Recent content by XServer

I had same problem and have no idea to solve it correctly, but here is at least some solution: You have to change mount point for VE102 from /var/lib/vz/root/102 to (for example) /var/lib/vz/root/YAHOO/102 mkdir -p /var/lib/vz/root/YAHOO/102 and in the file /etc/pve/openvz/102.conf (or...

There is another problem causing this error - VEs on this node can't stop. OS inside VE is completely stops, but "private" doesn't unmount. Therefore VE is being paralyzed, vzctl stop VEID returns error "operation timed out". And vzctl start VEID returns "container is already running"...

Hello! On my new HP DL320G6 server there are some strange errors occured while finishing migrating VEs: ------------[ cut here ]------------ WARNING: at fs/inode.c:793 unlock_new_inode+0x2c/0x4c() Hardware name: ProLiant DL320 G6 Modules linked in: vzethdev vznetdev simfs vzrst vzcpt...

Hi! I have last PVE with 2.6.32+KSM on server with 8Gb memory. cat /proc/meminfo is only display 7395836kb (in proxmox web-interface it is 7.05Gb) Where is the remained ~1Gb? :o PVE is installed on Debian Lenny (not from Proxmox VE ISO).

Re: Proxmox VE 1.5 released I know :) I mean maybe the disk cache of VMs is not the same, because this VMs is on another disks. I'm just looking for the reason why ksm doesn't work as i would like to... Now after reboot 3 of VMs i have: s1:~# cat /sys/kernel/mm/ksm/pages_shared 26411 s1:~# cat...

Re: Proxmox VE 1.5 released maybe it's because all VMs are on different physical drives? We have server with follow scheme: VMID Physical Drive OS 101 0 SUSE11.2 102 1 Debian 5.0 amd64 103 2 RHEL5.4 104...

anybody has the same troubles?

Re: Proxmox VE 1.5 released No, memory is only user for diskcache. Applications in VM doesn't use so many memory.

Re: Proxmox VE 1.5 released And I have KSM does not work :( First time it works, and then after couple of hours it is consumed all memory.

Yes, it is 3.0.23-1pve6. I try it later update: Problem persists. Here (http://forum.openvz.org/index.php?t=msg&goto=37009&) is recommended to update the kernel.

Where it (cpulimit) can be set in web-interface?

openvz with bridged eth0. for reproduce - start many ve's (i have about 40) and after some time to stop one of them.

No, it's only increases the time of container stopping to about 2 minutes.