Downgrade 7.0 to 6.4

[vactis]

Is it possible to downgrade Proxmox VE 7.0 to 6.4 without too many problems? I've been running version 6 for quite some time and it has been very stable but since I upgraded to 7.0 beta I get random kernel panics after an uptime of 1-2 days.

 

[tom]

s it possible to downgrade Proxmox VE 7.0 to 6.4 without too many problems?

No, you need to re-install. Debian/Proxmox do not provide a clean downgrade path.

But please provide more details about your issues with 7.0 beta, send hardware details.

 

[vactis]

Here is an output of lshw:

H/W path Device Class Description
======================================================
system PRIMERGY TX100 S3P (S26361-Kxxx-Vyyy)
/0 bus D3009-B1
/0/0 memory 64KiB BIOS
/0/28 memory 1MiB L2 cache
/0/29 memory 256KiB L1 cache
/0/2a memory 8MiB L3 cache
/0/2b memory 24GiB System Memory
/0/2b/0 memory 8GiB DIMM DDR3 Synchronous 1067 MHz (0.9 ns)
/0/2b/1 memory 4GiB DIMM DDR3 Synchronous 1067 MHz (0.9 ns)
/0/2b/2 memory 8GiB DIMM DDR3 Synchronous 1067 MHz (0.9 ns)
/0/2b/3 memory 4GiB DIMM DDR3 Synchronous 1067 MHz (0.9 ns)
/0/2c processor Intel(R) Xeon(R) CPU E3-1220 V2 @ 3.10GHz
/0/100 bridge Xeon E3-1200 v2/Ivy Bridge DRAM Controller
/0/100/19 enp0s25 network 82579LM Gigabit Network Connection (Lewisville)
/0/100/1a bus 6 Series/C200 Series Chipset Family USB Enhanced Host Controller #2
/0/100/1a/1 usb1 bus EHCI Host Controller
/0/100/1a/1/1 bus Integrated Rate Matching Hub
/0/100/1a/1/1/1 input Back-UPS RS 900G FW:879.L4 .I USB FW:L4
/0/100/1a/1/1/4 communication Aeotec Z-Stick Gen5 (ZW090) - UZB
/0/100/1c bridge 6 Series/C200 Series Chipset Family PCI Express Root Port 1
/0/100/1c.6 bridge 6 Series/C200 Series Chipset Family PCI Express Root Port 7
/0/100/1c.6/0 enp2s0 network 82574L Gigabit Network Connection
/0/100/1d bus 6 Series/C200 Series Chipset Family USB Enhanced Host Controller #1
/0/100/1d/1 usb2 bus EHCI Host Controller
/0/100/1d/1/1 bus Integrated Rate Matching Hub
/0/100/1d/1/1/5 communication TI CC2531 USB CDC
/0/100/1e bridge 82801 PCI Bridge
/0/100/1e/7 display ES1000
/0/100/1f bridge C202 Chipset LPC Controller
/0/100/1f.2 scsi0 storage 6 Series/C200 Series Chipset Family 6 port Desktop SATA AHCI Controller
/0/100/1f.2/0 /dev/sda disk 2TB ST2000VN000-1HJ1
/0/100/1f.2/0/1 /dev/sda1 volume 1863GiB EFI partition
/0/100/1f.2/1 /dev/sdb disk 2TB ST2000VN000-1HJ1
/0/100/1f.2/1/1 /dev/sdb1 volume 1863GiB EFI partition
/0/100/1f.2/2 /dev/sdc disk 1TB WDC WD10EADS-00M
/0/100/1f.2/2/1 /dev/sdc1 volume 931GiB EFI partition
/0/100/1f.2/3 /dev/sdd disk 1TB WDC WD10EADS-00M
/0/100/1f.2/3/1 /dev/sdd1 volume 931GiB EFI partition
/0/100/1f.2/4 /dev/sde disk 500GB ST500DM002-1BD14
/0/100/1f.2/4/1 /dev/sde1 volume 465GiB EFI partition
/0/100/1f.2/5 /dev/sdf disk 500GB ST500DM002-1BD14
/0/100/1f.2/5/1 /dev/sdf1 volume 1006KiB BIOS Boot partition
/0/100/1f.2/5/2 /dev/sdf2 volume 511MiB Windows FAT volume
/0/100/1f.2/5/3 /dev/sdf3 volume 465GiB LVM Physical Volume
/0/100/1f.3 bus 6 Series/C200 Series Chipset Family SMBus Controller
/0/1 system PnP device PNP0c01
/0/2 system PnP device PNP0c02
/0/3 system PnP device PNP0b00
/0/4 generic PnP device INT3f0d
/0/5 system PnP device PNP0c02
/0/6 communication PnP device PNP0501
/0/7 system PnP device PNP0c02
/0/8 system PnP device PNP0c02
/1 power To Be Filled By O.E.M.
/2 vmbr0 network Ethernet interface
/3 vmbr1 network Ethernet interface
/4 tap101i0 network Ethernet interface
/5 fwbr101i0 network Ethernet interface
/6 fwpr101p0 network Ethernet interface
/7 fwln101i0 network Ethernet interface

