Seti limits access due to perceived on time of computer

Message boards : Number crunching : Seti limits access due to perceived on time of computer
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile John

Send message
Joined: 5 Jun 99
Posts: 30
Credit: 77,663,734
RAC: 236
United States
Message 557048 - Posted: 30 Apr 2007, 23:21:48 UTC

I have gotten a message on one of my computers that in affect says: No work Sent: won't finish in time-computer on 17%, Boinc on 100% Projects gets 50%
I am a Snowbird (a person who goes South to warmer climates in the winter) and I leave two computers at home. It does run 100% after I get back, but Seti is stingy with its allocation - see above message. It seems to me that the powers that be in Seti have out smarted themselves as they can not correctly figure out the computer on time. Yes , I know they are figuring it based on the last 6 months or so....That is the problem. I have moved on to Rosetta which is what I guess they want, but I would like them to fix this glaring error. John
ID: 557048 · Report as offensive
Alinator
Volunteer tester

Send message
Joined: 19 Apr 05
Posts: 4178
Credit: 4,647,982
RAC: 0
United States
Message 557067 - Posted: 30 Apr 2007, 23:44:33 UTC
Last modified: 30 Apr 2007, 23:48:33 UTC

Hmmmm, you are correct that an extended off period will cause BOINC to make a big correction to the percent uptime metric. However, that's no reason to bash SAH because when you start it back up it thinks it can't make the deadline.

In fact all you need to do is just let the machine run with the project(s) that will DL and as soon as the uptime percentage rises enough SAH will resume on it's own. Another workaround to force BOINC to resume immediately is to edit the uptime parameter in the client_state file back to 1.000.

IOW, it's not an error, both ends are doing exactly what they were designed to do.

Alinator
ID: 557067 · Report as offensive
Profile John

Send message
Joined: 5 Jun 99
Posts: 30
Credit: 77,663,734
RAC: 236
United States
Message 557105 - Posted: 1 May 2007, 2:47:38 UTC

Thanks for your reply. Think I will give your suggestion to edit the uptime parameter a try. Later.
ID: 557105 · Report as offensive
Alinator
Volunteer tester

Send message
Joined: 19 Apr 05
Posts: 4178
Credit: 4,647,982
RAC: 0
United States
Message 557109 - Posted: 1 May 2007, 3:08:27 UTC
Last modified: 1 May 2007, 3:13:14 UTC

Just a note on editing the client_state file, you need to put six digits after the decimal point in order for it to 'take', so replace the value that's there with 1.000000. This is just a quirk with BOINC, but had me scratching my head for a while the first time I was poking around in there.

The xml element you want to modify is <on_frac> in the <time_stats> section, and you can use notepad, or the edit mode of IE to perform the modification.

Alinator
ID: 557109 · Report as offensive
Profile John

Send message
Joined: 5 Jun 99
Posts: 30
Credit: 77,663,734
RAC: 236
United States
Message 557851 - Posted: 2 May 2007, 2:44:37 UTC
Last modified: 2 May 2007, 2:48:05 UTC

Thanks again. I discovered the "name" of the parameter by the process of deduction. Since it started out with .17xxxx, I just changed the first digit to a 9 making it over 90% active and just in case modified the backup as well. It worked as both computers immediately requested additional work. Today being Tuesday the scheduled down time has made contact, at best, sporadic. At least I now have enough work to carry me through. One was on OS/2 the other was on Windows XP. I like the fact this project can be done with multiple operating systems. Later. John
ID: 557851 · Report as offensive

Message boards : Number crunching : Seti limits access due to perceived on time of computer


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.