Questions and Answers :
Unix/Linux :
New % on tghe result
Message board moderation
Author | Message |
---|---|
Send message Joined: 10 Dec 05 Posts: 5 Credit: 175,291 RAC: 0 |
Hello I have crunch a hadsm3 wu, and after the phase 2, the % is starting new from 0% - is this right or is the wu crashed? http://climateapps2.oucs.ox.ac.uk/cpdnboinc/result.php?resultid=7299083 Edit: Now i have a error message: Fr 14 Mär 2008 23:38:55 CET|climateprediction.net|Task hadsm3fub_025k_005933905_8 exited with zero status but no \'finished\' file Fr 14 Mär 2008 23:38:55 CET|climateprediction.net|If this happens repeatedly you may need to reset the project. Fr 14 Mär 2008 23:38:55 CET|climateprediction.net|Restarting task hadsm3fub_025k_005933905_8 using hadsm3 version 510 |
Send message Joined: 5 Sep 04 Posts: 7629 Credit: 24,240,330 RAC: 0 |
the % is starting new from 0% - is this right or is the wu crashed?It depends on where you see this %. On some types of model, such as the slab model that you\'re running, the % WILL be shown as starting again, and will do so for each phase. However, it is possible for the model to have a problem and start again from scratch if it gets interrupted at the phase change, and before it\'s had a chance to reach the first savepoint in the new phase. After this, it\'s fairly safe. This is currently being investigated. exited with zero status but no \'finished\' fileThis is well known, and is documented in the READMES, which are here in Crashes and other problems, 4th from the top. Backups: Here |
Send message Joined: 6 Aug 04 Posts: 124 Credit: 9,195,838 RAC: 0 |
Hello Look at the timesteps, are they still growing? Linux Users Everywhere @ BOINC |
Send message Joined: 10 Dec 05 Posts: 5 Credit: 175,291 RAC: 0 |
Yes, they are growing - i have start the wu again after the Race on primegrid - we will see The WU is running on OpenSuse 10.3 64bit - Boinc 5.10.28 |
Send message Joined: 13 Jan 06 Posts: 1498 Credit: 15,613,038 RAC: 0 |
It sounds like the model may have been interrupted (gone out of memory) during the post-processing phase. Sometimes this causes it to go all the way back to the start. This restarting-after-phase-change-interruption problem has been analysed in depth by Thyme and will be Tolu\'s main focus once he gets back from holiday: http://cpdnbeta.oerc.ox.ac.uk/forum_thread.php?id=41 I'm a volunteer and my views are my own. News and Announcements and FAQ |
Send message Joined: 10 Dec 05 Posts: 5 Credit: 175,291 RAC: 0 |
What should i do? Trickles are sending to cpdn - but no one is showing on the site. I don\'t know that they arrive the server |
Send message Joined: 7 Aug 04 Posts: 2168 Credit: 64,541,825 RAC: 6,664 |
What should i do? Trickles are sending to cpdn - but no one is showing on the site. I don\'t know that they arrive the server That\'s up to you. If you continue running it, it will likely continue to the end and succeed. However, you will get no new trickles listed (and no new credits on it) until it passes the point at which it went back to zero. The speed of that work unit, and the benchmarks for that PC, suggest significant overclocking. While that may have not had anything to do with the problem, if more problems crop up, it\'d be something to look at. |
©2024 climateprediction.net