I peformed the solvation for a protein using pre-equilibrated water box named tip125.crd. The CHARMM I used is c27b4 version with xlarge size installation. Here is the part of my script
....
incr seg by 1
read sequ tip3 @nwat
generate w@seg setup noangle nodihe
....

the output is
CHARMM > incr seg by 1
Parameter : SEG <- "259"
CHARMM> read sequ tip3 @nwat
Parameter: NWAT -> "125"
CHARMM> generate w@seg setup noangle nodihe
Parameter: SEG -> "259"
NO PATCHING WILL BE DONE ON THE FIRST RESIDUE
NO PATCHING WILL BE DONE ON THE LAST RESIDUE
GENPSF> Segment 259 has been generated. Its identifier is W259.

***** LEVEL -4 WARNING FROM <ALLDT> *****
***** DATA STRUCTURE REALLOCATED IN ALLDT
******************************************
BOMLEV ( -1) IS REACHED - TERMINATING. WRNLEV IS 5

ABNORMAL TERMINATION
MAXIMUM STACK SPACE USED IS 419004
STACK CURRENTLY IN USE IS 0
MOST SEVERE WARNING WAS AT LEVEL -4
HEAP PRINTOUT- HEAP SIZE 10240000
SPACE CURRENTLY IN USE IS 480
MAXIMUM SPACE USED IS 120196
FREE LIST


any hint for solving this problem?

I met with this error before. My understanding is, you are trying to build too much atoms. Check the maximum number of atoms that you can build, please. Apologise in advance if it is not helpful.

Good luck!

It is a memory overwrite error. It is usually a coding error.
Even when it is the user's fault (too many something), this is not the proper way to notify the user...

Please submit it as a bug report.

Usually this message indicates that you have exceeded the maximum for some size parameter for the charmm executable you are using. Sometimes there is a more useful message that shows the maxima and what you are attempting to use so you can see which one is causing the trouble.

The easiest solution, when it works, is to compile a larger version of charmm.

Mike

Acutally, I test my script with a samller protein. It works well. Thus, as you expected, it may be caused by exceeding the maximum usage of some data strcture.
I will complie the CHARMM one more time with the xlarge size. If the problem still exists, I will submit a bug report.
thanks.