Message boards : Number crunching : 1% Progress and Then?
Author | Message |
---|---|
Penguirl Send message Joined: 5 Nov 05 Posts: 3 Credit: 364,381 RAC: 0 |
Each of my three Macs has started a WU today and after 1.5 - 6.5 hours, they all are at 1.xxx% completion. Should I Suspend & Resume them, or Abort them? I saw something about this somewhere but was under the impression that it was a bug that had been addressed. |
Cureseekers~Kristof Send message Joined: 5 Nov 05 Posts: 80 Credit: 689,603 RAC: 0 |
Don't abort it. If it is an error in the WU, it will be detected and aborted automatically. But it's possible that it's a complete normal WU. To explain this (you can find this story in several postings on this board): * When a WU is started, it is getting initialized. If this is finished, you get 1.xx% done. * After the first model is finished, it percentage increases. * If it is a complex model, it will last longer than a simple model. * Depending on your settings (for the default runtime), after this first model it will take a look if it is going to start a second model * etc So it is possible that you'll see the WU 'hangs' at 1% and after several hours it goes almost immidiatly to 100% Member of Dutch Power Cows |
Penguirl Send message Joined: 5 Nov 05 Posts: 3 Credit: 364,381 RAC: 0 |
Thank you. Hopefully when I get up in the morning all will be good! |
Jose Send message Joined: 28 Mar 06 Posts: 820 Credit: 48,297 RAC: 0 |
Each of my three Macs has started a WU today and after 1.5 - 6.5 hours, they all are at 1.xxx% completion. Should I Suspend & Resume them, or Abort them? If they are stuck: the Terminator function in the program will catch it. Do not abort!!!. Please remember that unless a "decoy is "built"progress seems to stop at 1.00..%. The program only updates progress % after a model/decoy is "built. This and no other is the root from which a Tyrant springs; when he first appears he is a protector.†Plato |
Message boards :
Number crunching :
1% Progress and Then?
©2024 University of Washington
https://www.bakerlab.org