OK, Clarification Requested!

Message boards : Number crunching : OK, Clarification Requested!
Message board moderation

To post messages, you must log in.

AuthorMessage
Alinator
Volunteer tester

Send message
Joined: 19 Apr 05
Posts: 4178
Credit: 4,647,982
RAC: 0
United States
Message 367039 - Posted: 14 Jul 2006, 16:09:33 UTC

What is the maximum deferral interval BOINC will impose when communication fails?

I seem to recall 1 week being mentioned, but one of my remote boxes got "gagged" for about 8 days for SAH after last weeks maintenance outage.

Alinator
ID: 367039 · Report as offensive
Profile Steve Cressman
Volunteer tester
Avatar

Send message
Joined: 6 Jun 02
Posts: 583
Credit: 65,644
RAC: 0
Canada
Message 367040 - Posted: 14 Jul 2006, 16:13:33 UTC - in response to Message 367039.  
Last modified: 14 Jul 2006, 16:14:05 UTC

What is the maximum deferral interval BOINC will impose when communication fails?

I seem to recall 1 week being mentioned, but one of my remote boxes got "gagged" for about 8 days for SAH after last weeks maintenance outage.

Alinator

If it fails to get the master file list four times then it backs off for one week or 604800 sec as seen in message tab or 168 hours as seen under the projects tab.
:)
98SE XP2500+ @ 2.1 GHz Boinc v5.8.8

And God said"Let there be light."But then the program crashed because he was trying to access the 'light' property of a NULL universe pointer.
ID: 367040 · Report as offensive
Alinator
Volunteer tester

Send message
Joined: 19 Apr 05
Posts: 4178
Credit: 4,647,982
RAC: 0
United States
Message 367045 - Posted: 14 Jul 2006, 16:19:01 UTC

That's what I thought, so I figured after seven days went by for this remote box the science app had faulted.

I finally had a chance to get to it last night and discovered it was fine but still had 17 hours+ left on a comm deferral, which would have made it about 9 days total.

As you said, the trigger for the "biggie" was a failure to get the master file.

Alinator
ID: 367045 · Report as offensive
Andrew Waddington
Volunteer tester

Send message
Joined: 1 Jun 99
Posts: 69
Credit: 55,557
RAC: 0
Australia
Message 367463 - Posted: 15 Jul 2006, 0:44:25 UTC - in response to Message 367039.  

What is the maximum deferral interval BOINC will impose when communication fails?

I seem to recall 1 week being mentioned, but one of my remote boxes got "gagged" for about 8 days for SAH after last weeks maintenance outage.

Alinator


I've seen one of my projects defer for a month.

Andrew
ID: 367463 · Report as offensive
Alinator
Volunteer tester

Send message
Joined: 19 Apr 05
Posts: 4178
Credit: 4,647,982
RAC: 0
United States
Message 367482 - Posted: 15 Jul 2006, 0:52:05 UTC - in response to Message 367463.  

What is the maximum deferral interval BOINC will impose when communication fails?

I seem to recall 1 week being mentioned, but one of my remote boxes got "gagged" for about 8 days for SAH after last weeks maintenance outage.

Alinator


I've seen one of my projects defer for a month.

Andrew


UGH, that seems a bit excessive! :-)

Alinator
ID: 367482 · Report as offensive
John McLeod VII
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 15 Jul 99
Posts: 24806
Credit: 790,712
RAC: 0
United States
Message 368931 - Posted: 16 Jul 2006, 2:21:03 UTC

The max deferral length depends on the version of BOINC. In some versions there was no max (but I believe these are all long gone 2.x?). It was reduced to a month I don't remember when this occurred. Then it was reduced to 2 weeks, and later to 1 week.


BOINC WIKI
ID: 368931 · Report as offensive
Alinator
Volunteer tester

Send message
Joined: 19 Apr 05
Posts: 4178
Credit: 4,647,982
RAC: 0
United States
Message 369348 - Posted: 16 Jul 2006, 15:07:05 UTC - in response to Message 368931.  

The max deferral length depends on the version of BOINC. In some versions there was no max (but I believe these are all long gone 2.x?). It was reduced to a month I don't remember when this occurred. Then it was reduced to 2 weeks, and later to 1 week.


That's what I thought, but the 9 day deferral came on a 5.2.13 version. It was Trux' client, but I don't see why he would have changed the algorithm. I suppose it's possible though.

The records have been purged so you can't see for yourself, but for this unit I had done a manual update on 7/05 at about 14:00 UTC, and the next contact was on was on 7/13 at 23:41:07 UTC when I did another manual update, and as I mentioned it had 17+ hours to go on the deferral, which by my reckoning would have been ~9days 2 hours.

Curious.

Just as a side note here, with the reduction on server resources with the release of enhanced, I fail to see why we have to be hustling the old results out of the database so quickly. It makes it a lot harder to detect and track anomalies, especially with slower machines.

This could be a partial reason for user retention problems, since there seems to be a trend for a lot projects to devise new apps and set the project configuration with the assumption everyone is running a quad dual core Opty 285. Logic clearly indicates the vast majority of machines in the field are far less capable than that. (Note: This is a slight exaggeration for dramatic impact!) :-)

Alinator
ID: 369348 · Report as offensive

Message boards : Number crunching : OK, Clarification Requested!


 
©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.