climateprediction.net home page
Visual Fortran Runtime error
Visual Fortran Runtime error
log in

Advanced search

Questions and Answers : Windows : Visual Fortran Runtime error

1 · 2 · Next
Author Message
Matt
Send message
Joined: 20 Mar 14
Posts: 1
Credit: 4,749,697
RAC: 10,191
Message 49792 - Posted: 20 Aug 2014, 17:02:52 UTC

Getting the following error messages for the last week whenever I've included CPDN tasks in BOINC.

"forrtl: severe (17) syntax error in NAMELIST input, unit 5, file: <snip>, line 393, position 19"

I'm wondering if anyone else is having the same issue, or if this is just something peculiar going on with my machine.

Thanks,
Matt

[img][/img]

SuperSluether
Send message
Joined: 6 Jul 14
Posts: 11
Credit: 353,865
RAC: 0
Message 49843 - Posted: 23 Aug 2014, 20:46:51 UTC - in response to Message 49792.

Oh thank goodness, I'm not crazy. I have been having the same exact error all day! I have no idea what's causing it. When I abort the task, it shows the error a few more times, then says Computation Error on the BOINC Manager.

Les Bayliss
Volunteer moderator
Send message
Joined: 5 Sep 04
Posts: 6408
Credit: 16,340,531
RAC: 19,391
Message 49844 - Posted: 23 Aug 2014, 20:55:49 UTC

This is something that happens to some computers, some times.
There's no known reason, or cure. On those which get it, it usually stops by itself eventually.

It's "Visual Fortran Runtime error", because this is the program language that's used for most of the programs here.

There are several threads about it somewhere, but possibly not for a few years now.


SuperSluether
Send message
Joined: 6 Jul 14
Posts: 11
Credit: 353,865
RAC: 0
Message 49845 - Posted: 23 Aug 2014, 22:06:18 UTC - in response to Message 49844.


There's no known reason, or cure.


Why do computers have to be so random... (headshake)



It's "Visual Fortran Runtime error", because this is the program language that's used for most of the programs here.



It looks like Intel Visual Fortran Runtime is a version of C++.

Profile Iain Inglis
Volunteer moderator
Send message
Joined: 16 Jan 10
Posts: 866
Credit: 36,969
RAC: 968
Message 49846 - Posted: 23 Aug 2014, 23:06:43 UTC

The missing image from the first post may look something like this:


I changed one Windows BOINC to a non-service install to avoid the potential PNW 7.22 problem (and HADCM3S 7.24?). However, to no avail for HADCM3S - as the image shows. Crashed four straight HADCM3S, in what looks like a model configuration problem to me. The FORTRAN namelist line number should help. However, I've run out of models and quota for today: perhaps someone could dig out the namelist and see what's on line 393. (Should have kept one: silly me.)

Albert H.
Send message
Joined: 18 Feb 06
Posts: 40
Credit: 12,903,706
RAC: 26,078
Message 49855 - Posted: 25 Aug 2014, 16:10:40 UTC

I had the same run-time error, but only on the machine crunching "short" models.
They all have computation error after a few minutes.
Stopped download of this application and am waiting for better times.
Maybe someone can help.
Thanks
____________

Tristis Oris
Avatar
Send message
Joined: 29 Aug 14
Posts: 4
Credit: 8,985
RAC: 0
Message 49973 - Posted: 2 Sep 2014, 19:18:44 UTC

get this error many times for each of "short" jobs what i taken.
line 393 position 19.

Pete McCann
Send message
Joined: 8 Mar 06
Posts: 24
Credit: 6,620,013
RAC: 11,529
Message 50179 - Posted: 15 Sep 2014, 12:24:01 UTC

Hi.
I'm getting these same error messages on all of my short models which try and run on my machine. Do I leave them running or shall I just abort all the short models from the queue? They are currently processing just one at a time as I have other models running on the other cores, but I have a queue of about 5 or 6 of these short models coming up. Please advise.
Cheers.
Pete

Mike.Gibson
Send message
Joined: 2 May 07
Posts: 20
Credit: 657,542
RAC: 0
Message 50183 - Posted: 15 Sep 2014, 17:12:56 UTC - in response to Message 49846.

Iain

That is exactly the message I have been getting - same line & position number, so it is indicating where the failure is, whatever it is at that point of the program.

I was having no problems on any of the other applications - just HADCM3S.

My cache either failed with a calculation error or were aborted by me. All of them registered a % done, some over 10%, and elapsed time but no CPU time. That resulted in an empty cache which was filled with the same application, so the process was repeated. Is there any way to restrict which applications are downloaded? It won't let me have any more until tomorrow.

Pete, if you have other applications available I would abort all HADCM3S units.

Cheers

Mike

Profile Iain Inglis
Volunteer moderator
Send message
Joined: 16 Jan 10
Posts: 866
Credit: 36,969
RAC: 968
Message 50184 - Posted: 15 Sep 2014, 18:09:01 UTC - in response to Message 50183.

