We are in the process of building a new cluster that's not yet in production. We're trying to hold off on going live with it until v4 is released since it will contain several features we will want to use. Any idea when that may be? Ballpark estimates would be great.
v4 ETA?
- Thread starter churnd
- Start date
Well, I guess it's the debian way :" it's done when it's done"
Ha, yeah, would be good to know how much is done then.
IMHO the base system is stable, and qemu/kvm is basically finished. We are now try to implement missing LXC features...
After few days of using 4b1: ZFS became usable only after upgrading to 0.6.4.2 (0.6.4.1 is shipped with beta1), qemu q35 machines don't work correctly (failing to reboot), and few minutes ago node rebooted for no reason at all. PCIe cards behave erratically after some uptime. (virtual machines refuse to boot, E1 card is stuttering)
We fixed some bugs since the beta iso was released, so if your installing it from the iso, the first thing after that should be:
The node that rebooted may have lost quorum, and so the watchdog reset the node, look in the logs. Never saw a reboot without provoking it and think I can say I have spent some time with pve4
The problem with q35 was mentioned (and I think also related to the PCI problems) and someone solved it (or had a workaround) also, if I recall correctly, I have to search the answer later though.
We would be really happy if you open bugreports on https://bugzilla.proxmox.com/ with detailed description of the problem, your system specs and, if possible, a way to reproduce the issue.
Code:
apt-get update
apt-get dist-upgradeThe node that rebooted may have lost quorum, and so the watchdog reset the node, look in the logs. Never saw a reboot without provoking it and think I can say I have spent some time with pve4
The problem with q35 was mentioned (and I think also related to the PCI problems) and someone solved it (or had a workaround) also, if I recall correctly, I have to search the answer later though.
We would be really happy if you open bugreports on https://bugzilla.proxmox.com/ with detailed description of the problem, your system specs and, if possible, a way to reproduce the issue.
There's always a quorum, since it's a single node - I'm observing and reporting what I can. Right now I can only report that 0.6.4.2 has fixed major problems with ZFS compared to 0.6.4.1 -- hot-pulling a drive caused a kernel panic on 0.6.4.1 (shipped with Proxmox), and is working as intended on 0.6.4.2. I'd love to hear about this q35 fix.
- I'm observing and reporting what I can. Right now I can only report that 0.6.4.2 has fixed major problems with ZFS compared to 0.6.4.1 -- hot-pulling a drive caused a kernel panic on 0.6.4.1 (shipped with Proxmox), and is working as intended on 0.6.4.2. I'd love to hear about this q35 fix.