Search results

Well, our nodes are very different hardware wise, yet the problem surfaces on all of them (but not on all guests). We have experienced the issue on single and dual socket servers, sporting 32GB to 96GB RAM, but there is one thing in common: all of them use ZFS (KVM guests are running on...

If you read carefully, the 4GB of RAM belongs to the the KVM guest that produced the errors. These CPU stalls are happening on all of our nodes, regardless of hardware configuration. Yes, we tried monitoring the hosts and the guest nodes as well, no particular reason has shown up.

Right, so I disabled the iothread option on the individual disks, yet the controller type is still set to virtio single, and the guest Windows still shows 3 controllers connected to the 3 disks. Backup completed successfully. Does this config mean I'm still getting one IO thread for each disk...

@dietmar @fabian is this true? what is supposed to happen, does the backup stop with an error?

This is the essential CPU stall console message, but as it passed relatively fast it didn't freeze up the kernel: [301703.260023] INFO: rcu_sched detected stalls on CPUs/tasks: { 1} (detected by 2, t=8978 jiffies, g=2886633, c=2886632, q=265) [301711.937277] [<ffffffff810d259a>] ...

Really? No one else has seen CPU stalls / freezes in the entire Proxmox community?

Does that mean that vzdump won't be able to take live backups of my VM that's stored on ZFS?

So I need to select VirtIO SCSI single in Options, and also enable IO Thread for every disk in Hardware to ensure iothread is working? Is there any other limitation (cache type, etc.) for iothread? Right now my VM is on zfs-local, with cache=writeback. Is there any reliable way to verify that...

Since we upgraded our cluster to PVE 4.3 from 3.4, all our OpenVZ containers have been converted to KVM virtual machines. In many of these guests we get frequent console alerts about CPU stalls, usually when the cluster node is under high IO load (for example when backing up or restoring VMs to...

Any news on Ceph Jewel landing in Proxmox? We are really looking forward to using CephFS.

So how is this coming along? Has anyone tested it? I would gladly test...

As both issues are resolved now, when can we expect a new Ceph version in PVE? The pve-no-subscription repo for PVE 4.3 still contains Ceph 0.80.7, which is exactly 2 years old... And what about GlusterFS, where 3.5.2 is included, which is EOL for 4 months now, as 3.8 LTM is out since June...

Sounds safe enough to me. So are you going to provide a patch to Proxmox that gets into the test repository, or how is this going to work? I will be more than happy to test.

Yes, PVE 3.x shows 12 rows where 4.x shows only 10. Doesn't sound like a lot, but when you have to manage a lot of guests, it quickly adds up and you have to scroll a lot more. So I would very much like an option to set font (and list item) size in the new GUI.

I think there is a more important issue here than the new navigation, which looks a bit unconventional and takes some getting used to, but I personally think is actually better from a usability standpoint (your eyes need to scan a much smaller area to find the next click). What I personally...

This has been talked to death before in this thread, you might want to read first and comment later. Ceph and Gluster are very useful, but only for backups or very low IO guests (at least on our bonded gigabit infrastructure). Also, this has nothing to to with the fact that block migration code...

Why would it start if a node is down? No one said it would be automatic, or in any way connected to HA. It would only be faster than it currently is, with much less downtime. In Proxmox 4.x a KVM guest can only be migrated fully offline from local storage, which is unacceptable and lags behind...

Actually, LXD + LXC 2.0 has some solutions for this problem, now the question is can they be implemented in Proxmox? More in this thread: https://forum.proxmox.com/threads/revisiting-cpu-limit-for-lxc-2-0-lxd.28736/ I have also commented the relevant link in the bugzilla.

Hyper-V also supports it: "You can also perform a live migration of a virtual machine between two non-clustered servers running Hyper-V when you are only using local storage for the virtual machine. (This is sometimes referred to as a “shared nothing” live migration. In this case, the virtual...

I have created a feature request in the Proxmox bugzilla, I hope we get support for this feature: https://bugzilla.proxmox.com/show_bug.cgi?id=1118