errors in run with mutiple domains

bolus13 · November 22, 2016, 03:13

Hi All,

I am trying to solve the flow through a rotating domain with several subdomains inside. The subdomains are included to model flaps that move while the larger domain rotates. See the images for an idea. I was advised by the guys from ansys to use moving meshes, which works well for if I model a small number of subdomains

I started this with 6 flaps and this worked well. In that case I had a total of 13 moving meshed.

Now when I increase to more flaps I start getting the error as given below. I have checked and rechecked my setup and it seems good. The error says a file is locked, but I do not get this error when I use a small number of domains.

Could this be a memory issue. I am running this from workbench, could that be an issue

Details of error:-
----------------
Error detected by routine MAKLNK
COLDNM = CLOOP0 CNEWNM = LAST
CRESLT = NONE

Current Directory : /FLOW/MESH/TSTEP2

+--------------------------------------------------------------------+
| Writing crash recovery file |
+--------------------------------------------------------------------+

Details of error:-
----------------
Error detected by routine PEEKD
CDANAM = DRTIME_DTEND
CRESLT = NONE

Current Directory : /FLOW/SOLUTION/TSTEP3/CLOOP0/ZN1

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| The ANSYS CFX solver exited with return code 1. No results file |
| has been created. |
+--------------------------------------------------------------------+

End of solution stage.

+--------------------------------------------------------------------+
| The following transient and backup files written by the ANSYS CFX |
| solver have been saved in the directory D:/ansys |
| cfx/DWE/flapModelNineFlapVenturi_pending/dp0_CFX_4_Solution_4/CFX- |
| _002: |
| |
| 2.trn, 1.trn |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Error reported by IO module: iif_set_lock: file D:\ansys |
| cfx\DWE\flapModelNineFlapVenturi_pending\dp0_CFX_4 _Solution_4\CFX- |
| _002.dir\crash is currently locked for writing, and cannot be |
| locked for read access. Either another program is using this |
| file, or the lock file D:\ansys |
| cfx\DWE\flapModelNineFlapVenturi_pending\dp0_CFX_4 _Solution_4\CFX- |
| _002.dir\crash.lck is bad. |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Error reported by IO module: iocnt: open the primary file failed |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Error from IO module while opening D:\ansys |
| cfx\DWE\flapModelNineFlapVenturi_pending\dp0_CFX_4 _Solution_4\CFX- |
| _002.dir\crash: |
| |
| "iocnt: open the primary file failed" |
| |
| IO_LOCKED(21): File is locked. Remove the lock (lck) file if this |
| shouldn't be the case. |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Error reported by IO module: iif_set_lock: file D:\ansys |
| cfx\DWE\flapModelNineFlapVenturi_pending\dp0_CFX_4 _Solution_4\CFX- |
| _002.dir\crash is currently locked for writing, and cannot be |
| locked for read access. Either another program is using this |
| file, or the lock file D:\ansys |
| cfx\DWE\flapModelNineFlapVenturi_pending\dp0_CFX_4 _Solution_4\CFX- |
| _002.dir\crash.lck is bad. |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Error reported by IO module: iocnt: open the primary file failed |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Error from IO module while opening D:\ansys |
| cfx\DWE\flapModelNineFlapVenturi_pending\dp0_CFX_4 _Solution_4\CFX- |
| _002.dir\crash: |
| |
| "iocnt: open the primary file failed" |
| |
| IO_LOCKED(21): File is locked. Remove the lock (lck) file if this |
| shouldn't be the case. |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| Warning! |
| |
| The ANSYS CFX Solver has written a crash recovery file. This file |
| has been saved as D:/ansys |
| cfx/DWE/flapModelNineFlapVenturi_pending/dp0_CFX_4_Solution_4/CFX- |
| _002.res.err and may be an aid to diagnosing the problem or |
| restarting the run. More details should be available in the |
| solver output section of the output file. Note that a lock file |
| was left for the crash file when the solver exited, and so it is |
| probably incomplete. |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| The following user files have been saved in the directory D:/ansys |
| cfx/DWE/flapModelNineFlapVenturi_pending/dp0_CFX_4_Solution_4/CFX- |
| _002: |
| |
| mon |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| For CFX runs launched from Workbench, the final locations of |
| directories and files generated may differ from those shown. |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Error reported by IO module: create_indextable: stored index is |
| wrong: missing start tag |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Error reported by IO module: iif_open: unable to read the stored |
| index, attempting to rebuild it |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| Error reported by IO module: recreate_indextable: warning: file |
| was not closed correctly, data may be inconsistent |
+--------------------------------------------------------------------+

ghorrocks · November 22, 2016, 04:19

Are all those errors from the same run? If so then only the first error is relevant. And I can't make out what that means from the information posted here.

Also: Are you doing this using a rotating frame of reference with moving mesh (as part of the one domain)? Or is it rotating frame of reference with the flaps as separate domains?

bolus13 · November 22, 2016, 04:32

yes, all the errors are from the same run.

The domains are all separate. Within each domain I add a subdomain, for which I specify the mesh motion.

The strange thing to me is that it works for a smaller number of domains. And I am pretty convinced I checked all settings carefully.

ghorrocks · November 22, 2016, 04:47

Are you doing it with rotating frames of reference or moving mesh?

bolus13 · November 22, 2016, 04:52

moving mesh

ghorrocks · November 22, 2016, 05:15

Are all the domains moving mesh?

Are you using displacement diffusion or are you specifying the mesh motion?

Is the motion coupled to the flow at all? Or is the motion of flap known in advance?

bolus13 · November 22, 2016, 05:23

There is one static domain. Inside the static domain is the larger circular moving mesh. Inside this are the smaller moving meshes for the flaps.

I am specifying the mesh motion by means of the 'specified location' option. This is applied to each single subdomain inside each domain.

The motion of the flaps is known in advance. I am prescribing it.

ghorrocks · November 22, 2016, 06:03

OK, I understand. For the mesh motion are you using displacement diffusion or specifying the motion of the internal nodes?

Can you post an image of the 6 flap model and the 13 flap model? Also post an image of the mesh in the 13 flap model on time step 2, which is the time step before it crashed.

bolus13 · November 22, 2016, 08:25

I specify the motion of the internal nodes. Actually it is done in such a way that the mesh of the larger circular domain does not deform during rotation. Only the mesh in the smaller domains deform.

Attached is an image of the mesh on time it crashed, including a close up of one of the flaps. This is a 9 flap model build out of 20 domains. 1 static domain, 1 rotating domain and 18 domains for the flaps.

ghorrocks · November 22, 2016, 16:31

Thanks. That looks like a good way of doing it, but note the mesh quality is not very high. I would expect there to be some convergence problems from this, but if you have a simple fluid model (ie incompressible, constant properties, single phase) it should be OK. If you have a complex fluid model you might need to consider improvements to mesh quality.

Another option I can see for this simulation which might improve mesh quality is dynamic remeshing.

Back to your original question - The error message is unusual and it is not something I have seen before. It is going to take some debugging to find the cause.
Was this error a one-off, or does it do it every time? If you delete a few of your flaps does it run OK? Are you sure every one of your flaps are correctly set up?

bolus13 · November 23, 2016, 04:01

Maybe good to note here, that I am not even solving the fluid here. Ansys advised to use the following expert parameters

FLOW:
EXPERT PARAMETERS:
solve energy = f
solve fluids = f
solve masfrc = f
solve meshdisp = t
solve turbulence = f
solve volfrc = t
solve wallscale = f
END
END

The error itself is a bit temperamental. For different number of flaps I get the error at different time steps.

- For a 5 flap model it works fine.
- Then I moved to a 6 flap model, also worked fine.
- Then moved to 9 flaps, this produced the error, but then when I changed the wall boundary in my static domain to free slip, it started working. To me this is very strange. (note here that my model is 3 elements thick)
- Then moved to 15 flaps, which is the mesh I showed in the previous post. This stopped after 6 iterations. Then I changed the mesh and it stopped after 9 iterations. Also when I change settings in the solver, like number of cores and memory usage, it also stops at a different time step.

I added a graph of my current mesh quality, which might help. In the meanwhile I keep checking my setup; cannot find any mistakes.

ghorrocks · November 23, 2016, 17:32

Are you using double precision numerics? You will probably need that.

Also I note that you appear to be solving the volume fraction equations but no other fluid equations. Is that intentional? That would lead to some stability problems I would expect.

bolus13 · November 24, 2016, 03:07

I changed to double precision and turned of the volume fraction equation. This ran for a longer time, but eventually failed at time step 18 with the same error.

Then I tried improving the mesh, and this time it failed after 9 time steps.

Then, just to try a few things, I changed to parallel run with two cores and it failed after 6 time steps. In this last case it gave a different error:

Details of error:-
----------------
Error detected by routine MAKLNK
COLDNM = CLOOP0 CNEWNM = LAST
CRESLT = NONE

Current Directory : /FLOW/MESH/TSTEP6

+--------------------------------------------------------------------+
| ERROR #001100279 has occurred in subroutine ErrAction. |
| Message: |
| Stopped in routine MEMERR |
| |
| |
| |
| |
| |
+--------------------------------------------------------------------+

+--------------------------------------------------------------------+
| An error has occurred in cfx5solve: |
| |
| The ANSYS CFX solver exited with return code 1. No results file |
| has been created. |
+--------------------------------------------------------------------+

Lance · November 24, 2016, 03:57

Quote:

Originally Posted by bolus13

Have you tried to increase memory size? I find that CFX often need more memory than it initially allocates when running mesh deformation

bolus13 · November 24, 2016, 05:19

I did, this also did not work.

I started trouble shooting and simplifying until things work.

bolus13 · December 16, 2016, 10:07

So, after contacting the guys at ansys, they instructed me to run on version 17.2 instead of the 16.0 I was using (for legacy reasons). This worked well. I did not have to increase memory size, though I did run with double precision.

They said that "the error had something to do with the mesh data access within the solver. The error is not known in this form"

So to me it seemed there was some bug in version 16.0, but it is solved now.

Thanks for the help though

sam_cfdd · December 26, 2016, 23:40

Hi,

I am running tutorial number 29 (spray dryer) and I enter all step based on tutorial but when I click run after 5 iteration I faced with error. I tried to increase allocate memory but still error is take place. Do u have any idea to what is error and how I can solve it. see please snapshot of error.

Thanks

December 16, 2016, 10:07		#16
bolus13 Member Ruud Caljouw Join Date: Dec 2012 Posts: 45 Rep Power: 13	So, after contacting the guys at ansys, they instructed me to run on version 17.2 instead of the 16.0 I was using (for legacy reasons). This worked well. I did not have to increase memory size, though I did run with double precision. They said that "the error had something to do with the mesh data access within the solver. The error is not known in this form" So to me it seemed there was some bug in version 16.0, but it is solved now. Thanks for the help though jiaji lu likes this.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
AMI speed performance	danny123	OpenFOAM	21	October 24, 2020 04:13
time step continuity problem in VAWT simulation	lpz_michele	OpenFOAM Running, Solving & CFD	5	February 22, 2018 19:50
High Courant Number @ icoFoam	Artex85	OpenFOAM Running, Solving & CFD	11	February 16, 2017 13:40
Upgraded from Karmic Koala 9.10 to Lucid Lynx10.04.3	bookie56	OpenFOAM Installation	8	August 13, 2011 04:03
Convergence moving mesh	lr103476	OpenFOAM Running, Solving & CFD	30	November 19, 2007 14:09

November 22, 2016, 04:19		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,695 Rep Power: 143	Are all those errors from the same run? If so then only the first error is relevant. And I can't make out what that means from the information posted here. Also: Are you doing this using a rotating frame of reference with moving mesh (as part of the one domain)? Or is it rotating frame of reference with the flaps as separate domains?

November 22, 2016, 04:32		#3
bolus13 Member Ruud Caljouw Join Date: Dec 2012 Posts: 45 Rep Power: 13	yes, all the errors are from the same run. The domains are all separate. Within each domain I add a subdomain, for which I specify the mesh motion. The strange thing to me is that it works for a smaller number of domains. And I am pretty convinced I checked all settings carefully.

November 22, 2016, 04:47		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,695 Rep Power: 143	Are you doing it with rotating frames of reference or moving mesh?

November 22, 2016, 04:52		#5
bolus13 Member Ruud Caljouw Join Date: Dec 2012 Posts: 45 Rep Power: 13	moving mesh

November 22, 2016, 05:15		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,695 Rep Power: 143	Are all the domains moving mesh? Are you using displacement diffusion or are you specifying the mesh motion? Is the motion coupled to the flow at all? Or is the motion of flap known in advance?

November 22, 2016, 05:23		#7
bolus13 Member Ruud Caljouw Join Date: Dec 2012 Posts: 45 Rep Power: 13	There is one static domain. Inside the static domain is the larger circular moving mesh. Inside this are the smaller moving meshes for the flaps. I am specifying the mesh motion by means of the 'specified location' option. This is applied to each single subdomain inside each domain. The motion of the flaps is known in advance. I am prescribing it.

November 22, 2016, 06:03		#8
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,695 Rep Power: 143	OK, I understand. For the mesh motion are you using displacement diffusion or specifying the motion of the internal nodes? Can you post an image of the 6 flap model and the 13 flap model? Also post an image of the mesh in the 13 flap model on time step 2, which is the time step before it crashed.

November 22, 2016, 16:31		#10
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,695 Rep Power: 143	Thanks. That looks like a good way of doing it, but note the mesh quality is not very high. I would expect there to be some convergence problems from this, but if you have a simple fluid model (ie incompressible, constant properties, single phase) it should be OK. If you have a complex fluid model you might need to consider improvements to mesh quality. Another option I can see for this simulation which might improve mesh quality is dynamic remeshing. Back to your original question - The error message is unusual and it is not something I have seen before. It is going to take some debugging to find the cause. Was this error a one-off, or does it do it every time? If you delete a few of your flaps does it run OK? Are you sure every one of your flaps are correctly set up?

November 23, 2016, 17:32		#12
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,695 Rep Power: 143	Are you using double precision numerics? You will probably need that. Also I note that you appear to be solving the volume fraction equations but no other fluid equations. Is that intentional? That would lead to some stability problems I would expect.

November 24, 2016, 03:07		#13
bolus13 Member Ruud Caljouw Join Date: Dec 2012 Posts: 45 Rep Power: 13	I changed to double precision and turned of the volume fraction equation. This ran for a longer time, but eventually failed at time step 18 with the same error. Then I tried improving the mesh, and this time it failed after 9 time steps. Then, just to try a few things, I changed to parallel run with two cores and it failed after 6 time steps. In this last case it gave a different error: Details of error:- ---------------- Error detected by routine MAKLNK COLDNM = CLOOP0 CNEWNM = LAST CRESLT = NONE Current Directory : /FLOW/MESH/TSTEP6 +--------------------------------------------------------------------+ \| ERROR #001100279 has occurred in subroutine ErrAction. \| \| Message: \| \| Stopped in routine MEMERR \| \| \| \| \| \| \| \| \| \| \| +--------------------------------------------------------------------+ +--------------------------------------------------------------------+ \| An error has occurred in cfx5solve: \| \| \| \| The ANSYS CFX solver exited with return code 1. No results file \| \| has been created. \| +--------------------------------------------------------------------+

November 24, 2016, 05:19		#15
bolus13 Member Ruud Caljouw Join Date: Dec 2012 Posts: 45 Rep Power: 13	I did, this also did not work. I started trouble shooting and simplifying until things work.