Proxmox Team - again

[pille99]

again the whole network was down. Ping was possible, Website open, not possible, runs always in timeout.
it happened 3 times already. to be fair, the first time was my fault. the second time the firewall got activated, and now i change the Machine from q35-7.0 to 7.2. the same issue - network was not working proper anymore.
changed it back to 7.0 (nothing else changed) and website is working again.

once again thx, which costs me hours to figure out why my DNS was not working anymore, why the connection per scp was slow like hell

this is a problem for me
1. your updates are not relaiable
2. best option is: NEVER UPDATE, when its running

i will change all VMs back to 7.0.

 

[pille99]

7.1 is working too

 

[dcsapak]

and now i change the Machine from q35-7.0 to 7.2. the same issue - network was not working proper anymore.

hi, was that by any chance a windows guest?

if not what's running inside? (also the complete config would be interesting)

if it is a windows, then that's not surprising, since hardware changes often 'reset' the network settings inside windows
also for other guests it can happen (but most often doesn't, we do not see any problems in our test or production environments AFAIK, so it's not a general issue)

also, why did you manually update the machine version?

 

[pille99]

hi, was that by any chance a windows guest?

if not what's running inside? (also the complete config would be interesting)

if it is a windows, then that's not surprising, since hardware changes often 'reset' the network settings inside windows
also for other guests it can happen (but most often doesn't, we do not see any problems in our test or production environments AFAIK, so it's not a general issue)

also, why did you manually update the machine version?

it was an existing centos 9
i setuped a new ubunto server 22.04 - just to test.
windows didnt see anything,except it cant resolve the network (DNS), and its very slow (i discovered today but i am not sure from which site it comes (copy from windows per scp to linux was im byte range)

 

[pille99]

hi, was that by any chance a windows guest?

if not what's running inside? (also the complete config would be interesting)

if it is a windows, then that's not surprising, since hardware changes often 'reset' the network settings inside windows
also for other guests it can happen (but most often doesn't, we do not see any problems in our test or production environments AFAIK, so it's not a general issue)

also, why did you manually update the machine version?

which config files you want to see ?
do you want any logs from the inside VM ?
i made a copy and with that we can play around

what do you mean with "manuelly update"
how else i should update ?

 

[dcsapak]

which config files you want to see ?

the vm config (qm config ID)

do you want any logs from the inside VM ?

yes that would be helpful

what do you mean with "manuelly update"
how else i should update ?

you wrote:

and now i change the Machine from q35-7.0 to 7.2.

why did you do that?

note that the machine version has not really anything to do with the version of the running qemu binary, and e.g. for windows guests we intentionally set the machine version manually (when creating over the gui) so that the guest does not get (slightly) different hardware each time a qemu update is installed

for linux guest it does not make a problem most of the time, but there was an instance where e.g. the names of the interfaces changed (which was later fixed again)
AFAIK libvirt does the same, hardcoding the machine version into the config on creation, for hardware stability

EDIT: also please the output of the pveversion -v would be interesting

 

[Neobin]

again the whole network was down. Ping was possible, Website open, not possible, runs always in timeout.
it happened 3 times already. to be fair, the first time was my fault. the second time the firewall got activated, and now i change the Machine from q35-7.0 to 7.2. the same issue - network was not working proper anymore.
changed it back to 7.0 (nothing else changed) and website is working again.

once again thx, which costs me hours to figure out why my DNS was not working anymore, why the connection per scp was slow like hell

this is a problem for me
1. your updates are not relaiable
2. best option is: NEVER UPDATE, when its running

i will change all VMs back to 7.0.

Simply stop playing around with things you seemingly have no knowledge of and/or experience with and/or do not know what their consequences are on your production systems/guests and instead set up a test-lab for your shenanigans...

