Network freezes after recent distro upgrades

After recent distro/kernel upgrades: network may freeze after some some period of work: interface is up but no response to any network requests. Wicked is used. Tried to switch to NetworkManager - same thing. Some post-install scripts are very incorrect. Force to reboot or restart network (latter sometimes does not help). How to fix?

Welcome.

On how to fix, please provide more details :wink:

Is ping working? Try “ping forums.opensuse.org” and if that is not working try “ping 195.135.221.161” and if that does not work try “tracepath -m 8 195.135.221.161” and share the output of that command Using Code Tags Around Your Paste

No, nothing works - no responses on any networks requests. On that moment in gkrellm only transmit packages are shown - no receive. Now network works fine, can’t provide more details.

Had to search what GKrellM means, but clear, no Rx counters are increasing.

Can you check the journal around the time the network goes down?

Yep!

Ok, but how: command, path?

To see the journal of the last boot use:

sudo journalctl -b

Scroll to the time when the netwerk did go down and have a look at the messages.
If you need help, post the interesting part of the log here Using Code Tags Around Your Paste

Shit, after today’s upgrade network disappeared completely: regardless it is wicked or NM. Guys, could You test your post-install scripts better please!!!

journal contains short one:

Dec 19 23:25:00 User kernel: 8139too 0000:04:07.0 enp4s7: link down

However, unlike wicked, NM restored network in 20 seconds:

Dec 19 23:25:00 User kernel: 8139too 0000:04:07.0 enp4s7: link down
Dec 19 23:25:06 User NetworkManager[1239]: <info>  [1671485106.4230] device (enp4s7): state change: activated -> unavailable (reason 'carrier-changed', sys-iface-state: 'managed')
Dec 19 23:25:06 User NetworkManager[1239]: <info>  [1671485106.4233] dhcp4 (enp4s7): canceled DHCP transaction
Dec 19 23:25:06 User NetworkManager[1239]: <info>  [1671485106.4233] dhcp4 (enp4s7): activation: beginning transaction (timeout in 45 seconds)
Dec 19 23:25:06 User NetworkManager[1239]: <info>  [1671485106.4233] dhcp4 (enp4s7): state changed no lease
Dec 19 23:25:06 User avahi-daemon[1014]: Withdrawing address record for 192.168.0.110 on enp4s7.
Dec 19 23:25:06 User avahi-daemon[1014]: Leaving mDNS multicast group on interface enp4s7.IPv4 with address 192.168.0.110.
Dec 19 23:25:06 User avahi-daemon[1014]: Interface enp4s7.IPv4 no longer relevant for mDNS.
Dec 19 23:25:06 User avahi-daemon[1014]: Withdrawing address record for fe80::52c0:75ad:4033:60d0 on enp4s7.
Dec 19 23:25:06 User avahi-daemon[1014]: Leaving mDNS multicast group on interface enp4s7.IPv6 with address fe80::52c0:75ad:4033:60d0.
Dec 19 23:25:06 User avahi-daemon[1014]: Interface enp4s7.IPv6 no longer relevant for mDNS.
Dec 19 23:25:22 User kernel: 8139too 0000:04:07.0 enp4s7: link up, 100Mbps, full-duplex, lpa 0x41E1
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.2791] device (enp4s7): carrier: link connected
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.2796] device (enp4s7): state change: unavailable -> disconnected (reason 'carrier-changed', sys-iface-state: 'managed')
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.2809] policy: auto-activating connection 'enp4s7' (8b97d285-fc1b-3b9d-9b28-3b4f2cd4a74e)
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.2819] device (enp4s7): Activation: starting connection 'enp4s7' (8b97d285-fc1b-3b9d-9b28-3b4f2cd4a74e)
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.2821] device (enp4s7): state change: disconnected -> prepare (reason 'none', sys-iface-state: 'managed')
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.2834] device (enp4s7): state change: prepare -> config (reason 'none', sys-iface-state: 'managed')
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.3353] device (enp4s7): state change: config -> ip-config (reason 'none', sys-iface-state: 'managed')
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.3362] dhcp4 (enp4s7): activation: beginning transaction (timeout in 45 seconds)
Dec 19 23:25:22 User avahi-daemon[1014]: Joining mDNS multicast group on interface enp4s7.IPv6 with address fe80::52c0:75ad:4033:60d0.
Dec 19 23:25:22 User avahi-daemon[1014]: New relevant interface enp4s7.IPv6 for mDNS.
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.3384] dhcp4 (enp4s7): state changed new lease, address=192.168.0.110
Dec 19 23:25:22 User avahi-daemon[1014]: Registering new address record for fe80::52c0:75ad:4033:60d0 on enp4s7.*.
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.3398] policy: set 'enp4s7' (enp4s7) as default for IPv4 routing and DNS
Dec 19 23:25:22 User avahi-daemon[1014]: Joining mDNS multicast group on interface enp4s7.IPv4 with address 192.168.0.110.
Dec 19 23:25:22 User avahi-daemon[1014]: New relevant interface enp4s7.IPv4 for mDNS.
Dec 19 23:25:22 User avahi-daemon[1014]: Registering new address record for 192.168.0.110 on enp4s7.IPv4.
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.5850] device (enp4s7): state change: ip-config -> ip-check (reason 'none', sys-iface-state: 'managed')
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.6268] device (enp4s7): state change: ip-check -> secondaries (reason 'none', sys-iface-state: 'managed')
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.6270] device (enp4s7): state change: secondaries -> activated (reason 'none', sys-iface-state: 'managed')
Dec 19 23:25:22 User NetworkManager[1239]: <info>  [1671485122.6279] device (enp4s7): Activation: successful, device activated.

Good you found the culprit in the log.

reason ‘carrier-changed’

That means I think that the link low-level “died”.

enp4s7 looks like a plain cabled Ethernet interface, true?

What is on the other side of the cable and can it be that some other electric equipment is now and then putting interference on the Ethernet cable?

Yes

Tp-Link Wi-Fi router on the other side. And I have it for a long time, works fine, no other interferences.

Can you have a look at the log of the TP-Link? Was it maybe doing a restart?

I would still have a better look at the cable, even trying to replace it for a test. If you search for “carrier-changed” on the Internet you will find that at least in some cases that was the problem.

No logs for that day.

No possible: it is inside the wall. And I had not problems with it like I said.