[molpro-user] ERROR IN WRITE DUMP error
Jacek Antoni Klos
jklos at umd.edu
Wed Sep 19 19:44:26 BST 2012
Dear Molpro users and developers,
I have encountered the following error in new Molpro 2012.1.
When running in a loop and having assigned record to store MRCI densities after a while the job exits with an error:
?ERROR IN WRITE_DUMP: TOO MANY ENTRIES IN RECORD 4133.3 WHEN WRITING TRDEN/TRANSITION SET= 0
I went and increased maxset parameter from 100 to 600 in src/common/dumpinfor and src/common/dumpinfow and recompiled
and the job went for 20 loop steps longer but died again. The job runs OK with no problems in 2010.1 version without increasing
dumpinfor/w maxset parameter. I have noticed in the output that in 2012.1 molpro the above 4133.3 record is repeated:
3 107 14.22 1000 5011 5021 5111 5121 4011 6021 4021 4181 4161
BASIS MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
5311 5321 5411 5421 4211 4221 6221 6321 4181(1) 4161(1)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
4181(2) 4161(2) 4381 4361 4051 4133 5021(1) 4051(1) 4133(1) 4181(3)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
4161(3) 4133(2) 4181(4) 4381(1) 4361(1) 4133(3) 4181(5) 4381(2) 4361(2) 4133(4)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
4181(6) 4381(3) 4361(3) 4133(5) 4181(7) 4381(4) 4361(4) 4133(6) 4181(8) 4381(5)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
4361(5) 4133(7) 4181(9) 4381(6) 4361(6) 4133(8) 4181(*) 4381(7) 4361(7) 4133(9)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
4181(*) 4381(8) 4361(8) 4133(*) 4181(*) 4381(9) 4361(9) 4133(*) 4181(*) 4381(*)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
4361(*) 4133(*) 4181(*) 4381(*) 4361(*) 4133(*) 4181(*) 4381(*) 4361(*) 4133(*)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
4181(*) 4381(*) 4361(*) 4133(*) 4181(*) 4381(*) 4361(*) 4133(*) 4181(*) 4381(*)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
4361(*) 4133(*) 4181(*) 4381(*) 4361(*) 4133(*) 4181(*) 4381(*) 4361(*) 4133(*)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
4181(*) 4381(*) 4361(*) 4133(*) 4181(*) 4381(*) 4361(*)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI
in Molpro 2010.1 it is not repeated like that:
3 27 6.41 700 1000 5011 5021 5111 5121 4011 6021 4021 4181
GEOM BASIS MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
4161 5311 5321 5411 5421 4211 4221 6221 6321 4181(1)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI MRCI
4161(1) 4381 4361 4051 4133 5021(1) 4051(1)
MRCI MRCI MRCI MRCI MRCI MRCI MRCI
Does it mean that in Molpro 2012.1 this record is getting build up and stores data for every loop step while in 2010.1 molpro
it overrides the values and doesn't go over maxset parameter? How to avoid this behavior in 2012.1, I can't go and recompile
molpro after increasing maxset each time.
In the input the record is assigned to ddrdm_trans=4133.3 variable.
Best regards
Jacek
Jacek Klos, Research Assistant Professor
Department of Chemistry
University of Maryland
College Park, MD 20742
More information about the Molpro-user
mailing list