 

[pille99]

Simply stop playing around with things you seemingly have no knowledge of and/or experience with and/or do not know what their consequences are on your production systems/guests and instead set up a test-lab for your shenanigans...

ths for your non_sense input.
what people update and what not its not your business. i am not the only one which update the Operatingsystems and/or Hypervisior software. but at least i read the changelogs and if something looks fishy to me i dont update (and it wasnt anything written what would point me to network issues).

from the dokumentation:
Proxmox VE allows to boot VMs with different firmware and machine types, namely SeaBIOS and OVMF. In most cases you want to switch from the default SeaBIOS to OVMF only if you plan to use PCIe pass through. A VMs Machine Type defines the hardware layout of the VM’s virtual motherboard. You can choose between the default Intel 440FX or the Q35 chipset, which also provides a virtual PCIe bus, and thus may be desired if one wants to pass through PCIe hardware.

usually between sub_versions arejust minor changes like a big fix.
i expect a new "virtual motherboards" with the version release like 6 and 7.

i am sure you do your OS updates - do you know exactly what are every update doing ? no, so, seemingly you dont know what you are talking. but in one thing you are right: should be tested before rollout.
plz keep your opinion about yourself, thx

btw: it doesnt work with a new VM (Ubunto server 22.04) also not, the machine type 7.2. switching to 7.1 worked perfectly. (a blank installed OS, nothing on it)

 

[pille99]

root@hvirt01:~# qm config 101
agent: 1,fstrim_cloned_disks=1,type=virtio
balloon: 2048
boot: order=scsi0
cores: 2
cpu: host
description: CentOS 9%0A%0AInfra server %0ADNS Server, Main DNS%0A%0AIn Progress%0AAnsible, Update server
hotplug: disk,network,memory,cpu
localtime: 1
machine: pc-q35-7.1
memory: 8192
meta: creation-qemu=6.2.0,ctime=1664371726
name: infra
net1: virtio=16E:0C:26:25:96,bridge=Net001
numa: 1
onboot: 1
ostype: l26
parent: auto_HOURLY_230418110001
protection: 1
scsi0: ceph_01:vm-101-disk-0,cache=writeback,discard=on,size=40G
scsihw: virtio-scsi-pci
smbios1: uuid=6dc28f58-7b8a-46ee-95ff-05d8edb50ab1
sockets: 1
startup: order=2
tags: infra.eyonis;infra.service;linux
vmgenid: 257247cb-25fa-404d-a691-32a23470ab6c

 

[fabian]

you are mistaken - for Qemu, 7.1 to 7.2 is just a big of a jump as 7.2 to 8.0. the major version simply gets bumped at the first release of a given year.

 

[dcsapak]

@pille99 would you mind posting the journal/syslog from inside the guest?
i tried reproducing by installing a centos9 vm with machine type 6.2 and changed the machine version to 7.2, but everything worked after (did some basic tests with networking etc.)

 

[Dark26]

Do you have a linux bond in the proxmox configuration?

 

[pille99]

@pille99 would you mind posting the journal/syslog from inside the guest?
i tried reproducing by installing a centos9 vm with machine type 6.2 and changed the machine version to 7.2, but everything worked after (did some basic tests with networking etc.)

as mentioned / an existing centos9 (changed from x.x to 7.2) and a new installed ubuntu server 22.04 (created the VM directly with 7.2). only network configured. will bundle you all the requested logs on the evening.
the result, and this is the result is; 1. performance decrease (ping works, so host is in generally reachable, but open a website is: a, with chrone i can see 10% of the page, the rest wont load, b, firefox runs in timeout, a copy with winscp from an windows to the affected centos results in 100 biytes and disconnects). the windows performance went down as well

 

[pille99]

Do you have a linux bond in the proxmox configuration?

no

auto lo
iface lo inet loopback

auto enp41s0
iface enp41s0 inet manual
#1GB UPLINK

auto enp1s0f0
iface enp1s0f0 inet static
address 10.10.15.10/24
mtu 9000
#10GB SDN

auto enp1s0f1
iface enp1s0f1 inet static
address 10.10.12.10/24
#10GB Corosync

auto enp33s0
iface enp33s0 inet manual
#10GB Ceph

auto vmbr99
iface vmbr99 inet static
address 10.10.10.10/24
bridge-ports enp33s0
bridge-stp off
bridge-fd 0
bridge-vlan-aware yes
bridge-vids 2-4094
#10GB Ceph Public

auto vmbr0
iface vmbr0 inet static
address pub ip
gateway pub gw
bridge-ports enp41s0
bridge-stp off
bridge-fd 0
bridge-vlan-aware yes
bridge-vids 2-4094
#1GB UPLINK Public

auto vmbr99.11
iface vmbr99.11 inet static
address 10.10.11.10/24
#10GB Ceph Cluster

source /etc/network/interfaces.d/*