Recent content by dirks

Thanks. :) Adding "tag=0" removes the network interface. The same is true if I add the trunks parameter without without vlan ids to it, e.g. trunks,tag=0. If I read man qm correctly trunks itself requires vlan ids seperated by ";", like trunks=0;5. I tried net0...

I have a bond binding two nics connected to two switch ports each with an untagged and multiple tagged vlans (802.1q). On top of the bond is a bridge (linux ifupdown2 not ovs). Now I add some VMs. VMs which receive a vlan tag see only the vlan tagged traffic, all good. VMs without a vlan tag see...

I am currently testing integration of AD with PVE including syncing and found the documentation and wiki somewhat incomplete. Especially there were two points can I use bind_dn and similar with AD as well (yes) how to filter a set of users and/or groups. Luckily there are options documented in...

Thank you for your suggestion, but I am using sparse disks on both sides anyway, i.e. while the disk was configured with a max size of 80G the vhdx had <30G and the rbd is now just shy of 20G. Copying was done in <1 min. While qm importdisk took a little bit longer than that (a few minutes) and...

https://www.kernel.org/doc/html/latest/networking/bonding.html And yes bringing an interface down via OS does not necessarily mean the link is seen as down on the switch side, see https://forum.proxmox.com/threads/802-3ad-failover-time-when-bringing-a-slave-interface-down.87390/#post-383527

Since I needed to document the process anyway I thought I share my experience of migrating a domain controller from a hyper-v cluster to a proxmox cluster with ceph storage. I wrote this with these 2 wiki articles as background information...

On our nodes interfaces got renamed from predictable to old naming scheme, i.e. eno1 -> eth0, eno33p0 -> eth3 … But why? No /dev/null links in /etc/systemd/network. No net.ifnames=0 in /proc/cmdline. No … wait, why is there a /etc/systemd/network/99-default.link file. That is new. Contents: #...

Just to finish this up if anyone ever comes across something similar: A link is not necessarily powered down by ip link set $NIC down. We found that across multiple cards from various vendors - and an equivalent for the windows version of disabling of a link. For our Mellanox cards we could...

Just discovered the PBS Prune Simulator while looking at the pve documentation, really nice. Maybe add the option to create not only scheduled backups but also adhoc backups? That would help to visualize the scenario mentioned in the pve-docu:

Thanks for the pointer. What basically happened on my systems is a rename of several ID_NET parameters sudo udevadm info -e |grep ID_NET > interfaces_5.11 # do the same on kernel 5.4 with interfaces_5.4 diff interfaces_5.4 interfaces_5.11 3,4c3,4 < E: ID_NET_NAME_ONBOARD=eno33 < E...

Maybe hardware specific, with kernel pve-5.11 some Mellanox ConnectX-5 cards (onboard and pci) changed their names, which obviously caused havoc on our network config. What does it say on Predictable Network Interface Names again? > Stable interface names when kernels or drivers are...

With 5.11.7-1-pve DFS works fine (pve-kernel-5.11: Installed: 7.0-0+3~bpo10).

It seems at last some nics do not actually take the link down, when told so via ifdown/ip link down. [1], [2] I played around with mlxconfig yesterday to query and set the parameters. It did not resolve the issue yet, as KEEP_ETH_LINK_UP_P1=0 would result in a non functional network connection...

Will probably be fixed with 5.4.112 For reference: - https://cdn.kernel.org/pub/linux/kernel/v5.x/ChangeLog-5.4.112 - https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1923670 Should already work with the pve 5.11 kernel but had no time to test on actual proxmox installation yet (only arch...

The problem was probably introduced somewhere between 5.4.0-67 and 5.4.0-71 as it arrived on my client (Ubuntu 20.04 based) today as well. edit: with 5.4.0-70 it still works.