climateprediction.net home page
What happened to my WU?

What happened to my WU?

Questions and Answers : Windows : What happened to my WU?
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile old_user17289

Send message
Joined: 13 Sep 04
Posts: 228
Credit: 354,979
RAC: 0
Message 37919 - Posted: 31 Aug 2009, 8:03:40 UTC

http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=8367882

It seems that it just finished the year 2020, and uploaded a ZIP file, then just stopped.

I made a backup shortly before that, and now restored it. Same thing happened.

Any suggestion what happened to the WU...?
ID: 37919 · Report as offensive     Reply Quote
Profile astroWX
Volunteer moderator

Send message
Joined: 5 Aug 04
Posts: 1496
Credit: 95,522,203
RAC: 0
Message 37920 - Posted: 31 Aug 2009, 17:56:38 UTC

I lost two with similar diagnostic (except that they lost only the ocean file). Don\'t know the underlying reason but this is the diagnostic:
No Process Handle
CPDN process is not running, exiting, bRetVal = 1, checkPID=2588, selfPID=2588, iMonCtr=1
cpdnmonitor: cannot open input file C:\\Documents and Settings\\pwillener.CINCOM\\Application Data\\BOINC/projects/climateprediction.net/hadcm3istd_cp7t_1920_160_06016813/dataout/atmos_restart.day
cpdnmonitor: cannot open input file C:\\Documents and Settings\\pwillener.CINCOM\\Application Data\\BOINC/projects/climateprediction.net/hadcm3istd_cp7t_1920_160_06016813/dataout/ocean_restart.day

Exit code 22 is a catch-all.

"We have met the enemy and he is us." -- Pogo
Greetings from coastal Washington state, the scenic US Pacific Northwest.
ID: 37920 · Report as offensive     Reply Quote
Profile old_user17289

Send message
Joined: 13 Sep 04
Posts: 228
Credit: 354,979
RAC: 0
Message 37921 - Posted: 1 Sep 2009, 0:36:51 UTC - in response to Message 37920.  

Thanks; I will try to restore the previous backup from 2010.
ID: 37921 · Report as offensive     Reply Quote
Profile old_user17289

Send message
Joined: 13 Sep 04
Posts: 228
Credit: 354,979
RAC: 0
Message 37928 - Posted: 3 Sep 2009, 7:57:24 UTC

The WU has been restarted at 2010, and has been running for a while now. Last night it sent the first trickle since the restart. At the end of the trickle-up it issued this message
2009-09-03 02:47:40 climateprediction.net Generated new computer cross-project ID: cf4cce97478fa3629aca6b45503349fc

Any idea what this means...?
ID: 37928 · Report as offensive     Reply Quote
Profile Iain Inglis

Send message
Joined: 9 Jan 07
Posts: 467
Credit: 14,549,176
RAC: 317
Message 37930 - Posted: 3 Sep 2009, 13:23:22 UTC - in response to Message 37928.  

Any idea what this means...?

BOINC doesn\'t really understand the concept of a backup and can do several things when it notices that the model is repeating something that has already been done.

One thing BOINC can do is to generate a new computer record. If you can see multiple identical computer records in your computer list (as appears to be the case here) then these can be merged back into a single record. In any event, any projects sharing the computer for which the new ID has been generated will synchronise on the new ID after a time.

Another thing that BOINC will do is to mark the model as \'client detached\'. This is irritating but harmless.
ID: 37930 · Report as offensive     Reply Quote
Profile old_user17289

Send message
Joined: 13 Sep 04
Posts: 228
Credit: 354,979
RAC: 0
Message 37945 - Posted: 4 Sep 2009, 7:14:30 UTC

Thank you for the explanation!
ID: 37945 · Report as offensive     Reply Quote
Profile old_user17289

Send message
Joined: 13 Sep 04
Posts: 228
Credit: 354,979
RAC: 0
Message 38026 - Posted: 24 Sep 2009, 7:07:55 UTC

Just an update: when I came back from my vacation after 2½ weeks, the WU has now well passed the original point of failure, and is now processing the year 2025.

This is thanks to making regular backups, and keeping them all for a while!
ID: 38026 · Report as offensive     Reply Quote
Profile mo.v
Volunteer moderator
Avatar

Send message
Joined: 29 Sep 04
Posts: 2363
Credit: 14,611,758
RAC: 0
Message 38027 - Posted: 24 Sep 2009, 10:23:31 UTC

You\'re right, backups have saved a lot of models. I hope it will help you complete one model from that workunit. The other person in that workunit who was doing well appears to have abandoned the model in July. Really, CPDN needs people who don\'t abandon models except in extreme circumstances.
Cpdn news
ID: 38027 · Report as offensive     Reply Quote

Questions and Answers : Windows : What happened to my WU?

©2024 climateprediction.net