[PVE-User] (no subject)

[Shain Miley]

Every once in a while we have several Oracle instances that consume 100% of the cpu.  So if we have 4 then each use 25%.  If I kill 1..the other 3 consume 33% cpu each.  Once this happens...we are unable to access Oracle at all (via sqlplus,etc) we can access the VM though (ssh, ftp, etc).

Anyway...the only way I have been able to get oracle working again is to reboot the host node.  I can also migrate the VM to another node and reboot that node and it works again as well (it does not work after a migrate and stop/start,etc without the reboot).  

What I am wondering is what is going on during a reboot (even on a different node) that is not going on during a vzctl start/stop of the VM.

I would like to discover a solution that does not require a host node reboot.

Shain




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.proxmox.com/pipermail/pve-user/attachments/20090603/f690237e/attachment.htm>