Welcome to MilkyWay@home

Big issue : Milkiway doesn't work on Mac OS X.

Message boards : Number crunching : Big issue : Milkiway doesn't work on Mac OS X.
Message board moderation

To post messages, you must log in.

1 · 2 · 3 · Next

AuthorMessage
JiPaulB

Send message
Joined: 7 Feb 10
Posts: 2
Credit: 73,338
RAC: 0
Message 39053 - Posted: 23 Apr 2010, 13:09:53 UTC

Hello,

I'm on Mac OS X (Leopard 10.5.8)

Each time the Milkyway task is resuming (after suspending because of computer in use), the task restarts from beginning. That is very unefficient !...

Am I alone with this issue ?
Is this an issue with Mac OS X version of Milkyway only ?

Please, note that I haven't this problem with other projects.
ID: 39053 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Zydor
Avatar

Send message
Joined: 24 Feb 09
Posts: 620
Credit: 100,587,625
RAC: 0
Message 39056 - Posted: 23 Apr 2010, 14:22:40 UTC

Its a known issue - see:

OS X Checkpointing
Its just time & space to get it all done for all OS's for the new app

Regards
Zy
ID: 39056 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Martin P.

Send message
Joined: 21 Nov 07
Posts: 52
Credit: 1,756,052
RAC: 0
Message 39283 - Posted: 29 Apr 2010, 12:18:25 UTC - in response to Message 39056.  

Zydorg,

this discussion is about Milkyway3, but I see the same problem with Milkyway 0.26!

ID: 39283 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 39291 - Posted: 29 Apr 2010, 20:23:18 UTC - in response to Message 39283.  

I should have the 0.26 binaries updated today.

--Travis
ID: 39291 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Zydor
Avatar

Send message
Joined: 24 Feb 09
Posts: 620
Credit: 100,587,625
RAC: 0
Message 39300 - Posted: 30 Apr 2010, 10:08:04 UTC - in response to Message 39283.  
Last modified: 30 Apr 2010, 10:08:40 UTC

Zydorg,

this discussion is about Milkyway3, but I see the same problem with Milkyway 0.26!


Milkyway 3 is the version number of the WU, 0.26 is the version number of the exe file running the WU (for your OS X). 0.26 needs an update to the binaries to get it to run the new V3 WU.

When a new WU runs ok on one OS and all looks good, it then takes time to go through the exe files for the other O/S to get them to be compatible, yours should be done real soon now (see Travis post above)

Regards
Zy
ID: 39300 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
MacRonin

Send message
Joined: 23 Jan 10
Posts: 4
Credit: 172,174
RAC: 0
Message 39391 - Posted: 3 May 2010, 4:58:38 UTC
Last modified: 3 May 2010, 5:10:28 UTC

I'm still having the work unit completion reset problem under OSX 10.5.8

Once again I've had a bunch of WU stuck in this mode for the last few days. I just canceled the ones currently on my system and temporarily suspended the project. I'll start it back up again when the checkpoint function is working correctly.
ID: 39391 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
paris
Avatar

Send message
Joined: 26 Apr 08
Posts: 87
Credit: 64,801,496
RAC: 0
Message 39420 - Posted: 4 May 2010, 22:45:38 UTC

My Mac mini (core duo, OS X 10.4.11, MW 0.26) has been slowly completing units successfully but for several days now it will not validate against other machines. Shouldn't a stock machine running a stock app validate properly? How about awarding half credit for successful completion but non-validation against the non-stock apps? :) I may be clear off base here and I fully realize that my contribution is minimal compared to the GPU cards completing in minutes what takes me hours to finish but I like the project and I would like to get at least some credit for doing successful work. Can someone straighten me out on this?


Plus SETI Classic = 21,082 WUs
ID: 39420 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 39440 - Posted: 5 May 2010, 20:44:41 UTC - in response to Message 39420.  

I've updated the OS X binaries, let me know if this fixes the problems.
ID: 39440 · 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 39444 - Posted: 6 May 2010, 0:58:48 UTC - in response to Message 39440.  

Looks like check-pointing is working on my iMac.
ID: 39444 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
paris
Avatar

Send message
Joined: 26 Apr 08
Posts: 87
Credit: 64,801,496
RAC: 0
Message 39474 - Posted: 6 May 2010, 23:08:49 UTC
Last modified: 6 May 2010, 23:31:28 UTC

Argh. After 20 hours of computing on each of two cores I'm getting an error:

<file_xfer_error>
<file_name>de_new_test2_98092_1273111747_0_0</file_name>
<error_code>-161</error_code>
</file_xfer_error>

I had allowed the automatic update to 0.27 to occur and aborted the units that required 0.26. The new work units were automatically downloaded and errored out at the very end - apparently because some file cannot be found. I did not make any manual changes to MilkyWay/BOINC. Am I supposed to hand modify the xml file(s)? Advice appreciated. (And for good measure, I've now reset the project.)

Mac OS X 10.4.11, core duo mini, BOINC 6.6.20, stock app.


Plus SETI Classic = 21,082 WUs
ID: 39474 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 39478 - Posted: 7 May 2010, 3:32:40 UTC - in response to Message 39474.  

Argh. After 20 hours of computing on each of two cores I'm getting an error:


de_new_test2_98092_1273111747_0_0
-161


I had allowed the automatic update to 0.27 to occur and aborted the units that required 0.26. The new work units were automatically downloaded and errored out at the very end - apparently because some file cannot be found. I did not make any manual changes to MilkyWay/BOINC. Am I supposed to hand modify the xml file(s)? Advice appreciated. (And for good measure, I've now reset the project.)

Mac OS X 10.4.11, core duo mini, BOINC 6.6.20, stock app.


Think I know what the problem is, I'll be upgrading the OS X applications again, shortly.
ID: 39478 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 39479 - Posted: 7 May 2010, 4:05:43 UTC - in response to Message 39478.  

Updated the applications. They should really work now. Sorry about that.
ID: 39479 · 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 39506 - Posted: 7 May 2010, 20:59:59 UTC - in response to Message 39479.  

This one completed just fine, and a little faster than 0.27 as well.

http://milkyway.cs.rpi.edu/milkyway/result.php?resultid=117108312
ID: 39506 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
MacRonin

Send message
Joined: 23 Jan 10
Posts: 4
Credit: 172,174
RAC: 0
Message 39522 - Posted: 8 May 2010, 10:59:03 UTC - in response to Message 39479.  

I just finished two 0.27 WU's. They grabbed the system and wouldn't let go for a long time, but the percentages did increase. But then they died with these errors.

Sat May 8 00:52:59 2010 Milkyway@home Computation for task de_new_test2_40575_1273100978_1 finished
Sat May 8 00:52:59 2010 Milkyway@home Output file de_new_test2_40575_1273100978_1_0 for task de_new_test2_40575_1273100978_1 absent
Sat May 8 01:31:00 2010 Milkyway@home Computation for task de_new_test2_40355_1273100887_1 finished
Sat May 8 01:31:00 2010 Milkyway@home Output file de_new_test2_40355_1273100887_1_0 for task de_new_test2_40355_1273100887_1 absent
ID: 39522 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 39535 - Posted: 8 May 2010, 21:01:32 UTC - in response to Message 39522.  

I just finished two 0.27 WU's. They grabbed the system and wouldn't let go for a long time, but the percentages did increase. But then they died with these errors.

Sat May 8 00:52:59 2010 Milkyway@home Computation for task de_new_test2_40575_1273100978_1 finished
Sat May 8 00:52:59 2010 Milkyway@home Output file de_new_test2_40575_1273100978_1_0 for task de_new_test2_40575_1273100978_1 absent
Sat May 8 01:31:00 2010 Milkyway@home Computation for task de_new_test2_40355_1273100887_1 finished
Sat May 8 01:31:00 2010 Milkyway@home Output file de_new_test2_40355_1273100887_1_0 for task de_new_test2_40355_1273100887_1 absent


This should be fixed in the new version 0.28.
ID: 39535 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
paris
Avatar

Send message
Joined: 26 Apr 08
Posts: 87
Credit: 64,801,496
RAC: 0
Message 39536 - Posted: 8 May 2010, 23:24:29 UTC

Yes, mine is working (sort of) now, too. That is to say, I'm am able to successfully complete and send in work units but they don't validate against the wingmen that I get. Is there a solution for this or am I wasting my time? I have read the threads that I think are relevant and I can't tell for sure if I have any real hope of helping the project, even if only marginally.


Plus SETI Classic = 21,082 WUs
ID: 39536 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 39539 - Posted: 9 May 2010, 0:07:15 UTC - in response to Message 39536.  
Last modified: 9 May 2010, 0:07:32 UTC

Yes, mine is working (sort of) now, too. That is to say, I'm am able to successfully complete and send in work units but they don't validate against the wingmen that I get. Is there a solution for this or am I wasting my time? I have read the threads that I think are relevant and I can't tell for sure if I have any real hope of helping the project, even if only marginally.


Are you running version 0.28?

Any information is appreciated I really want to get these issues fixed.
ID: 39539 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
paris
Avatar

Send message
Joined: 26 Apr 08
Posts: 87
Credit: 64,801,496
RAC: 0
Message 39541 - Posted: 9 May 2010, 1:15:57 UTC - in response to Message 39539.  

Yes, mine is working (sort of) now, too. That is to say, I'm am able to successfully complete and send in work units but they don't validate against the wingmen that I get. Is there a solution for this or am I wasting my time? I have read the threads that I think are relevant and I can't tell for sure if I have any real hope of helping the project, even if only marginally.


Are you running version 0.28?

Any information is appreciated I really want to get these issues fixed.


Yes I am. I reset yesterday after the problems (which you fixed - thank you) with 0.27 and had two units complete today but neither one validated against the GPU cards that the others were using. I can keep track of the other systems that don't agree with mine if that would help. Most of the ones I recall use ati 13(?) I think, but not all.

Mac OS X 10.4.11, core duo mini, BOINC 6.6.20, stock app 0.28



Plus SETI Classic = 21,082 WUs
ID: 39541 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 39543 - Posted: 9 May 2010, 1:30:49 UTC - in response to Message 39541.  

Yes, mine is working (sort of) now, too. That is to say, I'm am able to successfully complete and send in work units but they don't validate against the wingmen that I get. Is there a solution for this or am I wasting my time? I have read the threads that I think are relevant and I can't tell for sure if I have any real hope of helping the project, even if only marginally.


Are you running version 0.28?

Any information is appreciated I really want to get these issues fixed.


Yes I am. I reset yesterday after the problems (which you fixed - thank you) with 0.27 and had two units complete today but neither one validated against the GPU cards that the others were using. I can keep track of the other systems that don't agree with mine if that would help. Most of the ones I recall use ati 13(?) I think, but not all.

Mac OS X 10.4.11, core duo mini, BOINC 6.6.20, stock app 0.28


Updating the applications (AGAIN) in a few minutes. Really hope it should fix things this time around :P
ID: 39543 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
paris
Avatar

Send message
Joined: 26 Apr 08
Posts: 87
Credit: 64,801,496
RAC: 0
Message 39545 - Posted: 9 May 2010, 1:43:33 UTC - in response to Message 39543.  

I'll keep my fingers crossed and I'll let you know the results on my end as soon as I can (about 20 hrs). I hope I'm not causing you any undue trouble with this problem. I suspect that if this this happens on my machine, it affects others with similar setups, too. If there is anything else I can do to help, please let me know.


Plus SETI Classic = 21,082 WUs
ID: 39545 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
1 · 2 · 3 · Next

Message boards : Number crunching : Big issue : Milkiway doesn't work on Mac OS X.

©2024 Astroinformatics Group