XANSYS Forum Index
    Register    

FAQ    Search    Memberlist    Usergroups    SmartFeedSmartFeed    Profile    Log in
[Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )
 
Post new topic   Reply to topic    XANSYS Forum Index -> XANSYS
Author Message
joseph.metrisin
User


Joined: 07 May 2009
Posts: 304

PostPosted: Tue Mar 27, 2012 12:01 pm  Reply with quote

From the knowledge database:

#2007215 - I am getting a SIG$SEGV error when running DANSYS on a shared-memory machine. I tried turning MSAVE,OFF, and this seems to help.

Family: Structural Mechanics
Product: Mechanical APDL (f. ANSYS)
Version: 12.1
Area: General
Subarea: N/A

Joseph T Metrisin
Structures Lead
Florida Turbine Technologies, Inc
1701 Military Tr. Suite 110
Jupiter, FL 33458 U.S.A.
+1 (561)427-6346 Office
(561)427-6191 Fax
JMetrisin@fttinc.com

Visit our new website: www.fttinc.com

FTT's public email encryption keys are stored on the FTT Verified Directory at http://keys.fttinc.com

-----------------------------------------------------------------------------------------------------
Confidentiality Note:
The information contained in this transmission and any attachments are proprietary and may be privileged, intended only for the use of the individual or entity named above. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you received this communication in error, please delete the message and immediately notify the sender via the contact information listed above.
-----------------------------------------------------------------------------------------------------

________________________________________
Answer:

INTERNAL

The problem has been fixed for v130.

From Jeff B:

The error is unrelated to the MSAVE command or running models from older versions. It is a memory bug in the DSP solver, which is used as a kernel solver during the preconditioner creation for the PCG solver. It depends on where the memory allocation sits as to when the model will crash or not. Other options, like MSAVE, might simply move the memory around enough to avoid the bug in some cases but is not a guaranteed workaround.

In DSPmainPCG.F we call mf2LSave_sym to save the sym array. In mf2LSave_sym.F we were sometimes writing data beyond the allocated length for the sym array which can cause unexpected behavior. The code never uses those extra values, but accessing them can cause crashes in certain cases on certain platforms.

The mf2LSave_sym.F, mf2LRestore_sym.F and DSPmainPCG.F have been fixed and checked into R13 today to resolve this bug.

No good workaround exists for this bug that would cover all cases. Users can try different -m and -db allocation sizes to try to avoid the bug. Typically, larger -m values and smaller -db values should work best to avoid the bug. Avoid using the typical WB scheme of a -m value that is 50 MB bigger than the -db value. This is more likely to trigger the bug.



-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On Behalf Of Fischer, Richard
Sent: Tuesday, March 27, 2012 2:38 PM
To: ANSYS User Discussion List
Subject: [FTT_SPAM] - [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

The full text of the message is: " An unexpected ( SIG$SEGV ) has occurred... Ansys internal data has been corrupted. Ansys is unable to recover and will terminate"

This error has started occurring when I run scripts. It usually goes away after a reboot, and sometimes I can get it to run when I cut and paste at the command line instead of using /input. Any idea what is causing this? I've been looking for problems in the model, but cant find anything. I am running structural to preturb a structure with pressure, then resuming that database, doing an UPGEOM, meshing the inside with HF119 elements with the FVMESH command, and then it happens. Sometimes it goes for weeks without a problem, then it starts up again.

I googled SIG$SEGV and got a hit on Xansys from Sept 2009. Somebody was running coupled CFD Structural and had the same problem when he went to structural or thermal. His Ansys support guys said it was a memory issue. THis was back at V11-V12. (Deitmar, any update?)

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

Post generated using Mail2Forum (http://www.mail2forum.com)
Back to top
View user's profile Send private message
rick.fischer
User


Joined: 31 Aug 2011
Posts: 171

PostPosted: Tue Mar 27, 2012 12:20 pm  Reply with quote

Very interesting. The error occurs during an FVMESH command on a model in a resumed database. That database has 187 elements, static analysis, no NLGEOM or SSTIF (all the stuff listed as prerequisites for the MSAVE command per TFM) and the solution used the PCG solver. Thanks.

