossible since sudo systemctl restart NetworkManager hangs, and even restarting the computer is impossible without a hard reset.

I've managed to gather the syslog from boot till such a crash, and the problems seem to start after these kernel messages start appearing:

Dec 29 12:33:23 littlebeast whoopsie[1769]: [12:33:23] Cannot reach: https://daisy.ubuntu.com
Dec 29 12:36:23 littlebeast systemd-resolved[992]: Grace period over, resuming full feature set (UDP+EDNS0) for DNS server 8.8.8.8.
Dec 29 12:36:31 littlebeast systemd-resolved[992]: Using degraded feature set UDP instead of UDP+EDNS0 for DNS server 8.8.8.8.
Dec 29 12:36:36 littlebeast systemd-resolved[992]: Using degraded feature set TCP instead of UDP for DNS server 8.8.8.8.
Dec 29 12:47:24 littlebeast systemd[1]: Starting Refresh fwupd metadata and update motd...
Dec 29 12:47:24 littlebeast systemd[1]: fwupd-refresh.service: Succeeded.
Dec 29 12:47:24 littlebeast systemd[1]: Finished Refresh fwupd metadata and update motd.
Dec 29 12:53:19 littlebeast systemd-timesyncd[993]: Timed out waiting for reply from 91.189.89.199:123 (ntp.ubuntu.com).
Dec 29 12:53:30 littlebeast systemd-timesyncd[993]: Timed out waiting for reply from 91.189.89.198:123 (ntp.ubuntu.com).
Dec 29 12:53:40 littlebeast systemd-timesyncd[993]: Timed out waiting for reply from 91.189.91.157:123 (ntp.ubuntu.com).

I've included the full logs just in case the problem starts somewhere else: https://pastebin.com/0RDyffsB.

I'd love some tips on how to start tackling this issue, but since I'm unable to find anything relevant in the syslog, I'm not sure where else to start looking.

EDIT: Updated logs with something that looks more related to the issue. Seems to be a DNS problem.

Also, the output from journalctl -u systemd.resolved:

dec 29 12:36:23 littlebeast systemd-resolved[992]: Grace period over, resuming full feature set (UDP+EDNS0) for DNS server 8.8.8.8.
dec 29 12:36:31 littlebeast systemd-resolved[992]: Using degraded feature set UDP instead of UDP+EDNS0 for DNS server 8.8.8.8.
dec 29 12:36:36 littlebeast systemd-resolved[992]: Using degraded feature set TCP instead of UDP for DNS server 8.8.8.8.
dec 29 13:59:37 littlebeast systemd-resolved[992]: Server returned error NXDOMAIN, mitigating potential DNS violation DVE-2018-0001, retrying transaction with reduced feature level UDP.
dec 29 13:59:37 littlebeast systemd-resolved[992]: Server returned error NXDOMAIN, mitigating potential DNS violation DVE-2018-0001, retrying transaction with reduced feature level UDP.
BallzofFury
  • 101
  • 1
    What are you generally doing when this issue occurs? Anything that might consume a lot of I/O or memory? How much RAM does the notebook have? –  Dec 28 '20 at 12:21
  • That's the hard part, I can't reliably reproduce it. I've got 32Gb of memory, so I'd be surprised if that had anything to with it. – BallzofFury Dec 28 '20 at 12:48
  • And after some googling, I feel the above kernel messages might not be related to my problem after all – BallzofFury Dec 28 '20 at 12:48
  • Memory is meant to be used. I, too, have 32GB in my workstations and often operate at around 70% use during the day. Mind you, I do a lot of programming and database stuff, so this is to be expected. The I/O, however, could be an area to look into. If the storage device is saturated when the OS tries to flush its cache, this error can appear. I’ve seen it when using SMR hard drives, for example. If the storage cannot keep up, it could point to a poor driver or something more complex. –  Dec 28 '20 at 14:25

1 Answers1

0

Looks like this was the issue:

"Using degraded feature set (UDP) for DNS server 8.8.8.8" on new ubuntu installation on virtualbox over windows10

The problem was resolved by changing the symlink

/etc/resolv.conf -> ../run/systemd/resolve/stub-resolv.conf

to

/etc/resolv.conf -> /run/systemd/resolve/resolv.conf

BallzofFury
  • 101