[RC5] Partial work saved during time limit shutdown?

O.Mico odjmico2 at wanadoo.fr
Sat Jan 25 22:59:29 EST 2003


It is not a bug ! Packets remain in memory so when the client close
itself blocks being proecessed are not saved it's clear and logical...

> -----Original Message-----
> From: owner-rc5 at lists.distributed.net 
> [mailto:owner-rc5 at lists.distributed.net] On Behalf Of SvR Marty
> Sent: Saturday, January 25, 2003 7:16 PM
> To: rc5 at lists.distributed.net
> Subject: Re: [RC5] Partial work saved during time limit shutdown?
> 
> 
> search or submit a bug for this
> 
> bugs.distributed.net
> 
> 
> ----- Original Message ----- 
> From: "Michael Pelletier" <mvpel at nortelnetworks.com>
> To: <rc5 at lists.distributed.net>
> Sent: Thursday, January 23, 2003 7:03 PM
> Subject: [RC5] Partial work saved during time limit shutdown?
> 
> 
> > I'm running v2.9001-477-CTR-02111721 for Solaris, and will 
> > ocassionally set up a few machines for an overnight run 
> with an 8 hour 
> > time limit.
> > 
> > I set the remote-buffer configuration as follows:
> > 
> > [buffers]
> > buffer-only-in-memory=yes 
> alternate-buffer-directory=/my/homedir/dnetc
> > frequent-threshold-checks=4
> > 
> > [misc]
> > project-priority=OGR,RC5-72
> > run-time-limit=8:00
> > 
> > [rc5-72]
> > fetch-workunit-threshold=1
> > 
> > [ogr]
> > fetch-workunit-threshold=1
> > 
> > ========= Now, what seemed to happen is that the last two 
> packets from 
> > RC5-72 were loaded (I don't fetch those buffers) in the 6 
> CPU system, 
> > then 6 OGR packets were loaded, and crunching commenced - 
> but when the 
> > time limit was up, the RC5 packets were not returned to 
> buff-in.rc5, 
> > and as far as I can tell, the last 6 were not either:
> > 
> > [Jan 23 07:50:11 UTC] OGR: Transferred 1 packet (71.17 
> stats units) to 
> > file. [Jan 23 09:08:06 UTC] Shutdown - reached time limit.
> > [Jan 23 09:08:06 UTC] OGR: Saved 25/4-6-8-17-2-3 (33.42 Gnodes done)
> >                       0.03:08:00.18 - [2,962,841 nodes/s]
> > [Jan 23 09:08:06 UTC] OGR: Saved 25/7-2-3-8-32-15 (50.95 
> Gnodes done)
> >                       0.04:44:24.08 - [2,986,229 nodes/s]
> > [Jan 23 09:08:06 UTC] OGR: Saved 25/4-6-8-26-19-3 (35.70 
> Gnodes done)
> >                       0.03:17:51.98 - [3,007,334 nodes/s]
> > [Jan 23 09:08:06 UTC] OGR: Saved 25/4-6-8-17-2-26 (34.28 
> Gnodes done)
> >                       0.03:12:25.53 - [2,969,979 nodes/s]
> > [Jan 23 09:08:06 UTC] OGR: Saved 25/4-6-8-16-31-2 (25.15 
> Gnodes done)
> >                       0.02:20:34.65 - [2,982,283 nodes/s]
> > [Jan 23 09:08:06 UTC] OGR: Saved 25/4-6-8-16-35-1 (26.35 
> Gnodes done)
> >                       0.02:27:44.26 - [2,973,160 nodes/s]
> > [Jan 23 09:08:06 UTC] OGR: Summary: 10 packets (306.26 stats units)
> >                       0.08:00:03.89 - [10.63 Mnodes/s]
> > [Jan 23 09:08:06 UTC] OGR: 0 packets are in (memory-in)
> > [Jan 23 09:08:06 UTC] OGR: 0 packets are in (memory-out)
> > [Jan 23 09:08:06 UTC] *Break* Shutting down...
> > 
> > It says it "saved" the packets, but where?  Is there some different 
> > configuration I need to use for this approach?
> > 
> > -Michael Pelletier.
> > 
> > --
> > To unsubscribe, send 'unsubscribe rc5' to 
> > majordomo at lists.distributed.net rc5-digest subscribers replace rc5 
> > with rc5-digest
> > 
> --
> To unsubscribe, send 'unsubscribe rc5' to 
> majordomo at lists.distributed.net rc5-digest subscribers 
> replace rc5 with rc5-digest
> 

--
To unsubscribe, send 'unsubscribe rc5' to majordomo at lists.distributed.net
rc5-digest subscribers replace rc5 with rc5-digest



More information about the rc5 mailing list