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215

-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On Behalf Of Metrisin, Joe
Sent: Tuesday, March 27, 2012 2:01 PM
To: ANSYS User Discussion List
Subject: Re: [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

Quote:
From the knowledge database:

#2007215 - I am getting a SIG$SEGV error when running DANSYS on a shared-memory machine. I tried turning MSAVE,OFF, and this seems to help.

Family: Structural Mechanics
Product: Mechanical APDL (f. ANSYS)
Version: 12.1
Area: General
Subarea: N/A

Joseph T Metrisin
Structures Lead
Florida Turbine Technologies, Inc
1701 Military Tr. Suite 110
Jupiter, FL 33458 U.S.A.
+1 (561)427-6346 Office
(561)427-6191 Fax
JMetrisin@fttinc.com

Visit our new website: www.fttinc.com

FTT's public email encryption keys are stored on the FTT Verified Directory at http://keys.fttinc.com

-----------------------------------------------------------------------------------------------------
Confidentiality Note:
The information contained in this transmission and any attachments are proprietary and may be privileged, intended only for the use of the individual or entity named above. If the reader of this message is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you received this communication in error, please delete the message and immediately notify the sender via the contact information listed above.
-----------------------------------------------------------------------------------------------------

________________________________________
Answer:

INTERNAL

The problem has been fixed for v130.

Quote:
From Jeff B:

The error is unrelated to the MSAVE command or running models from older versions. It is a memory bug in the DSP solver, which is used as a kernel solver during the preconditioner creation for the PCG solver. It depends on where the memory allocation sits as to when the model will crash or not. Other options, like MSAVE, might simply move the memory around enough to avoid the bug in some cases but is not a guaranteed workaround.

In DSPmainPCG.F we call mf2LSave_sym to save the sym array. In mf2LSave_sym.F we were sometimes writing data beyond the allocated length for the sym array which can cause unexpected behavior. The code never uses those extra values, but accessing them can cause crashes in certain cases on certain platforms.

The mf2LSave_sym.F, mf2LRestore_sym.F and DSPmainPCG.F have been fixed and checked into R13 today to resolve this bug.

No good workaround exists for this bug that would cover all cases. Users can try different -m and -db allocation sizes to try to avoid the bug. Typically, larger -m values and smaller -db values should work best to avoid the bug. Avoid using the typical WB scheme of a -m value that is 50 MB bigger than the -db value. This is more likely to trigger the bug.



-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On Behalf Of Fischer, Richard
Sent: Tuesday, March 27, 2012 2:38 PM
To: ANSYS User Discussion List
Subject: [FTT_SPAM] - [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

The full text of the message is: " An unexpected ( SIG$SEGV ) has occurred... Ansys internal data has been corrupted. Ansys is unable to recover and will terminate"

This error has started occurring when I run scripts. It usually goes away after a reboot, and sometimes I can get it to run when I cut and paste at the command line instead of using /input. Any idea what is causing this? I've been looking for problems in the model, but cant find anything. I am running structural to preturb a structure with pressure, then resuming that database, doing an UPGEOM, meshing the inside with HF119 elements with the FVMESH command, and then it happens. Sometimes it goes for weeks without a problem, then it starts up again.

I googled SIG$SEGV and got a hit on Xansys from Sept 2009. Somebody was running coupled CFD Structural and had the same problem when he went to structural or thermal. His Ansys support guys said it was a memory issue. THis was back at V11-V12. (Deitmar, any update?)

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

Post generated using Mail2Forum (http://www.mail2forum.com)
_________________
Rick Fischer
Principal Engineer
Argonne National Laboratory
Back to top
View user's profile Send private message Send e-mail
seth.nickerson
User


Joined: 21 Oct 2008
Posts: 10

PostPosted: Tue Mar 27, 2012 1:06 pm  Reply with quote

FYI: I get similar issues when building large solid models (e.g. a
couple thousand area subtract Boolean operations in a row) in ANSYS via
APDL in batch mode. After speaking with tech help, the message
indicates a low level memory issue error code - but I don't believe it
is specific to one type of problem or solution...?

My problem is in V13.0 and is preprocessor only. I believe it stems
from the XOX geometry engine. No good work around was found yet...

Best of luck -

Seth Nickerson
Corning, Inc.

-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On
Behalf Of Fischer, Richard
Sent: Tuesday, March 27, 2012 3:20 PM
To: ANSYS User Discussion List
Subject: Re: [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

Very interesting. The error occurs during an FVMESH command on a model
in a resumed database. That database has 187 elements, static analysis,
no NLGEOM or SSTIF (all the stuff listed as prerequisites for the MSAVE
command per TFM) and the solution used the PCG solver. Thanks.

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215

-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On
Behalf Of Metrisin, Joe
Sent: Tuesday, March 27, 2012 2:01 PM
To: ANSYS User Discussion List
Subject: Re: [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

Quote:
From the knowledge database:

#2007215 - I am getting a SIG$SEGV error when running DANSYS on a
shared-memory machine. I tried turning MSAVE,OFF, and this seems to
help.

Family: Structural Mechanics
Product: Mechanical APDL (f. ANSYS)
Version: 12.1
Area: General
Subarea: N/A

Joseph T Metrisin
Structures Lead
Florida Turbine Technologies, Inc
1701 Military Tr. Suite 110
Jupiter, FL 33458 U.S.A.
+1 (561)427-6346 Office
(561)427-6191 Fax
JMetrisin@fttinc.com

Visit our new website: www.fttinc.com

FTT's public email encryption keys are stored on the FTT Verified
Directory at http://keys.fttinc.com

------------------------------------------------------------------------
-----------------------------
Confidentiality Note:
The information contained in this transmission and any attachments are
proprietary and may be privileged, intended only for the use of the
individual or entity named above. If the reader of this message is not
the intended recipient, you are hereby notified that any dissemination,
distribution or copying of this communication is strictly prohibited. If
you received this communication in error, please delete the message and
immediately notify the sender via the contact information listed above.
------------------------------------------------------------------------
-----------------------------

________________________________________
Answer:

INTERNAL

The problem has been fixed for v130.

Quote:
From Jeff B:

The error is unrelated to the MSAVE command or running models from older
versions. It is a memory bug in the DSP solver, which is used as a
kernel solver during the preconditioner creation for the PCG solver. It
depends on where the memory allocation sits as to when the model will
crash or not. Other options, like MSAVE, might simply move the memory
around enough to avoid the bug in some cases but is not a guaranteed
workaround.

In DSPmainPCG.F we call mf2LSave_sym to save the sym array. In
mf2LSave_sym.F we were sometimes writing data beyond the allocated
length for the sym array which can cause unexpected behavior. The code
never uses those extra values, but accessing them can cause crashes in
certain cases on certain platforms.

The mf2LSave_sym.F, mf2LRestore_sym.F and DSPmainPCG.F have been fixed
and checked into R13 today to resolve this bug.

No good workaround exists for this bug that would cover all cases. Users
can try different -m and -db allocation sizes to try to avoid the bug.
Typically, larger -m values and smaller -db values should work best to
avoid the bug. Avoid using the typical WB scheme of a -m value that is
50 MB bigger than the -db value. This is more likely to trigger the bug.



-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On
Behalf Of Fischer, Richard
Sent: Tuesday, March 27, 2012 2:38 PM
To: ANSYS User Discussion List
Subject: [FTT_SPAM] - [Xansys] MAPDL: Recurring unexpected error (
SIG$SEGV )

The full text of the message is: " An unexpected ( SIG$SEGV ) has
occurred... Ansys internal data has been corrupted. Ansys is unable to
recover and will terminate"

This error has started occurring when I run scripts. It usually goes
away after a reboot, and sometimes I can get it to run when I cut and
paste at the command line instead of using /input. Any idea what is
causing this? I've been looking for problems in the model, but cant
find anything. I am running structural to preturb a structure with
pressure, then resuming that database, doing an UPGEOM, meshing the
inside with HF119 elements with the FVMESH command, and then it happens.
Sometimes it goes for weeks without a problem, then it starts up again.

I googled SIG$SEGV and got a hit on Xansys from Sept 2009. Somebody was
running coupled CFD Structural and had the same problem when he went to
structural or thermal. His Ansys support guys said it was a memory
issue. THis was back at V11-V12. (Deitmar, any update?)

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

Post generated using Mail2Forum (http://www.mail2forum.com)
Back to top
View user's profile Send private message
dietmar.kolbeck
User


Joined: 10 Nov 2008
Posts: 4

PostPosted: Wed Mar 28, 2012 7:38 am  Reply with quote

Dear Rick,

at that time i made i workaround by using an older Ansys version for my problem. Due to that i wasted a lot of time, but i solved my problem. There was no chance for me to reduce the size of the model and i never got the problem solved on the specific Ansys Version, which caused the problem, running at that time. Also no real help from Ansys. The funny thing is, that my ASD forwarded my request at that time to Ansys (i still see it when i visit the customer portal) and the status is closed without getting any help.

Using V13 and today V14 i never got the same error again ... but others ;-).
_________________
Dietmar Kolbeck

KME Germany AG & Co. KG
Engineered Products Division
Melting and Casting
Tel.: ++49 541 321 3367
Fax: ++49 541 321 83367
mailto:Dietmar.Kolbeck@kme.com

Visit our Homepage: http://www.kme.com
Back to top
View user's profile Send private message Send e-mail
robert.martini
User


Joined: 27 Oct 2011
Posts: 2

PostPosted: Wed Mar 28, 2012 11:11 am  Reply with quote

We had the same problem with a V13 model here last year. The model ran on an older (slower) solver PC with Windows XP-64bit and on our contractor's Windows 7-64 bit machine. On our newer solver (also Windows 7 64 bit) we kept getting the SIG$SEGV error. I tried re-installing ANSYS, but the problem persisted. It turns out that the computer was one of the first Windows 7-64 bit machines in the company. We re-installed Windows 7 with a more current image, re-installed ANSYS, and we never saw the problem again. It seems like an extreme solution, but it worked.
_________________
Robert C. Martini
Mechanical System Designer
CAE Inc.
St. Laurent, Quebec, CANADA
Back to top
View user's profile Send private message
rick.fischer
User


Joined: 31 Aug 2011
Posts: 171

PostPosted: Wed Mar 28, 2012 11:28 am  Reply with quote

Thanks to all for the replies.

Robert, do you know what it was doing when it crashed? Was it meshing?

Seth, were you running XP or Windows 7? Also, you said it was preprocessing only. Meshing?

Dietmar, back in 2009, where you on XP?

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215

-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On Behalf Of robert.martini
Sent: Wednesday, March 28, 2012 1:11 PM
To: xansys@xansys.org
Subject: Re: [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

We had the same problem with a V13 model here last year. The model ran on an older (slower) solver PC with Windows XP-64bit and on our contractor's Windows 7-64 bit machine. On our newer solver (also Windows 7 64 bit) we kept getting the SIG$SEGV error. I tried re-installing ANSYS, but the problem persisted. It turns out that the computer was one of the first Windows 7-64 bit machines in the company. We re-installed Windows 7 with a more current image, re-installed ANSYS, and we never saw the problem again. It seems like an extreme solution, but it worked.

------------------------
Robert C. Martini
Mechanical System Designer
CAE Inc.
St. Laurent, Quebec, CANADA






+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

Post generated using Mail2Forum (http://www.mail2forum.com)
_________________
Rick Fischer
Principal Engineer
Argonne National Laboratory
Back to top
View user's profile Send private message Send e-mail
robert.martini
User


Joined: 27 Oct 2011
Posts: 2

PostPosted: Wed Mar 28, 2012 11:42 am  Reply with quote

Hi Rick,

The model was already meshed, CE&CP applied and boundary conditions set. The error occurred while solving a static analysis.
_________________
Robert C. Martini
Mechanical System Designer
CAE Inc.
St. Laurent, Quebec, CANADA
Back to top
View user's profile Send private message
chuck.ritter
User


Joined: 21 Oct 2008
Posts: 147

PostPosted: Wed Mar 28, 2012 11:47 am  Reply with quote

Until yesterday, this was of purely academic interest to me. Not trying
to thread hijack, but while solving I got an error indicating possible
full disk. Hard to believe a 12000DOF model could fill up the remaining
2/3rds terabyte available or that ANSYS would project that it could. In
solution and had already written about 100 results sets and was using
the sparse solver. My first thought was that I'd had a disk failure (USB
external and most likely the boxes' cheap internal drive to USB
electronics rather than a real drive failure). I was running Win 7
Professional 64 . I took the offending drive to a system running an
early copy of XP64 and had no problem solving all the steps. Which in
itself is a problem. When I get a chance, I'm going to try it on a Win
7 Ultimate 64 system - both are Dell T3500s. Both T3500s have solved
much larger models using the sparse solver.

Chuck Ritter
JAR Associates
North Kingstown,RI
401-294-4589
www.jar.com

On 3/28/2012 2:28 PM, Fischer, Richard wrote:
Quote:
Thanks to all for the replies.

Robert, do you know what it was doing when it crashed? Was it meshing?

Seth, were you running XP or Windows 7? Also, you said it was preprocessing only. Meshing?

Dietmar, back in 2009, where you on XP?

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215

-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On Behalf Of robert.martini
Sent: Wednesday, March 28, 2012 1:11 PM
To: xansys@xansys.org
Subject: Re: [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

We had the same problem with a V13 model here last year. The model ran on an older (slower) solver PC with Windows XP-64bit and on our contractor's Windows 7-64 bit machine. On our newer solver (also Windows 7 64 bit) we kept getting the SIG$SEGV error. I tried re-installing ANSYS, but the problem persisted. It turns out that the computer was one of the first Windows 7-64 bit machines in the company. We re-installed Windows 7 with a more current image, re-installed ANSYS, and we never saw the problem again. It seems like an extreme solution, but it worked.

------------------------
Robert C. Martini
Mechanical System Designer
CAE Inc.
St. Laurent, Quebec, CANADA






+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+





--
Chuck Ritter JAR Associates North Kingstown,RI 401-294-4589 www.jar.com

Post generated using Mail2Forum (http://www.mail2forum.com)
Back to top
View user's profile Send private message
rick.fischer
User


Joined: 31 Aug 2011
Posts: 171

PostPosted: Wed Mar 28, 2012 11:54 am  Reply with quote

Did you get the SIG$SEGV error or the one Input/output error on Unit=20, Please check to see if the disk is full?

If the latter, my IT guy fixed it by turning off "windows search indexing", whatever that is.

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215


-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On Behalf Of Chuck Ritter
Sent: Wednesday, March 28, 2012 1:47 PM
To: ANSYS User Discussion List
Subject: Re: [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

Until yesterday, this was of purely academic interest to me. Not trying to thread hijack, but while solving I got an error indicating possible full disk. Hard to believe a 12000DOF model could fill up the remaining 2/3rds terabyte available or that ANSYS would project that it could. In solution and had already written about 100 results sets and was using the sparse solver. My first thought was that I'd had a disk failure (USB external and most likely the boxes' cheap internal drive to USB electronics rather than a real drive failure). I was running Win 7 Professional 64 . I took the offending drive to a system running an early copy of XP64 and had no problem solving all the steps. Which in itself is a problem. When I get a chance, I'm going to try it on a Win
7 Ultimate 64 system - both are Dell T3500s. Both T3500s have solved much larger models using the sparse solver.

Chuck Ritter
JAR Associates
North Kingstown,RI
401-294-4589
www.jar.com

On 3/28/2012 2:28 PM, Fischer, Richard wrote:
Quote:
Thanks to all for the replies.

Robert, do you know what it was doing when it crashed? Was it meshing?

Seth, were you running XP or Windows 7? Also, you said it was preprocessing only. Meshing?

Dietmar, back in 2009, where you on XP?

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215

-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On
Behalf Of robert.martini
Sent: Wednesday, March 28, 2012 1:11 PM
To: xansys@xansys.org
Subject: Re: [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

We had the same problem with a V13 model here last year. The model ran on an older (slower) solver PC with Windows XP-64bit and on our contractor's Windows 7-64 bit machine. On our newer solver (also Windows 7 64 bit) we kept getting the SIG$SEGV error. I tried re-installing ANSYS, but the problem persisted. It turns out that the computer was one of the first Windows 7-64 bit machines in the company. We re-installed Windows 7 with a more current image, re-installed ANSYS, and we never saw the problem again. It seems like an extreme solution, but it worked.

------------------------
Robert C. Martini
Mechanical System Designer
CAE Inc.
St. Laurent, Quebec, CANADA






+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+





--
Chuck Ritter JAR Associates North Kingstown,RI 401-294-4589 www.jar.com

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

Post generated using Mail2Forum (http://www.mail2forum.com)
_________________
Rick Fischer
Principal Engineer
Argonne National Laboratory
Back to top
View user's profile Send private message Send e-mail
chuck.ritter
User


Joined: 21 Oct 2008
Posts: 147

PostPosted: Wed Mar 28, 2012 12:04 pm  Reply with quote

I got the latter message - Input/output error on unit=20. I'll try
turning off windows search indexing. I wish Windows wouldn't be so helpful.

Thanks

Chuck Ritter
JAR Associates
North Kingstown,RI
401-294-4589
www.jar.com



On 3/28/2012 2:54 PM, Fischer, Richard wrote:
Quote:
Did you get the SIG$SEGV error or the one Input/output error on Unit=20, Please check to see if the disk is full?

If the latter, my IT guy fixed it by turning off "windows search indexing", whatever that is.

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215


-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On Behalf Of Chuck Ritter
Sent: Wednesday, March 28, 2012 1:47 PM
To: ANSYS User Discussion List
Subject: Re: [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

Until yesterday, this was of purely academic interest to me. Not trying to thread hijack, but while solving I got an error indicating possible full disk. Hard to believe a 12000DOF model could fill up the remaining 2/3rds terabyte available or that ANSYS would project that it could. In solution and had already written about 100 results sets and was using the sparse solver. My first thought was that I'd had a disk failure (USB external and most likely the boxes' cheap internal drive to USB electronics rather than a real drive failure). I was running Win 7 Professional 64 . I took the offending drive to a system running an early copy of XP64 and had no problem solving all the steps. Which in itself is a problem. When I get a chance, I'm going to try it on a Win
7 Ultimate 64 system - both are Dell T3500s. Both T3500s have solved much larger models using the sparse solver.

Chuck Ritter
JAR Associates
North Kingstown,RI
401-294-4589
www.jar.com

On 3/28/2012 2:28 PM, Fischer, Richard wrote:

Quote:
Thanks to all for the replies.

Robert, do you know what it was doing when it crashed? Was it meshing?

Seth, were you running XP or Windows 7? Also, you said it was preprocessing only. Meshing?

Dietmar, back in 2009, where you on XP?

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215

-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On
Behalf Of robert.martini
Sent: Wednesday, March 28, 2012 1:11 PM
To: xansys@xansys.org
Subject: Re: [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

We had the same problem with a V13 model here last year. The model ran on an older (slower) solver PC with Windows XP-64bit and on our contractor's Windows 7-64 bit machine. On our newer solver (also Windows 7 64 bit) we kept getting the SIG$SEGV error. I tried re-installing ANSYS, but the problem persisted. It turns out that the computer was one of the first Windows 7-64 bit machines in the company. We re-installed Windows 7 with a more current image, re-installed ANSYS, and we never saw the problem again. It seems like an extreme solution, but it worked.

------------------------
Robert C. Martini
Mechanical System Designer
CAE Inc.
St. Laurent, Quebec, CANADA






+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+





--
Chuck Ritter JAR Associates North Kingstown,RI 401-294-4589 www.jar.com

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+





--
Chuck Ritter JAR Associates North Kingstown,RI 401-294-4589 www.jar.com

Post generated using Mail2Forum (http://www.mail2forum.com)
Back to top
View user's profile Send private message
dietmar.kolbeck
User


Joined: 10 Nov 2008
Posts: 4

PostPosted: Thu Mar 29, 2012 3:18 am  Reply with quote

Rick,

i was and i'm still on XP-64bit. Unfortunately our company is a little bit conservative in changing to a newer version.
_________________
Dietmar Kolbeck

KME Germany AG & Co. KG
Engineered Products Division
Melting and Casting
Tel.: ++49 541 321 3367
Fax: ++49 541 321 83367
mailto:Dietmar.Kolbeck@kme.com

Visit our Homepage: http://www.kme.com
Back to top
View user's profile Send private message Send e-mail
seth.nickerson
User


Joined: 21 Oct 2008
Posts: 10

PostPosted: Thu Mar 29, 2012 8:19 am  Reply with quote

Rick,

We had the error occur in pre-processing (not even meshing yet, just
doing ASBA (area subtraction Boolean operations). It was repeatable on
several machines (including at ANSYS tech help) including Win XP 64 and
Win 7 64bit systems. The models would crash in interactive mode (but
not necessarily throw that error code). However in batch mode it would
crash as well and throw that error code...it's been a confusing issue...

Best of luck,

Seth Nickerson
Corning, Inc.

-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On
Behalf Of Fischer, Richard
Sent: Wednesday, March 28, 2012 2:28 PM
To: ANSYS User Discussion List
Subject: Re: [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

Thanks to all for the replies.

Robert, do you know what it was doing when it crashed? Was it meshing?

Seth, were you running XP or Windows 7? Also, you said it was
preprocessing only. Meshing?

Dietmar, back in 2009, where you on XP?

Rick Fischer
Principal Engineer
Argonne National Laboratory
630-252-2215

-----Original Message-----
From: xansys-bounces@xansys.org [mailto:xansys-bounces@xansys.org] On
Behalf Of robert.martini
Sent: Wednesday, March 28, 2012 1:11 PM
To: xansys@xansys.org
Subject: Re: [Xansys] MAPDL: Recurring unexpected error ( SIG$SEGV )

We had the same problem with a V13 model here last year. The model ran
on an older (slower) solver PC with Windows XP-64bit and on our
contractor's Windows 7-64 bit machine. On our newer solver (also Windows
7 64 bit) we kept getting the SIG$SEGV error. I tried re-installing
ANSYS, but the problem persisted. It turns out that the computer was one
of the first Windows 7-64 bit machines in the company. We re-installed
Windows 7 with a more current image, re-installed ANSYS, and we never
saw the problem again. It seems like an extreme solution, but it worked.

------------------------
Robert C. Martini
Mechanical System Designer
CAE Inc.
St. Laurent, Quebec, CANADA






+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

+-------------------------------------------------------------+
| XANSYS web - www.xansys.org/forum |
| The Online Community for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
| Send administrative requests to xansys-mod@tynecomp.co.uk |
+-------------------------------------------------------------+

Post generated using Mail2Forum (http://www.mail2forum.com)
Back to top
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    XANSYS Forum Index -> XANSYS
All times are GMT - 7 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You cannot download files in this forum


Powered by phpBB © 2001, 2005 phpBB Group

sleek template created by Andrew Charron