 

[Dunuin]

I would bet syslog with the kernel panics would be useful too.

 

[vactis]

I gave up and reinstalled 6.4 and now it's working fine again. The only thing I noticed when I was still at 7.0 was that after I started a btrfs scrub it would only take 5-10 minutes for the kernel panic to occur.

 

[alexskysilk]

So the problem is the kernel, not pve specifically. you need to update your bios.

 

[hvisage]

Why btrfs and not ZFS?

 

[PeterRoux]

So the problem is the kernel, not pve specifically. you need to update your bios.

I think there is an issue with virtio and amd cpus. I too was facing loads of crashes. After lots of testing, got it working fine in 7 by changing disk from no cache to write back and and take ballooning off. See https://forum.proxmox.com/threads/kernel-panic-whole-server-crashes-about-every-day.91803/page-2. I am hoping that this virtio issue can be looked at but someone more knowledgeable than I.

 

[alexskysilk]

My affected nodes were AMD based, true- but I have no issue with crashing. I believe this is microcode related.

 

[Bottsie]

I know this is not the correct place however, I'm hoping to add to this.

I'm coming from EXSI for many stable years.

I jumped into PVE 7.0.8 (after reading good things about PVE) on new 'much more power full' hardware, think 10+ yo HP Microserver to full-on rack stuff (gluten) and so far I'm not impressed. Full-on PVE crashes no syslog entries that I can see, container environments stressing the hardware to crash. Random stuff like Radarr crash, but Sonarr is OK. Missing basic GUI functions etc.

Why should I stick with PVE. I was doubting the hardware but now PVE is in the spotlight.

How do I easily downgrade, if it's not easier I might as well go back to ESXi, I don't need the premium features, let alone ZFS.

 

[hvisage]

ZFS is one of the great features/etc. to use with PVE I.M.O.
Containers isn't great 'cause off Linux's NIHS - yes I'm using them, gone a while, but you want to use KVM/qemu/VMs to compare to ESXi

Sounds like you need no reason so just go back to ESXi ;(

 

[t.lamprecht]

Now downgrade is not really possible, as it was already mentioned in this thread...

I jumped into PVE 7.0.8 (after reading good things about PVE) on new 'much more power full' hardware, think 10+ yo HP Microserver to full-on rack stuff (gluten) and so far I'm not impressed. Full-on PVE crashes no syslog entries that I can see, container environments stressing the hardware to crash. Random stuff like Radarr crash, but Sonarr is OK. Missing basic GUI functions etc.

I mean you posted about zero details regarding VM/CT configs, CPU in use, are you already on the latest versions, how does the crash look like (reset out of the blue or hangs with some OOPS on the IPMI/TTY consoles, ...), when (roughly) does it happen, did you checked the journal too, what services run in the containers and what processes cause load, ...

We can give it a try to find a potential underlying issue (as good as it's possible in a community forum) but for that we'd need at least above details in a new thread (you can mention my username with @ to "summon" me there)..

 

[alexskysilk]

How do I easily downgrade, if it's not easier I might as well go back to ESXi, I don't need the premium features, let alone ZFS.

If you dont understand why you would need zfs, you're probably right. More to the point, if you're not interested or able to assist in your own troubleshooting, you're not likely to be able to obtain the utility you expect.

What do I mean by this? you posted NO details. no hardware specs. no logs (the fact you're not seeing it doesnt mean there's nothing to be seen) no journal entries from affected services. no "missing" gui functions. What, exactly, are you expecting in reply?