On Fri, Jul 21, 2006 at 10:54:04AM +0100 or thereabouts, Alessandra Forti wrote:
> Hi,
>
> I was checking the network connections on manchester CE with netstat and
> I found
>
> >3000 TIME_WAIT connections
> pbs_server - pbs_server and
> pbs_mom to pbs
>
> any idea what be causing the problem?
> I couldn't find anything related to torque itself.
The MOMs check into the server regularly with their
status but I notice we only have 267 TIME_WAITs currently.
With server settings of
set server node_ping_rate = 300
set server node_check_rate = 600
set server tcp_timeout = 20
set server job_stat_rate = 120
set server poll_jobs = True
so I don't know why yours is so hight.
One thing I meant to ask you, you upgraded
torque I know but did you also upgrade maui. Maui
has staic libpbs in it so it is important to use matching
versions which I know I should hard code into the dependencies
and will do so next time.
# rpm -qa | egrep 'maui|torque'
torque-clients-2.0.0p7-1.sl3.st
torque-config-servername-ral-1.0-1
torque-2.0.0p7-1.sl3.st
torque-devel-2.0.0p7-1.sl3.st
maui-3.2.6p14-3_SL30X_ratio01
maui-server-3.2.6p14-3_SL30X_ratio01
torque-server-2.0.0p7-1.sl3.st
maui-client-3.2.6p14-3_SL30X_ratio01
>
> cheers
> alessandra
>
> --
> *******************************************
> * Dr Alessandra Forti *
> * Technical Coordinator - NorthGrid Tier2 *
> * http://www.hep.man.ac.uk/u/aforti *
> *******************************************
--
Steve Traylen
work email: [log in to unmask]
personal email: [log in to unmask]
jabber: xmpp:[log in to unmask]
|