Search results

Ceph writes are slowed down by network latency that does not improve with higher speeds cards. Best options are: krbd on, write back cache and iothread=1 but I see others have already suggested them to you.

Upgraded to latest version. It seems that if I have many vm to migrate ( I set policy type migration for the cluster) it probably use too much time to reboot and so ha-manager gets crazy.

It is doing it again. Fenced node and other nodes gave the same result: quorum OK master pvedell (active, Fri Jan 8 18:19:10 2021) lrm pvedell (active, Fri Jan 8 18:19:13 2021) lrm pvedell1 (active, Fri Jan 8 18:19:06 2021) lrm pvehpbig (active, Fri Jan 8 18:19:15 2021) service vm:100...

Hello, I have log full of messages like: Dec 29 11:07:26 pvehpbig ceph 2020-12-29T11:07:26.978+0100 7fbfa75b1700 10 mgr tick tick Dec 29 11:07:26 pvehpbig ceph 2020-12-29T11:07:26.978+0100 7fbfa75b1700 20 mgr send_beacon standby Dec 29 11:07:26 pvehpbig ceph 2020-12-29T11:07:26.978+0100...

I would like to explain better: here we are not choosing among several power profiles. Here we are choosing between "let linux kernel manage powersave" or "let hp ilo manage powersave". The second one gives problems!

Infact I got crazy for several months too. It is incredible that "dynamic power savings" is default in hp ilo. It should not have this behaviour.

I have several proxmox clusters on HP G8 and G9 servers, with Windows Server VMs. Terminal servers users complains about "disk slowness" (I have ssd and nvme...), "micro interruptions" and so on. It seemed an unsolvable problem until I discovered that changing in ilo4/power settings from...

Hello, I would like to monitor all my proxmox/ceph installations: - check that backups start; - check errors after backup run; - check if ceph osd is down; - check if some ha vm is in error state I plan to use influxdb with telegraf or opendistro for elasticsearch or others. I started with...

I have the same error on a zfs proxmox backup server. And NO I do not have an hardware raid controller. After reboot upgrading from 0.8.x to 0.9-1 it does not boot anymore. In addition to this I can see that with an ubuntu live cd I can mount and see my zfs pool so it is not corrupted. it is the...

Great trick! Very easy! No need to recover from backup! Thanks!

Hello, I have removed a proxmox node. By mistake I forgot to migrate some VMs. The disks for these machines are on ceph, so I have only lost their .conf. I have backups so I try to write a xxx.conf (recovered with vma config) in /etc/pve/qemu-server of another node. Unfortunately even if...

I have chosen to use batman protocol and now I have a full mesh with all links in the server working (full bandwidth, no waste) and at level 2. If you want to enhance wiki I can post details.

I explain better. In proxmox documentation I read that stp must be off. On proxmox created bridges I see they put stp off. In this case stp must be on to avoid loops: the standard way to connect several servers without using switches is to put nics in bridges with stp on. Now in the link above...

Hi, I have just read this: mesh And I do not understand why I cannot put simply a linux bridge with stp on. I have done it and it works. Mario

Anyway I have tried enabling multiqueue=8 in some test windows VM and freenas VM and I obtained total damage: network up and down, packet loss and so on

I would like to underline that proxmox official documentation says that you must put mulitqueue=number of threads BUT then proxmox gui behaviour is different from official documentation

There is another thread about this

Sorry to be rude but reading again you have 20 osd but only 13 up so 7 are down. Then 27 pgs inactive means that ceph has no more copies and so it has lost your data. So it is obvious backup will hang: ceph has stopped working.

Some questions: 1) are your server new? On used server when you overload scsi controller it may happen that it loses drives showing them not connected. 2) ceph is not magical: if you loses too many hdds and it discovers that it has not enough copies of data it will stop immediately accepting...

Thanks for reply, I have checked in nic options and there is nothing specific to multiqueue.