Search results

Note that qemu share a single thread by default, for disk,nic,.... maybe can you try to enable iothread on disk ? (virtio or scsi + virtio-scsi-single controller). (note that it's not yet compatible with proxmox backup)

>>While normal usage I have ping response times between 20-35 mSec on my local bridge Is it really from host bridge to to vm ? I'm around 0.1ms from host bridge to guest vm. more than 1ms locally is really anormal.

Maybe related: https://lists.gnu.org/archive/html/qemu-devel/2016-01/msg03973.html

does the back migration (newnode->oldnode) works if you only update pve-qemu-kvm package on oldnode ? (apt-get install pve-qemu-kvm should be enough to upgrade only qemu) Note : And just to be sure, the vm has not been stop/start on new node, before trying migrate back ? Because in this...

apt-get upgrade only upgrade installed packages apt-get dist-upgrade also install new packages from dependencies. So you should always use dist-upgrade, or it could give you breakage like this.

watchdog is now disable by default, but in proxmox 4.1 only. Do you have done last updates ?

you should enable igmp snooping on your network switch to avoid mutlicast packets from other cluster coming to theses nodes. (also be carefull to not use same clustername, because the multicast address is generated from clustername)

you can try to do a "ps -aux" on host, and compare qemu command of proxmox vs ubuntu.

Is the server bios and cpu microcode update on last version ?

Hi, Can you post result of "ipmitool mc watchdog get" ? (ipmitool package need to be installed) what is your server model ? which watchdog do you use ?

Hi, this is not a real kernel panic, this is your vms kernel which send error message, because the storage is not responding or too slow. Maybe your storage is overloaded when this happen ?

Hi,another way could be to disable motherboard watchdog,to use the hp ilo watchdog by default. edit: /etc/default/grub GRUB_CMDLINE_LINUX_DEFAULT="nmi_watchdog=0" #update-grub #reboot

I also found a note here: https://lkml.org/lkml/2014/4/25/184 "hpwdt can not work as expected if hp-asrd is running simultaneously.+Because both hpwdt and hp-asrd update same iLO watchdog timer." Do you have an hp-asrd daemon running ? (maybe from some hp management packages ?)

ubuntu has also disable it by default. https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1432837 But, It could be a problem of ilo configuration when watchdog is enable by hpwdt. Maybe they are a ilo timeout configuration somewhere in ilo ?

I think we could try qemu 2.3 with kernel 4.2. It could help to see if the problem come from qemu 2.4 or kernel. If it's don't work, try again qemu 2.3 but with kernel 4.1 http://download.proxmox.com/debian/dists/jessie/pvetest/binary-amd64.beta1/pve-kernel-4.1.3-1-pve_4.1.3-7_amd64.deb or...

Note that you don't need to mount your fs with discard=on, to use fstrim. if you use discard=on, trim will be done at each delete in your filesystem, and this can be quite slow. It's better to simply add a cron, daily for example, which launch fstrim. (without any discard=on)

It'll not be stable until 1.X . Currently they are still cluster format changes between release 0.8 -> 0.9, need to backup/restore for example.

Proxmox don't add nothing special to the distro, only qemu 2.4 is used instead default qemu 2.1 in debian jessie. you can install proxmox 3.4, qemu 2.2 , on proxmox 4.0 if you want. wget...

I could be great to see if it's working with proxmox qemu 2.4 + jessie 3.16 kernel for example. I think it's a kvm bug in last kernels, but I would like to be sure that's it's not a regression in qemu.

you can change the repository in /etc/apt/sources.list.d/ceph.list (https://download.ceph.com/debian-infernalis/) Please read the upgrade procedure: http://ceph.com/releases/v9-2-0-infernalis-released/ because they are some file permissions change with infernalis. (was root:root previously...