That is exactly the message I have been getting - same line & position number, so it is indicating where the failure is, whatever it is at that point of the program.
... I did come up with a theory as to what the error might have been in the relevant input file, but convinced myself that it wasn't supposed to be an error in a compliant FORTRAN environment. Too bad.
Is there any way to restrict which applications are downloaded?
Select the "Your account" option in the menu to the left of this window, log in if required, then select "climateprediction.net preferences". That allows you to pick which models to run.

Pete McCann
Send message
Joined: 8 Mar 06
Posts: 24
Credit: 6,620,013
RAC: 11,529
Message 50185 - Posted: 15 Sep 2014, 18:11:12 UTC

Hi Mile.
Thanks for the input. There is obviously something about these "short" models that my PC doesn't like. I think I'll ditch the ones in my queue and hope they are replaced with something more reliable. I'm sure there are plenty of other people out there who can run these short models without any trouble.
Cheers.
Pete

Mike.Gibson
Send message
Joined: 2 May 07
Posts: 20
Credit: 657,542
RAC: 0
Message 50187 - Posted: 15 Sep 2014, 20:27:37 UTC - in response to Message 50184.

Thanks, Iain

I had already received an ANZ unit and that is working fine. I have unchecked HADCM3S as you suggested.

Regards

Mike

Mike.Gibson
Send message
Joined: 2 May 07
Posts: 20
Credit: 657,542
RAC: 0
Message 50188 - Posted: 15 Sep 2014, 20:32:13 UTC - in response to Message 50185.

Pete

There is no need to hope you will get something different. I have just followed Iain's advice and unchecked HADCM3S (see 3 messages earlier than this one), so should never get one unless there is nothing else.

Mike

Les Bayliss
Volunteer moderator
Send message
Joined: 5 Sep 04
Posts: 6408
Credit: 16,340,531
RAC: 19,391
Message 50189 - Posted: 15 Sep 2014, 20:41:25 UTC - in response to Message 50188.

And you don't need to "get something else" either. Just uncheck the isolated box below the others.

Mike.Gibson
Send message
Joined: 2 May 07
Posts: 20
Credit: 657,542
RAC: 0
Message 50193 - Posted: 15 Sep 2014, 23:47:47 UTC - in response to Message 50189.

Les

I was thinking that maybe there might be new applications about by then. Leaving it checked would allow for that.

Mike

Mike.Gibson
Send message
Joined: 2 May 07
Posts: 20
Credit: 657,542
RAC: 0
Message 50195 - Posted: 16 Sep 2014, 1:23:09 UTC - in response to Message 50193.

However, before I could uncheck it, 14 downloaded because there wasn't anything else available at the time.

It is strange but 13 of the new units had the error and one is running correctly. This means overall only one out of over 30.

Mike

Mike.Gibson
Send message
Joined: 2 May 07
Posts: 20
Credit: 657,542
RAC: 0
Message 50196 - Posted: 16 Sep 2014, 1:27:39 UTC - in response to Message 50195.

Sorry, I spoke too soon. It was seemingly running ok because it was racking up CPU time which the others didn't but it still crashed with a computation error.

Mike

Pete McCann
Send message
Joined: 8 Mar 06
Posts: 24
Credit: 6,620,013
RAC: 11,529
Message 50215 - Posted: 16 Sep 2014, 17:21:13 UTC

Hi Mike. Thanks for the pointer. I found the checkbox after a bit of hunting about, so hopefully that is the last we shall see of the short models and all their error messages. I've had to abort about 20 of them though! They just kept on coming. While I was hunting about I came across the place where you can change the number of cores used for project work. I had been wondering why I was only processing 4 models at once when I have 8 available. Hopefully that will up my game a bit.
Cheers.
Pete

Michael Polla
Avatar
Send message
Joined: 24 Jul 06
Posts: 3
Credit: 980,790
RAC: 0
Message 52389 - Posted: 6 Aug 2015, 13:19:16 UTC

Got the same message recently, and thanks to the advice provided here, I unchecked the "HADCM3 Small" choice in project preferences, aborted the ones actually running and now all is doing great again.

Davissimo
Avatar
Send message
Joined: 30 Jul 08
Posts: 3
Credit: 254,585
RAC: 0
Message 53076 - Posted: 12 Dec 2015, 4:47:55 UTC

It seems climateprediction is a bit too fragile. An application was running for almost a week then Microsoft updates arrived requiring a restart. CliPred was suspended first then the restart done. The fortran error began occurring once the system was back up. I had to abort the task. Perhaps I might try another task but these problems are discouraging and I don't want to waste good tasks. I am using Windows 10 and Kaspersky internet security. Are these a problem currently?

____________

1 · 2 · Next

Questions and Answers : Windows : Visual Fortran Runtime error


Main page · Your account · Message boards


Copyright © 2017 climateprediction.net