Hello friends,

I am running a 3D case in parallel using Fluent on 30 nodes. I ran a case (Eps-0.41.cas) in the morning and it ran well. Again I ran the same case in the evening and I got this error. I am seeing this error quite often when I run in parallel. We are using BEOWULF cluster. Error is as follows

Emergency: received SIGHUP signal.

The application is busy. Please wait a moment.
: Emergency: received SIGHUP signal.

The application is busy. Please wait a moment.
: Error: = (equal): invalid argument [1]: wrong type [not a number] Error Object: #t

> invalid command [yes]

> Halting due to end of file on input. parent: Failed to read len_buf

Please anyone who have some idea, help me... I badly need your help...

Thanks Ravi

[dokeun]

I got a problem maybe the same as Ravi's.

Emergency: received SIGHUP signal.
The application is busy.
Please wait a moment.
================================================== ============================
Stack backtrace generated for process id 23031 on signal 1 :
fluent[0x4b68aa]
/lib64/libc.so.6[0x353e2302d0]
/lib64/libc.so.6(__select+0x62)[0x353e2cced2]
fluent(listen_input+0x1c2)[0x4b97b2]
fluent[0x4ba025]
fluent(lreadf+0x2d)[0x5705dd]
fluent(eval+0x7cc)[0x56b8dc]
fluent(eval+0x125b)[0x56c36b]
fluent(eval+0x906)[0x56ba16]
fluent[0x56cd58]
fluent(eval_errprotect+0x32)[0x56cde2]
fluent(eval+0x2ef)[0x56b3ff]
fluent(eval+0x125b)[0x56c36b]
fluent(eval+0x906)[0x56ba16]
fluent(eval+0x125b)[0x56c36b]
fluent(eval+0x906)[0x56ba16]
Please include this information with any bug report you file on this issue!
================================================== ============================
No error handler available

Is anybody who has the answer?

[step]

Hello Dokeun,

I made the experience that it sometimes helps to read in the case on a single processor. Afterwards partition it manually for, let's say 30 knots, and then save the case and read in parallel. Sometime it helps, because it appears to me that the autopartitioning feature is not as "stable".

Best regards

[dokeun]

Thank you for your advice, step

I think I will try it =)

Originally Posted by step

Hello Dokeun,

I made the experience that it sometimes helps to read in the case on a single processor. Afterwards partition it manually for, let's say 30 knots, and then save the case and read in parallel. Sometime it helps, because it appears to me that the autopartitioning feature is not as "stable".

Best regards

Dear step

It works. After 1 iteration with single processor, I ran the case in parallel.

Thank you so much again.

[step]

I'm happy that I was able to help. You are always welcome!

Best regards