Hello everyone,
I’m using qchem 5.4.1, i’m using NEMO with qchem to calculate ISC rates, however when submitting a batch of jobs i get lots of strange errors:
For example:
*** Error in `/cm/shared/apps/qchem/qchem54/exe/qcprog.exe_s’: double free or corruption (!prev): 0x00000000371b7e70 ***
*** Error in `/cm/shared/apps/qchem/qchem54/exe/qcprog.exe_s’: free(): invalid next size (normal): 0x00000000382148e0 ***
*** Error in `/cm/shared/apps/qchem/qchem54/exe/qcprog.exe_s’: malloc(): memory corruption: 0x0000000037a0d620 ***
Is it possible this is due to disk space? I am only using 50 GB out of 2 TB so that would be weird.
It always seems to crash before the printing of the orbital energies, after the STATE-TO-STATE TRANSITION MOMENTS have been printed.
That looks like a Q-Chem problem not a batch job problem. Can you please post a complete input file that exhibits the error? Use the formatted text button </> to avoid losing special characters.
$comment
EMISPCT
$end
$rem
cis_n_roots 5
cis_singlets true
cis_triplets true
calc_soc true
STS_MOM true
CIS_RELAXED_DENSITY TRUE
solvent_method PCM
EXCHANGE LRC-wPBE
OMEGA 184
RPA false
BASIS 6-31G(d,p)
THRESH 14
$end
$pcm
theory IEFPCM
ChargeSeparation Marcus
StateSpecific Perturb
$end
$solvent
Dielectric 37.5
OpticalDielectric 1.8066048100000003
$end
$molecule
0 1
C 1.27377551664317 1.69461248014133 -0.53196590925741
C 1.14891923958454 0.64962097139574 0.43817976172310
C 2.58938254687196 2.21124313261416 -0.76909007238453
C 3.67232924804091 1.50203455946636 -0.29523842881273
C 3.54496092842987 0.53449172846677 0.65667596709262
C 2.26871997830362 0.01263314839975 1.01409351499730
C 5.10867494728464 1.85394042872957 -0.41500238349855
C 5.41509695153314 3.29035560421130 0.01624931188216
C 5.71564893477905 1.77199401430941 -1.81530515226767
C 5.83109016585376 0.80661633093045 0.48452466768767
N 4.83406330329287 0.05048580758413 1.14582932942266
C 4.95640754099721 -0.85770383926689 2.28918252635682
S 8.15883019347720 -1.30867698099547 -0.76985425361321
C 7.20139222484259 0.40447338413368 0.66377729298978
C 8.32234512083268 1.36221639831556 0.50223068517498
C 9.63668482430370 0.80287492737406 0.14573579387472
C 9.67866337541142 -0.50176194936325 -0.40801081286863
C 10.97807620052607 -1.10256929942235 -0.63202085166272
C 10.85415806896645 1.64346361030839 0.23300800682473
C 12.00526756269975 0.97152484552600 -0.29447567812508
C 12.08220156581073 -0.30677744321962 -0.69244697242140
N 13.10726739651277 1.98259164362935 -0.32286851660776
O 12.87095005036340 3.21420931606783 -0.08756383557016
O 14.19796566615205 1.56407298805960 -0.79308001494797
H 0.36415274543137 2.17896221517842 -0.92789233161983
H 0.21725627591957 0.31288154738483 0.73222978548603
H 2.43968735922604 3.09841791458018 -1.36016764043007
H 2.07907074252682 -0.82440962664480 1.76194342920113
H 5.00163141773958 4.00841539281049 -0.79086298757088
H 6.43508688323929 3.24363989082623 0.17978907744596
H 4.99691631522621 3.64314821505966 1.09692820261748
H 5.11845658044236 2.61923582351130 -2.15663402604157
H 5.65656394969590 0.72415516745454 -2.26615333968845
H 6.80085367064321 1.94297713504508 -1.69430259154829
H 5.82744081442201 -0.44983599487632 2.96924420631895
H 5.48312599212096 -1.71893918214267 1.82017120999877
H 4.06049887070731 -1.05185642779775 2.93868348335497
H 7.40365570400844 -0.37197703288191 1.43321520977277
H 8.27305123312518 2.52615427235979 0.66977281824739
H 11.13363020515501 -2.02743205607122 -1.20467988485005
H 10.73138580250827 2.49003346820822 0.69451039529536
H 12.99611181712918 -0.93245985935317 -0.96142096786638
$end
I don’t know NEMO at all, but just running Q-Chem 6.1 on our cluster, your input runs to completion in 416 sec on 20 processors. It’s only 454 basis functions so memory/disk requirements should be quite modest. Do you have the ability to just run Q-Chem from a linux shell?
there are 500 of these jobs (which different geometries, displaced according to the normal modes), and i can only run qchem 5.4.1 on a cluster it seems to be okay if i run them one by hand, but when i submit all of the at the same time sometimes they crash before printing the final part of the qchem output file and I get these strange errors as above.
Acually scratch some of that - if i run them 1 by 1 they still crash for me sometimes
Strange that this only seems to happen with these types of jobs, I can run other jobs fine. They just sometimes crash at the point after “Transition Moments Between Triplet Excited States” are printed. Is there something special qchem does after this?
what is the last part of the output?
-----------------------------------------------------
State X Y Z(a.u.)
-----------------------------------------------------
0 2483.856526 285.477123 -15.741104
-----------------------------------------------------
Within CIS/TDA Excited States:
Electron Dipole Moments of Triplet Excited State
-----------------------------------------------------
State X Y Z(a.u.)
-----------------------------------------------------
1 2483.182964 286.788981 -15.053309
3 2481.988968 287.639653 -14.402910
4 2490.244852 288.148640 -15.885217
6 2486.106596 286.596149 -15.508314
8 2486.630431 287.554415 -15.545583
-----------------------------------------------------
Transition Moments Between Ground and Triplet Excited States
--------------------------------------------------------------------------------
States X Y Z Strength(a.u.)
--------------------------------------------------------------------------------
0 1 0.000000 0.000000 0.000000 -9.432241E-43
0 3 0.000000 0.000000 -0.000000 3.646884E-39
0 4 0.000000 -0.000000 -0.000000 2.310974E-38
0 6 0.000000 0.000000 -0.000000 3.104072E-37
0 8 0.000000 0.000000 0.000000 4.118646E-39
--------------------------------------------------------------------------------
Transition Moments Between Triplet Excited States
--------------------------------------------------------------------------------
States X Y Z Strength(a.u.)
--------------------------------------------------------------------------------
1 3 0.003077 0.185958 0.046226 0.00147793
1 4 0.163834 0.470487 0.138782 0.01174803
1 6 0.476248 -1.000953 -0.703023 0.1058322
1 8 -0.974537 0.036864 0.074823 0.06391053
3 4 -0.451032 0.081307 0.133315 0.0008390954
3 6 0.813702 0.041789 -0.145655 0.01451224
3 8 -0.471338 0.379340 0.229695 0.01112522
4 6 -0.376975 -0.302664 -0.055987 0.00414494
4 8 -0.459309 0.369070 0.164894 0.008565555
6 8 0.233313 -0.240004 -0.078075 0.0006355328
--------------------------------------------------------------------------------
This is the last part of the output, it just abruptly ends
so you don’t get
Transition Moments Between Singlet Excited States
which is below that?
Ah yes indeed! It seems that all jobs either go to completion, or crash right before Transition moments between singlet excited states
Is that part of the calculation particularly memory or disk intensive?
Shouldn’t be. Have you contacted Q-Chem support? I’m afraid this is beyond my abilities to diagnose at the moment, apologies for the slow follow-up.
Ah no problem! I’ll contact Q-Chem support!
