WARNING! Upgrade to 7 network problem, do not upgrade!!!

Kordian

Member
Mar 31, 2018
26
0
21
50
Well, since I am after 6 hours of fighting with getting my production system working, first of all many thanks to the whole Proxmox team for screwing the upgrade process up and not informing users of it! I hope karma will thank you better than I can.
The story: 2 days ago I decided to upgrade my proxmox from 6.4 to 7.2. I have a test system, so went for upgrade there first. Followed the instructions and actually the upgrade process went with no problems. Problems occurred afterwards. First of all, the network stopped working. I have Dell R620 with 4 port network card and bridge. The ip link showed the eno1 eno2... correctly, all up, but vmbr0 down. Tried few things, did not find much in internet, however as the ip link started showing a comic altname by every link (enp1s0f0 etc), I replaced the eno1 names with these alternate names in /etc/network/interfaces and strangely enough (ip command still returned the old names) the network was back.
I shut the test system down and went to sleep.
The next day was the production system upgrade time. Backed up all vms, started the test system and copied the backups there (test system still working ok) and did the upgrade to the production system. Again, got no network afterwards, so wanted to do the exact steps, as the day before. To my surprise, after changing the names to enp1s0f0 in /etc/network/interfaces and restarting the service, the ip returned suddenly 42 interfaces (!!!!!!!!!!!!!). The network stopped working, and, strangely enough, also the test system network went down!!! Here just to mention, I use fast ip, so no dhcp could mess up. The next 4 hours (till midnight!!!, thank you proxmox guys!) I tried to get my prod system back to life. What is worth mentioning, with vmbr0 down and all other interfaces up, there was absolutely no error in the logs! The network started, status active, vmbr0 down.
At the end I decided to resign from bridge and have at least one working interface, so adjusted the settings and finally could start proxmox gui.
And here another surprise - the gui interface was actually then more helpful, than the shell, which for me happened the very first time in my entire linux carrier!!!
Gui showed the 4 eno interfaces, but also three enp1s0f0-type ones (so in total 7). Removing the enp1s0f0 and clinking on apply brought the solution, as a message stating "you need ifupdown2 in order to restart" ensured, that there was the old ifupdown still present and the upgrade process skipped this CRUCIAL installation.
After installing the ifupdown2, everything went back to normal on both systems.
So once again, thank you proxmox for your lousy work! Now get back to work and do it properly this time. Or at least inform people of that!
To all the others: do not perform upgrade unless you have physical or kvm access to the systems!!!!!
 

Dunuin

Famous Member
Jun 30, 2020
9,039
2,321
156
Germany
first of all many thanks to the whole Proxmox team for screwing the upgrade process up and not informing users of it!

Did you read the change log with its "Known Issues" for PVE 7.0 before updating?:
Known Issues

  • Network: Due to the updated systemd version, and for most upgrades, the newer kernel version (5.4 to 5.11), some network interfaces might change upon reboot:
    • Some may change their name. For example, due to newly supported functions, a change from enp33s0f0 to enp33s0f0np0 could occur.We observed such changes with high-speed Mellanox models.
    • Bridge MAC address selection has changed in Debian Bullseye - it is now generated based on the interface name and the machine-id (5) of the system.
    Systems installed using the Proxmox VE 4.0 to 5.4 ISO may have a non-unique machine-id. These systems will have their machine-id re-generated automatically on upgrade, to avoid a potentially duplicated bridge MAC.
If you do the upgrade remotely, make sure you have a backup method of connecting to the host (for example, IPMI/iKVM, tiny-pilot, another network accessible by a cluster node, or physical access), in case the network used for SSH access becomes unreachable, due to the network failing to come up after a reboot.
 

Kordian

Member
Mar 31, 2018
26
0
21
50
Did you read the change log with its "Known Issues" for PVE 7.0 before updating?:
Yes, I read it. No word of ifupdown problems afterwards. Actually before the upgrades I added the hwaddress to the config, as suggested.
Still it all went wrong!
 

danielb

Well-Known Member
Jun 1, 2018
212
54
48
36
Bordeaux (france)
Before ranting, please take a closer look at the license under which PVE is released (and look for the warranties). Keep in mind a lot (most ?) users were able to handle the upgrade with no issue (including myself, on several servers). Major upgrades of any system requires basic administration skills. If you don't have them, you should consider buying a license to get support from the proxmox team.
 
  • Like
Reactions: Neobin

Kordian

Member
Mar 31, 2018
26
0
21
50
Before ranting, please take a closer look at the license under which PVE is released (and look for the warranties). Keep in mind a lot (most ?) users were able to handle the upgrade with no issue (including myself, on several servers). Major upgrades of any system requires basic administration skills. If you don't have them, you should consider buying a license to get support from the proxmox team.
What does this have to do with the problem, or the fact that the upgrade does not upgrade a critical component? If you have nothing to do, keep on doing nothing, instead of posting useless comments. Thank you in advance.
 

Pifouney

Member
Oct 17, 2021
148
12
18
33
Hey :)

First at all, without production ready cluster, you need to investigate more deeply yours OS-upgrade. You cannot drop on a staff team a personnal error. If you haven't tried more specificially your test area, it's not the problem of the dev team. :)

Another tips, blame all on this forum is a perfectly joke... Staff members are here when you do not speak to them like a dogs ;) (and some experimented users like @denuin come for help other when they have an idea. Your're just purging your hate here. That will doesn't help :)

In other word, your the only one responsive of your working overtime :)
 
Last edited:

About

The Proxmox community has been around for many years and offers help and support for Proxmox VE, Proxmox Backup Server, and Proxmox Mail Gateway.
We think our community is one of the best thanks to people like you!

Get your subscription!

The Proxmox team works very hard to make sure you are running the best software and getting stable updates and security enhancements, as well as quick enterprise support. Tens of thousands of happy customers have a Proxmox subscription. Get your own in 60 seconds.

Buy now!