On Mon, Jul 8, 2024 at 2:52 PM Jeffrey Ross <jeff@xxxxxxxxxx> wrote:
system is Fedora 40 with all packages updated, system is not running a
GUI, CLI only.
Which Fedora spin is it (server, workstation, etc.)? Is Fedora fully updated
(so you aren't chasing a solved problem)?
IPv6 works fine utilizing SLAAC and I can stay connected for hours, I
can move files around etc with no issues
IPv4 works briefly then freezes, I connect via SSH (on the local
network), I get connected and then maybe 30 seconds later the system
stops responding to IPv4, I can open up another IPv4 session and that
too will work for 30 seconds or so before it freezes. Note I do not get
a session closed it just stops responding.
I've used "netstat" to study this sort of issue in the past, but the current
tool is "ss". Can you run diagnostics on the system at the other side
of a connection?
I recall one case where connections were being dropped after a short
time that was ultimately traced to a misconfigured router sending
some traffic through some intrusion detection system. Network admins
at the remote site were able to see where the connection was being
dropped (my end was behind a firewall and required raising the issue
through a cumbersome process via chain of command).
Can you arrange to run "ss" on both ends of a session --
maybe start with ssh using IPv6, then use "ss" to see what happens
with IPv4 sessions.
journalctl may provide insights. You can use "journalctl -b" then search
for the lines where the interface is enabled, something like:
r8169 0000:01:00.0 enp1s0: renamed from eth0
but with "enp6s0f0".
Then search for mention of the interface around when the interface went offline.
I am confident it is not a network infrastructure issue nor a hardware
issue on the server, I strongly suspect there are some conflicting
configurations going on with the server and think the best way forward
would be to clean up any old network configurations and redeploy the
network configurations clean.
Makes sense. A fresh install of the same Fedora spin on another drive
could be used to compare with the broken configuration.
This server had been upgraded many times with the new releases of Fedora
over the years and when the static IP addressing was first setup I
believe the configuration was in /etc/sysconfig/network-scripts, the
directory still exists with the scripts but no longer contains my
interface configurations.
Current command-line Fedora network configuration uses nmcli:
This system is using a static IP address, has 4 ethernet interfaces
where only 1 has ever been in use (enp6s0f0).
Suggestions on where to start cleaning up the configuration or looking
for configuration conflicts?
--
George N. White III
-- _______________________________________________ users mailing list -- users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to users-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/users@xxxxxxxxxxxxxxxxxxxxxxx Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue
