Don't know if this is important or even related to E@H but thought I should report it just the same.
This is what I would expect if you are attached to projects whose applications have NOT been built using the very latest version of the BOINC API library (August 12, 2005 or later).
I have made the 0.12 App the official Mac OS X Application. It has been re-numbered version 4.82. I expect that should more-or-less end this message-board thread.
BETA TESTERS: PLEASE:
[1] FINISH YOUR CURRENT Einstein@Home WORK BUT DON'T GET ANY NEW WORK.
[2] THEN RENAME THE app_info.xml FILE TO app_info.xml.old OR DELETE IT.
[3] STOP AND RESTART BOINC.
THE NEXT TIME THAT YOU GET WORK, BOINC SHOULD DOWNLOAD AND USE THE 4.82 APPLICATION VERSION.
Well it didn't. It kept the old app and named the app for the new WU accordingly (12). If I reset the project when the queue empties will it load the right stuff?
It should. However I'm not sure what went wrong before. Bruce didn't mention it: you have to stop the client, then remove the app_info.xml, and the start the client again. The client looks for the file only at startup, so removing it while the client is running has no effect.
Apparently the App couldn't read a temp file it had written before. Looks like a problem with your disk (full?) or permissions (running as different user?).
minervini: I've seen your machine returned three successful results after the client errors. Did you do anything to fix is or was it just a temporary problem?
Phil: Yours is a "segmentation violation", roughly the Mac pendant of a "General Access Violation" on Windows. I don't see any sign of a software problem there, might be the hardware?
minervi: I occasionally see Results like those being returned. Unfortunately some of our error messages which would help us to track the problem seem to get lost on the Mac. We will build a new binary that should help us to track the problem, though it will not fix it immediately.
First Result from 4.19 client
)
First Result from 4.19 client and 0.12 App dropped in. No problems, no noticable change in computation time (8,5h on the Mac mini).
BM
BM
RE: Don't know if this is
)
This is what I would expect if you are attached to projects whose applications have NOT been built using the very latest version of the BOINC API library (August 12, 2005 or later).
Bruce
Unless I hear about problems
)
Unless I hear about problems coming up, in a few days I will make the 0.12 test application the standard Mac OS X application for all Mac OS X users.
Cheers,
Bruce
RE: Unless I hear about
)
No complaints?
BM
BM
I have made the 0.12 App the
)
I have made the 0.12 App the official Mac OS X Application. It has been re-numbered version 4.82. I expect that should more-or-less end this message-board thread.
BETA TESTERS: PLEASE:
[1] FINISH YOUR CURRENT Einstein@Home WORK BUT DON'T GET ANY NEW WORK.
[2] THEN RENAME THE app_info.xml FILE TO app_info.xml.old OR DELETE IT.
[3] STOP AND RESTART BOINC.
THE NEXT TIME THAT YOU GET WORK, BOINC SHOULD DOWNLOAD AND USE THE 4.82 APPLICATION VERSION.
Bruce
RE: Well it didn't. It
)
It should. However I'm not sure what went wrong before. Bruce didn't mention it: you have to stop the client, then remove the app_info.xml, and the start the client again. The client looks for the file only at startup, so removing it while the client is running has no effect.
BM
BM
Apparently the App couldn't
)
Apparently the App couldn't read a temp file it had written before. Looks like a problem with your disk (full?) or permissions (running as different user?).
BM
BM
Anyway, I'm looking into
)
Anyway, I'm looking into this. We've got similar errors from the latest Linux App, might be a bug in the latest BOINC or science code.
BM
BM
minervini: I've seen your
)
minervini: I've seen your machine returned three successful results after the client errors. Did you do anything to fix is or was it just a temporary problem?
BM
BM
Phil: Yours is a
)
Phil: Yours is a "segmentation violation", roughly the Mac pendant of a "General Access Violation" on Windows. I don't see any sign of a software problem there, might be the hardware?
minervi: I occasionally see Results like those being returned. Unfortunately some of our error messages which would help us to track the problem seem to get lost on the Mac. We will build a new binary that should help us to track the problem, though it will not fix it immediately.
BM
BM