I\'ve just had a Visual Fortran runtime error on what, up to now, has been a very reliable box - it\'s completed 4 or 5 classic models and is one third of it\'s way through it\'s second pair of BOINC models.

The error was a failed write to yabsd.out. I\'ve got a copy of the dialog and a dump of the entire CP folder if it helps...

It may be a co-incidence but it seemed to occur at more or less the same time as the automatically scheduled benchmark run.

I rebooted the box and both models (it\'s a P4 HT running XP) have restarted correctly without loss of work (phew!)

Ian

are you running any virus-checkers or anything like that? It could be some other process was checking out the yabsd.out file and the Fortran didn't like that.

> are you running any virus-checkers or anything like that? It could be some
> other process was checking out the yabsd.out file and the Fortran didn't like
> that.

I've got AV software running (who hasn't) but it wasn't doing anything special at the time and shouldn't have been scanning files. I wasn't even using the computer at the time - I was off making a cuppa :-).

As I said before, the only odd thing I could see was that the error occurred at roughly the same time as the scheduled BOINC benchmark run.

Ian

Carl Christensen (the software architect) sent me an email a few days back that the old (win 98, NT ME) op systems had some problems.
.........................
"in case you were trying to run climateprediction.net
on a Win98, WinME, or Win NT4 machine and did not have success,
we have identified and corrected a bug in the Fortran code that was
preventing older operating systems from being able to run.

So if you "Reset", or "Detach" and re-"Attach" to climateprediction.net using your
CPDN/BOINC authenticator key, it should get this new version (4.04) of
Windows code that will run fine on 98/ME/NT4."
...........................
Fortran error has not reappeared since. I how am seeing the
odd behaviour mentioned above (stalling out) once after a "run benchmarks" and once after a Seti Upload.
Not consistant in occurance or should I say I have not seen a pattern
to this except that it happens at process breakpoints.

> Carl Christensen (the software architect) sent me an email a few days back
> that the old (win 98, NT ME) op systems had some problems.
> .........................
> "in case you were trying to run climateprediction.net
> on a Win98, WinME, or Win NT4 machine and did not have success,
> we have identified and corrected a bug in the Fortran code that was
> preventing older operating systems from being able to run.
>
> So if you "Reset", or "Detach" and re-"Attach" to climateprediction.net using
> your
> CPDN/BOINC authenticator key, it should get this new version (4.04) of
> Windows code that will run fine on 98/ME/NT4."
> ...........................
> Fortran error has not reappeared since. I how am seeing the
> odd behaviour mentioned above (stalling out) once after a "run benchmarks" and
> once after a Seti Upload.
> Not consistant in occurance or should I say I have not seen a pattern
> to this except that it happens at process breakpoints.

I've got a Win 98SE machine with version 4.04 running which still gives that Visual Fortran error. It restarts compilation (I was at timestap 500+, and it now is running at timestep 3....).

No I/O error (uploading or downloading), just a dump in the runtime whilst calculating.

CPT..
My issue on the 1 Ghz AMD gateway turned out to be processor
workload and heat related.
Ran HOT CPU 4.2.2 on it and only made it 2 - 3 passes into a 6 hour run
before I started getting CPU related errors.
My Win 2000 and XP boxes have not had any issues.

Just the older 1 Ghz with the win 98.

Best HUnting

George Leaf

I got a Visual Fortran error in Phase 1 after timestep 259248.
Both the Boinc window (showing the 5 tabs) and the graphics window (showing the globe) were open. First a message about reorganizing the files was displayed in the graphics window, then the error message. Then the graphics window closed, and the results were uploaded, and the next work unit started being processed. Both messages were visible only shortly, so I don't remember the exact wordings. This was shown in the message tab:

climateprediction.net - 2004-10-28 22:39:50 - Unrecoverable error for result 2uks_100154282_0 ( - exit code -5 (0xfffffffb))
climateprediction.net - 2004-10-28 22:39:50 - Deferring communication with project for 1 minutes and 0 seconds
climateprediction.net - 2004-10-28 22:39:50 - Computation for result 2uks_100154282 finished
climateprediction.net - 2004-10-28 22:39:50 - Restarting result 2ukt_100154283_0 using hadsm3 version 4.04
climateprediction.net - 2004-10-28 22:39:50 - Started upload of 2uks_100154282_0_1.zip
climateprediction.net - 2004-10-28 22:39:50 - Started upload of 2uks_100154282_0_2.zip
climateprediction.net - 2004-10-28 22:39:56 - Finished upload of 2uks_100154282_0_1.zip
climateprediction.net - 2004-10-28 22:39:56 - Throughput 255 bytes/sec
climateprediction.net - 2004-10-28 22:39:56 - Started upload of 2uks_100154282_0_3.zip
climateprediction.net - 2004-10-28 22:39:57 - Finished upload of 2uks_100154282_0_2.zip
climateprediction.net - 2004-10-28 22:39:57 - Throughput 7073 bytes/sec
climateprediction.net - 2004-10-28 22:39:57 - Started upload of 2uks_100154282_0_4.zip
climateprediction.net - 2004-10-28 22:40:02 - Finished upload of 2uks_100154282_0_3.zip
climateprediction.net - 2004-10-28 22:40:02 - Throughput 255 bytes/sec
climateprediction.net - 2004-10-28 22:40:02 - Finished upload of 2uks_100154282_0_4.zip
climateprediction.net - 2004-10-28 22:40:02 - Throughput 255 bytes/sec
climateprediction.net - 2004-10-28 22:40:02 - Started upload of 2uks_100154282_0_5.zip
climateprediction.net - 2004-10-28 22:40:08 - Finished upload of 2uks_100154282_0_5.zip
climateprediction.net - 2004-10-28 22:40:08 - Throughput 9690 bytes/sec

Windows XP home, Boinc 4.13, hadsm3 4.04.