|
Horms,
ip_vs_sync.c:
885: __set_current_state(TASK_UNINTERRUPTIBLE);
886: add_wait_queue(&stop_sync_wait, &wait);
887: set_stop_sync(state, 1);
888: ip_vs_sync_state -= state;
889: wake_up(&sync_wait);
890: schedule();
891: __set_current_state(TASK_RUNNING);
892: remove_wait_queue(&stop_sync_wait, &wait);
Seems to have changed? Theres no "schedule_timeout(HZ);" throughout.
Regards
__________________________________________
Dean Manners
> -----Original Message-----
> From: lvs-users-bounces@xxxxxxxxxxxxxxxxxxxxxx
> [mailto:lvs-users-bounces@xxxxxxxxxxxxxxxxxxxxxx] On Behalf Of Horms
> Sent: Friday, November 10, 2006 12:15 PM
> To: LinuxVirtualServer.org users mailing list.
> Subject: Re: lvs sync, unusual load
>
> On Thu, Nov 09, 2006 at 10:17:36AM +0100, Sebastian Vieira wrote:
> > This happens on my systems as well. Running one sync daemon
> results in
> > an increase of 1.00 load avg. Both daemons: 2.00 load avg.
> Some time ago Horms
> > created a patch for ip_vs_sync.c It basicly involves changing:
> >
> > - __set_current_state(TASK_INTERRUPTIBLE);
> > - schedule_timeout(HZ);
> >
> > for
> >
> > + ssleep(1);
> >
> > Recompile kernel (and ipvsadm?) reboot and the load will be
> normal again.
> > I've tested this on six lvs's and all of them perform
> normally after
> > the patch.
>
> Ahh, that old gem. Dean, can you check to see if this change
> is present in ip_vs_sync.c?
>
> Also, its unrelated to ipvsadm, so you won't need to touch that.
>
> --
> Horms
> H: http://www.vergenet.net/~horms/
> W: http://www.valinux.co.jp/en/
>
> _______________________________________________
> LinuxVirtualServer.org mailing list -
> lvs-users@xxxxxxxxxxxxxxxxxxxxxx Send requests to
> lvs-users-request@xxxxxxxxxxxxxxxxxxxxxx
> or go to http://www.in-addr.de/mailman/listinfo/lvs-users
>
|
