Files camb_scalarcls.chk and camb_tensorcls.chk

Message boards : BOINC client : Files camb_scalarcls.chk and camb_tensorcls.chk
Message board moderation

To post messages, you must log in.

AuthorMessage
Ed Meadows
Avatar

Send message
Joined: 28 Jun 07
Posts: 22
United States
Message 15003 - Posted: 17 Jan 2008, 18:57:27 UTC
Last modified: 17 Jan 2008, 19:39:57 UTC

These files are in c:program files/boinc/BOINC/slots/2 and take up 54 and 11 mb, respectively. The file type is labeled as "Recovered File Fragments".

They are growing larger and are frequently badly fragmented. What are these, and is something going wrong that I need to know about? Should I do something to manage these files?

I don't see any evidence that anything is failing with my work units.

Thanks,
Ed
ID: 15003 · Report as offensive
Profile Jord
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 29 Aug 05
Posts: 15477
Netherlands
Message 15005 - Posted: 17 Jan 2008, 19:13:53 UTC

They are the Cosmology checkpoint files. If you delete them, you'll lose your tasks in progress, them having to restart from zero.

These files will get deleted once the Cosmology task is done.
ID: 15005 · Report as offensive
Ed Meadows
Avatar

Send message
Joined: 28 Jun 07
Posts: 22
United States
Message 15007 - Posted: 17 Jan 2008, 19:24:00 UTC

I take care to keep my system very well tuned, including defragmenting my hard drive every other day. That's what brought my attention to these two files in the first place - in the defragmenter's report.

So these are Cosmology files... since they are constantly growing and Cosmology WUs are flowing through my system normally, its apparent that this file isn't deleted once WUs are completed.

I'll just drain the Cosmology project queue, detach, and reattach to the project and this should get rid of the files, so they can reappear and start growing again :^(

Thanks for the information.

Ed
ID: 15007 · Report as offensive
Profile Jord
Volunteer tester
Help desk expert
Avatar

Send message
Joined: 29 Aug 05
Posts: 15477
Netherlands
Message 15009 - Posted: 17 Jan 2008, 20:28:02 UTC
Last modified: 17 Jan 2008, 20:29:10 UTC

It's possible that the same slot is used for the Cosmo tasks, especially if you have a multiprocessor machine. The checkpoint files cannot be deleted when one task is in progress. Checking your system I see you have two CPUs, so it's possible that while one task is done, uploaded & reported, another is still busy, adding to these files.
ID: 15009 · Report as offensive
Ed Meadows
Avatar

Send message
Joined: 28 Jun 07
Posts: 22
United States
Message 15010 - Posted: 17 Jan 2008, 20:41:53 UTC

Sekerob - thanks, I always suspend all projects and then completely exit BOINC manager before defragging. I also back up the BOINC directory while I'm out.

Ageless - this makes sense. There is always more Cosmology@Home WUs in queue, some of which may have some compute time, preventing these large files from resetting.

I've reposted my question to the Cosmology@Home forum to see what they say.

Ed
ID: 15010 · Report as offensive

Message boards : BOINC client : Files camb_scalarcls.chk and camb_tensorcls.chk

Copyright © 2024 University of California.
Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.