nscd inside CTs with 2.6.32-14-pve

[udo]

Hi,
like antubis wrote here http://forum.proxmox.com/threads/10629-New-Kernel-and-bug-fixes?p=59929#post59929 "eat" the nscd-process inside a CT more than 100% with the new kernel.

This happens also with an debian 6.0.5 (debian-6.0-x86_64.tar.gz).

top - 18:49:05 up 10 min,  2 users,  load average: 2.40, 2.25, 1.24
Tasks:  29 total,   1 running,  28 sleeping,   0 stopped,   0 zombie
Cpu(s): 16.5%us, 18.9%sy,  0.0%ni, 49.6%id,  0.0%wa,  0.0%hi,  0.0%si, 15.0%st
Mem:   3145728k total,    73508k used,  3072220k free,        0k buffers
Swap:  1048576k total,        0k used,  1048576k free,    27348k cached

  PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND                                                                                                        
  261 root      20   0  192m 1036  756 S  136  0.0  13:59.16 nscd                                                                                                           
    1 root      20   0  8360  784  652 S    0  0.0   0:00.00 init                                                                                                           
    2 root      20   0     0    0    0 S    0  0.0   0:00.00 kthreadd/300                                                                                                   
    3 root      20   0     0    0    0 S    0  0.0   0:00.00 khelper/300                                                                                                    
   27 root      20   0   104   12    4 S    0  0.0   0:00.00 init-logger                                                                                                    
  172 daemon    20   0  8100  516  396 S    0  0.0   0:00.00 portmap                                                                                                        
  245 root      20   0 54168 1456 1052 S    0  0.0   0:00.00 rsyslogd

In this case the nscd is not important and can switched off, but on other CTs it's different.

Udo

 

[dietmar]

I filed a bug here:

http://bugzilla.openvz.org/show_bug.cgi?id=2345

 

[udo]

I filed a bug here:

http://bugzilla.openvz.org/show_bug.cgi?id=2345

Hi Dietmar,
thanks a lot!

Udo

 

[dietmar]

Kir just posted a workaround: http://bugzilla.openvz.org/show_bug.cgi?id=2345

 

[udo]

Kir just posted a workaround: http://bugzilla.openvz.org/show_bug.cgi?id=2345

Fine,
just do a "sysctl kernel.vsyscall64_gettime_monotonic=1" on the proxmox-server and the cpu is calm.

Thanks again

Udo