Search results

Hello again everyone, been too long since my last post here. I have one server randomly locking up for over a month now, now a 2nd server is also having this problem. Unfortunately I've not captured all of the kernel messages that would help diagnose this but I have a couple screenshots from...

@Alwin I've tried setting the values back to default so I can test before and after but performance stays the same. Its possible that there is some other explanation to my initial results.

@Alwin I used dd and 'time cp' on a VM where I could only copy anything at about 40MB/sec and now I can get 140MB/sec.

Yup, that thread is what gave me the idea to try this. Found it this morning. I suspect these settings will help anyone facing IO related performance issues.

@nicko I'm curious to know if this helps you or not: https://forum.proxmox.com/threads/increase-performance-with-sched_autogroup_enabled-0.41729/ For me it made a huge difference in IO performance on numerous servers.

Changing sched_autogroup_enabled from 1 to 0 makes a HUGE difference in performance on busy Proxmox hosts Also helps to modify sched_migration_cost_ns I've tested this on Proxmox 4.x and 5.x: echo 5000000 > /proc/sys/kernel/sched_migration_cost_ns echo 0 >...

Have you rebooted into memtest and checked your RAM? That would be my first suggestion. If you cannot afford the downtime you might try installing and using memtester: apt-get install memtester

What is the IO Wait on the Proxmox host during the backup? If its high then you are starving the VM of disk IO causing the VM to think its disks are bad because they are not responding. I've only started using ZFS a few months ago so I am far from an expert. It seems that ZFS has its own...

Not enough contiguous free RAM to allocate the RAM requested. This will display how many contiguous allocations of each 'order' are available: cat /proc/buddyinfo From left to right each column represents the count of allocations available for each order starting with order 0. The size of each...

You can offline migrate when using storage replication, wish it worked live.

I've got DRBD setup on some 5.x servers using a setup similar to the old wiki article. @fwf DRBD will end up diskless on reboot when it cannot find the disk you specified in the configuration. How did you reference the disks in drbd config? I've found that using /dev/sdX is a bad idea because...

This is reported upstream already by someone else, I added my info there too. https://github.com/zfsonlinux/zfs/issues/6781 I setup DRBD on top of a ZVOL. When making heavy sequential writes on the primary, the secondary node throws a General Protection Fault error from zfs. The IO was from a...

@fabian I just ran into this problem myself Installing pve-kernel-4.13.8-3-pve_4.13.8-30_amd64.deb from pve-test seems to have resolved the issue. I would be happy to give Proxmox one of these cards to put into the test servers. Would you like me to ship it to you?

I have seen one report that with Luminous and Proxmox 5.0 the situation is much better: https://forum.proxmox.com/threads/ceph-luminous-backup-improvement.34678/ Apparently some fixes in librdb to reduce memory copies is the source of the improvement. If the 64k reads in QEMU backup were...

Looks like you are using zvols so all you need to do is copy the volume. Something like this: dd if=/dev/zvol/rpool/data/vm-100-disk-1 of=/dev/sdX bs=1M

Yes, disable disk swap and use zram. Every Monday testing VMs would be swapped to disk after being idle all weekend and testers complain their VMs are slow. Even setting swappiness=1 would not prevent the problem. Not had a single complaint or OOM event since removing disk swap and adding...

If your using raw the vm disk data can be copied directly to a drive. You could boot a linux live cd on the bare metal then use dd and netcat to clone the vm disk to the bare metal. Be sure to stop the vm first. The only tricky part, because of Windows, is making sure the driver for your...

I don't create new VMs frequently enough to invest time into automating the process. I just use the Windows and Ubuntu templates I created and then bootstrap Chef via SSH or winrm. In Proxmox 5 we will hopefully have cloudinit, that will make using templates even easier. Would be nice to see...

Is this with or without KRBD?

Each node has a dedicated group that looks like this: group: NodeName nodes Node_Name nofailback 0 restricted 0 Each node has a diskless VM like this: bootdisk: scsi0 cores: 1 freeze: 1...