vpsnetclean crashes PVE every 3-4 days

Hello,

I've been having major problems with the vpsnetclean cron script that manifests only after 2-4 days runtime on average. It ends with a complete lockup of the PVE server.
I notice some similar reports, but I do believe I am using the latest kernel with OVZ support? I run one OVZ container and two KVM's at light load.

If there is no way of improving this, would dropping OVZ and move to 2.6.35 kernel improve stability while running only KVM's. A shame because everything works perfectly otherwise. Or can these vps scripts be disabled?

---

Last line in log always:

Mar 3 06:25:01 proxmox /USR/SBIN/CRON[8956]: (root) CMD (/usr/share/vzctl/scripts/vpsnetclean)

---

---
pveversion -v:

pve-manager: 1.7-12 (pve-manager/1.7/5490)
running kernel: 2.6.32-4-pve
proxmox-ve-2.6.32: 1.7-30
pve-kernel-2.6.32-4-pve: 2.6.32-30
qemu-server: 1.1-28
pve-firmware: 1.0-10
libpve-storage-perl: 1.0-16
vncterm: 0.9-2
vzctl: 3.0.24-1pve4
vzdump: 1.2-11
vzprocps: 2.0.11-1dso2
vzquota: 3.0.11-1
pve-qemu-kvm: 0.13.0-3
ksm-control-daemon: 1.0-5

dietmar said:

I suggest to use kernel 2.6.18 if you want to run openvz containers.

Click to expand...

Thanks, confirmed my thoughts... Actually I just went to 2.6.35 and dropped openvz stuff entirely. Everything seems more stable and faster on KVM and now have KSM support.

I went with VZ containers at first due to convenience but ended up being more headache than it was worth.

Cheers,

Kevin