Welcome to MilkyWay@home

BOINC 6.10.4 released.

Message boards : Number crunching : BOINC 6.10.4 released.
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile [SETI.USA]Tank_Master

Send message
Joined: 2 Sep 07
Posts: 24
Credit: 26,244,596
RAC: 0
Message 30383 - Posted: 8 Sep 2009, 19:56:58 UTC

So far for windows only.
This build should finally have proper ATI support from the client site!

http://boinc.berkeley.edu/download_all.php
ID: 30383 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Paul D. Buck

Send message
Joined: 12 Apr 08
Posts: 621
Credit: 161,934,067
RAC: 0
Message 30388 - Posted: 8 Sep 2009, 21:31:50 UTC

I just got MW working again ... with 6.10.3 and the 19g CP ATI application ...

Now what would I have to change to keep it working with 6.10.4?

It has been months of me tinkering off and on when I had time and health to get this going ...

I sure would hate to mess it up ...
ID: 30388 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Emanuel

Send message
Joined: 18 Nov 07
Posts: 280
Credit: 2,442,757
RAC: 0
Message 30390 - Posted: 8 Sep 2009, 22:23:30 UTC - in response to Message 30388.  

If it works, I wouldn't mess with it. If and when the CAL app is officially integrated into the Milkyway@Home project, then I'd change to a version that has all the features it needs and as many bugs fixed as possible. Until then, why bother?
ID: 30390 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Divide Overflow
Avatar

Send message
Joined: 16 Feb 09
Posts: 109
Credit: 11,089,510
RAC: 0
Message 30391 - Posted: 8 Sep 2009, 22:25:55 UTC

I see some strange behavior with 6.10.4, it seems to want to crunch the most recently downloaded tasks. I don't really like it working this way.
ID: 30391 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
zpm

Send message
Joined: 27 Feb 09
Posts: 41
Credit: 123,828
RAC: 0
Message 30393 - Posted: 8 Sep 2009, 23:14:30 UTC - in response to Message 30391.  

a system that was turned off in 6 series is now turned back on.


perfect title to the thread.

http://boinc.berkeley.edu/dev/forum_thread.php?id=4268#27176

I recommend Secunia PSI: http://secunia.com/vulnerability_scanning/personal/
ID: 30393 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Divide Overflow
Avatar

Send message
Joined: 16 Feb 09
Posts: 109
Credit: 11,089,510
RAC: 0
Message 30397 - Posted: 9 Sep 2009, 3:24:16 UTC

The suggested fix did not correct the problem for me. After making the adjustment to <on_frac>, the task duration estimates look fine. I do not show any tasks running in high priority, yet BOINC continues to immediately switch to the most recently download task.
ID: 30397 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Paul D. Buck

Send message
Joined: 12 Apr 08
Posts: 621
Credit: 161,934,067
RAC: 0
Message 30405 - Posted: 9 Sep 2009, 10:59:39 UTC - in response to Message 30397.  

The suggested fix did not correct the problem for me. After making the adjustment to <on_frac>, the task duration estimates look fine. I do not show any tasks running in high priority, yet BOINC continues to immediately switch to the most recently download task.

That is actually another long standing bug/issue ...

The DCF error causes the tasks to go into high priority mode until the estimates are all back to normal.
ID: 30405 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Zanth
Avatar

Send message
Joined: 18 Feb 09
Posts: 158
Credit: 110,699,054
RAC: 0
Message 30530 - Posted: 12 Sep 2009, 2:06:44 UTC - in response to Message 30397.  

The suggested fix did not correct the problem for me. After making the adjustment to <on_frac>, the task duration estimates look fine. I do not show any tasks running in high priority, yet BOINC continues to immediately switch to the most recently download task.

This is causing me a lot of problems, it magically stops processing units every few hours. Its a quick fix to get it going again, but a bit annoying. But if it allows the BOINC client to properly send ATI info to the servers and all that, its a good step.
ID: 30530 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile arkayn
Avatar

Send message
Joined: 14 Feb 09
Posts: 999
Credit: 74,932,619
RAC: 0
Message 30532 - Posted: 12 Sep 2009, 2:17:44 UTC

I am going back to 6.10.3 until they fix the bug.
ID: 30532 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Joseph Stateson
Avatar

Send message
Joined: 18 Nov 08
Posts: 291
Credit: 2,461,693,501
RAC: 0
Message 30664 - Posted: 13 Sep 2009, 13:28:39 UTC - in response to Message 30530.  

The suggested fix did not correct the problem for me. After making the adjustment to <on_frac>, the task duration estimates look fine. I do not show any tasks running in high priority, yet BOINC continues to immediately switch to the most recently download task.

This is causing me a lot of problems, it magically stops processing units every few hours. Its a quick fix to get it going again, but a bit annoying. But if it allows the BOINC client to properly send ATI info to the servers and all that, its a good step.


I have the same problem. Last night I had all 24 milkyway tasks "waiting to run" plus another 6 or so that were stuck uploading. Any cuda type task (seti, milkway, collatz, einstein) that is downloaded gets started immediately. I got a computation error on seti cuda with 2 seconds of it downloading and starting. One bug I noticed is that the 24 task limit is exceeded if there are any "stuck uploading" milkyway tasks. I upgraded three GeForce (cuda) type systems to 6.10.4 last night (big mistake) and all have the same problem. Setting the on_frac to 1.0 did nothing. Only one system has the 2xx type GeForce so that one has the 24+ milkyway problem. The other can't run milkyway as they have 9800 type gpu's

I managed to clear the backlog by disableing all new tasking on all cuda type projects. However, none of the milkyway managed to upload. They were all stuck uploading this morning. I rebooted and cleared them. Right now I have 13 milkway "waiting to run" with a seti cuda running. When that seti swaps out I suspect the remaining milkyway will all go to "waiting to run". Fortunately, collatz is not reporting jobs or I would have even more "waiting to run".
ID: 30664 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : Number crunching : BOINC 6.10.4 released.

©2024 Astroinformatics Group