[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [condor-users] Condor sleeping
- Date: Thu, 17 Jun 2004 03:26:29 -1000
- From: "Ron Viloria" <Ron.Viloria@xxxxxxxxxxxxx>
- Subject: RE: [condor-users] Condor sleeping
Just wanted to let folks know that I did respond and emailed Colin my
config files and it so happened to be too big to post to the users list.
I believe he is looking at the issue right now.
I do not believe I explicitly used TCP Updates and I have seen the same
issue on clusters smaller than 25, and on old versions. Running W2KP,
XPP, W2K3S, on dual proc servers, single proc desktops, and laptops in a
dedicated configuration. I also explained that the nodes eventually do
come back online but at times after a long while.
> -----Original Message-----
> From: condor-users-bounces@xxxxxxxxxxx
> [mailto:condor-users-bounces@xxxxxxxxxxx] On Behalf Of Colin Stolley
> Sent: Wednesday, June 16, 2004 10:02 AM
> To: Condor-Users Mail List
> Subject: Re: [condor-users] Condor sleeping
> >I easily run a script across the pool from time to time that
> >the condor service. I have lived with that mechanism.
> I'm not sure what your configuration is, but it might
> interest everyone on this list that a bug was discovered this
> week that causes updates to the collector to be dropped if
> your pool is configured to use TCP updates, has more than
> about 25 machines in it, and the collector is running on
> Windows. This issue will be fixed in 6.6.6.
> Condor-users mailing list