Yeah that makes sense that it would still be encrypted... I had a moment. I think if I went that route I would have generated certs anyways since I could put it behind Traefik with my domain which then gives signed Let's Encrypt keys. Unless the...
Die FDB hat zwar einträge, jedoch noch lange in im Max bereich:
root@pmfw1:~# bridge fdb show br vmbr001 | wc -l
437
root@pmfw1:~# cat /sys/class/net/vmbr001/bridge/hash_max
4096
Das Deaktivieren des floodings wäre vermutlich der Workaround...
With a self signed cert the connection is encrypted, the remote has to accept the cert. it could be easier to set up a man in the middle attack but Proxmox requires the cert thumbprint if self signed.
Hello community,
We need an hand to solve an issue that is driving us crazy. Please help us to find the missing piece of the puzzle :)
We are unable to contact public hosts using HTTP and HTTPS from LXC containers, although DNS and ICMP work...
Die WRMSR-Meldung (Unhandled WRMSR(0x1d9)) ist ein Red Herring, das ist nur ein MSR (IA32_DEBUGCTL) den der Gast schreiben will aber der Host nicht durchreicht. Passiert ständig und ist harmlos.
Was den Reboot angeht: das Corosync-Log sieht...
Hi @Freddy Lege ,
We have seen no issues in our testing with PVE 9.2 and Kernel 7. Perhaps you can expand with more details, i.e.:
- storage vendor you are using
- exact PVE versions (pveversion / pveversion -v
- how the connections were...
Hallo.
Folgender Aufbau bei uns:
Host-Europe-Web-Server (hält die Domain) <---> OPNSense-Firwall <---> PMG (nimmt an) <---> interner Mail-Server (nimmt entgegen und hält die Postfächer).
Bisher dachte ich, dass ich sämtliche DNS-Einstellungen...
Thanks ! - the issue might be the output from your custom check script - could you verify that by - disabling the DEBUG: lines it prints?
or delete them from a copy of your syslog and run the locktracker command with `-i </path/to/copy-of-log>`
Thank you so much for showing me the way.
I've changed it to `my $debug = 0;`
The debug lines were cleared with the following commands, and the service was restarted.
systemctl stop rsyslog syslog.socket
grep -v "DEBUG:" /var/log/syslog >...
Thank you so much for showing me the way.
I've changed it to `my $debug = 0;`
The debug lines were cleared with the following commands, and the service was restarted.
systemctl stop rsyslog syslog.socket
grep -v "DEBUG:" /var/log/syslog >...
Dass das Flooding nicht nur kurz sondern dauerhaft auftritt ist schon komisch. Normalerweise lernt die Bridge die MAC ja beim nächsten Paket vom Client (ACK zurück etc.) sofort wieder. Check mal wie voll die FDB überhaupt ist:
bridge fdb show br...
@shanreich , I'm curious if you've had a chance to try NFTables in newer kernels and/or in PVE 9.x. I'm specifically wondering if the IPSet merge feature in NFTables isn't as broken as when you first worked on it.
Reference bug...
Das Muster ist ziemlich aufschlussreich: VMs laufen weiter, aber Host-Dienste und LXCs nicht. VMs haben ihren eigenen Netzwerk-Stack über QEMU, LXCs teilen sich aber den Host-Kernel. Wenn der Host hängt (Storage, OOM, Kernel-Bug), sind die LXCs...
Thanks for sharing the final outcome. Threads like this are valuable because they highlight how nested virtualization issues can sometimes be related to CPU configuration rather than Hyper-V itself.
One thing I'd recommend for anyone performing...