My "to completion" time seems to be going up.
Running a P-4 XPsp2 Otherwise everything's running smooth.

Edit: I'm not sure I understand what happened but I think this wu started over again wu_1204099514_212_0 using lalanne version 506
____________
A clear conscience is usually the sign of a bad memory!

My "to completion" time seems to be going up.
Running a P-4 XPsp2 Otherwise everything's running smooth.

Edit: I'm not sure I understand what happened but I think this wu started over again wu_1204099514_212_0 using lalanne version 506

I just watched one under XP. I noticed that the CPU time stopped accumulating for about 5 or 6 seconds (as timed by the ever so scientific and precise one-mississippi two-mississippi method). When it started to accumulate time again, I switched to the message tab. It had sent a trickle message during those few seconds. This may be why I'm seeing CPU times of around 50 minutes under XP (it's a dedicated cruncher right now).

Of course I'm out of work on the Linux box right now so I can't watch and see if it behaves the same way, but I'm guessing they do because CPU times were about 55 minutes. I can grep the log to make sure the wall time was an hour if you'd like.


[edit]Holy cow... did I just use grep as a verb :-)[/edit]
____________
Kathryn :o)
The BOINC FAQ Service
The Unofficial BOINC Wiki
The Trac System

This result ran 1.25 hours CPU time. It's also one that was running when I stopped the service to upgrade the Einstein power users app and the core client.

worker: CPU time: 4506.453125 seconds, wall clock time: 3634.750000 seconds for 100000000 iterations.

That WU at 102.286% to completion at the moment with 20 minutes or so crunch time. Sorry if this was mentioned elsewhere... does it run for a certain time or does it vary from platform to platform?

[KSMarksPsych]This result ran 1.25 hours CPU time. It's also one that was running when I stopped the service to upgrade the Einstein power users app and the core client.

worker: CPU time: 4506.453125 seconds, wall clock time: 3634.750000 seconds for 100000000 iterations.

I think (I'm 95% positive) it was roughly between 10 and 15 minutes in when I stopped the service.

Strange...

This one didn't trickle at all.

2008-02-29 23:36:36 [Pirates@Home] Starting wu_1204273265_314_0
2008-02-29 23:36:36 [Pirates@Home] Starting task wu_1204273265_314_0 using lalanne version 506
2008-03-01 00:39:43 [Pirates@Home] Computation for task wu_1204273265_314_0 finished
2008-03-01 00:39:45 [Pirates@Home] [file_xfer] Started upload of file wu_1204273265_314_0_0
2008-03-01 00:39:52 [Pirates@Home] [file_xfer] Finished upload of file wu_1204273265_314_0_0
2008-03-01 00:39:52 [Pirates@Home] [file_xfer] Throughput 1452 bytes/sec


These are the lines from Pirates that grep found, so I took a look at boinc.log and there indeed are no trickle messages for the result.
____________
Kathryn :o)
The BOINC FAQ Service
The Unofficial BOINC Wiki
The Trac System

Eww.

Wow, my first success, on the Linux box, but with TWO problems.

First: I had set the application cycle to 20 minutes, and it would
not realise it HAD already run for 20 minutes when doing its 60-minute
run, so it could not finish until I quit BOINC, edited client_state.xml,
added <suspended_via_gui/> to all <result> tags except the Pirate one,
and restarted BOINC.

Secondly, not a problem, but a cosmetic (for me), but maybe useful: