WRF SEGMENTATION ERROR

[LagnajeetRoy]

while i was running WRFv4.5.2, after running for 10-12 hours it is showing this error: "
===================================================================================
= BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
= PID 72052 RUNNING AT lagnajeet-MS-7E06
= EXIT CODE: 9
= CLEANING UP REMAINING PROCESSES
= YOU CAN IGNORE THE BELOW CLEANUP MESSAGES
===================================================================================
YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Killed (signal 9)
This typically refers to a problem with your application.
Please see the FAQ page for debugging suggestions

I am attaching the rsl.out and rsl.error and also the namelist.input. please take a look and let me know if there are any suggestions

 

[kwerner]

As a test, can you try running a single domain (just change max_dom=1) to see if that works? If so, then try 2 domains, and increase until you figure out which domain is causing the problem. Let me know the result.

Can you also check to make sure you're not running out of disk space or wall clock time (if you're submitting a job to a queuing system)? Thanks!

 

[LagnajeetRoy]

Hey @kwerner i have tried with single domain and then multiple domain set up there was issue when i am using BEM as urban physics scheme. Both SLUCM and BEP model run was fine with no error. There is enough disk space

 

[LagnajeetRoy]

I have ran the model for first two domain its running fine. There is some problem in the third domain i have changed the namelist.wps and input file and run again still showing the segmentation error for 3rd domain. I am attaching the rsl.error, rsl.out and the new namelist file. This time i used WRF 4.3. Please check. I saw this segmentation error quite frequently and sometime the model run for a long time and then show segmentation error. It hard to determine the specific reason for the segmentation error. I will be happy to know more about these kind of problems and their solutions.

 

[William.Hatheway]

I have ran the model for first two domain its running fine. There is some problem in the third domain i have changed the namelist.wps and input file and run again still showing the segmentation error for 3rd domain. I am attaching the rsl.error, rsl.out and the new namelist file. This time i used WRF 4.3. Please check. I saw this segmentation error quite frequently and sometime the model run for a long time and then show segmentation error. It hard to determine the specific reason for the segmentation error. I will be happy to know more about these kind of problems and their solutions.

@LagnajeetRoy



WPS looks good, the thing that I see in namelist.input is that dx/dy should have the all the grid spaces defined.

so

dx = 30000, 10000, 3333.33, 1111.11
dy = 30000, 10000, 3333.33, 1111.11

cu_physics = 5, 5, 0, 0, (you can also try to turn off the cu_physics for the ones below 4km as wrf can explicitly resolve clouds at the resolution, try the first suggestion before doing this)

 

[LagnajeetRoy]

Hello @William.Hatheway i have changed the namelist.input as i define the grid spaces dx/dy and also changed the cu_physics still i am getting the segmentation error. The problem arises with the third domain as first and second domain is running fine. Here are the attachments.

 

[William.Hatheway]

I'm in the hospital so I will try to look at it later.

 

[LagnajeetRoy]

@William.Hatheway please take care. In the mean time if i got any solution i will let you know.

 

[Abdullah Sami]

Respected All,

I am running WRF 4.6 for the first time. I'm also experiencing the same Segmentation Error:
===================================================================================
= BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
= PID 41157 RUNNING AT ROG-Strix-G15
= EXIT CODE: 9
= CLEANING UP REMAINING PROCESSES
= YOU CAN IGNORE THE BELOW CLEANUP MESSAGES ===================================================================================YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Segmentation fault (signal 11)This typically refers to a problem with your application.Please see the FAQ page for debugging suggestions

I have attached the rsl error and out files. The namelist.input is also attached. Kindly assist me on this. Thankyou.

 

[kwerner]

Respected All,

I am running WRF 4.6 for the first time. I'm also experiencing the same Segmentation Error:
===================================================================================
= BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES
= PID 41157 RUNNING AT ROG-Strix-G15
= EXIT CODE: 9
= CLEANING UP REMAINING PROCESSES
= YOU CAN IGNORE THE BELOW CLEANUP MESSAGES ===================================================================================YOUR APPLICATION TERMINATED WITH THE EXIT STRING: Segmentation fault (signal 11)This typically refers to a problem with your application.Please see the FAQ page for debugging suggestions

I have attached the rsl error and out files. The namelist.input is also attached. Kindly assist me on this. Thankyou.

Since a segmentation fault is a very generic error message, and can be the result of many different things, can you post your inquiry as a new thread - this will keep topics separated and easier to search/read. Thanks!

 

[Abdullah Sami]

Since a segmentation fault is a very generic error message, and can be the result of many different things, can you post your inquiry as a new thread - this will keep topics separated and easier to search/read. Thanks!

Okay, Thanks!

 

[kwerner]

i have changed the namelist.input as i define the grid spaces dx/dy and also changed the cu_physics still i am getting the segmentation error. The problem arises with the third domain as first and second domain is running fine. Here are the attachments.

Hi,
Apologies for the delay. It looks like the model stops immediately now, and before it was stopping after some time. Sometimes when the model stops right away, it indicates an issue with the input data. I assume you're still using the same data as before? If not, check the data to make sure there aren't any NaNs or anything that looks weird. Otherwise, you can try to do some debugging. To do this, you'll need to clean the code (./clean -a), then reconfigure using

./configure -D

Then recompile. This won't remove any of the files you have placed in the directories - everything should stay there. When you run WRF again, check the end of the rsl.error.0000 file to see the traceback. This may tell you exactly where in the code the model is stopping. Feel free to send those new rsl* files, as well (go ahead and send them all - packaged into a zipped file again).

P.S. - For the record, it's not actually necessary to set dx/dy to a resolution for each domain. This was necessary prior to V4.2, but it's not anymore.