misterbrandonwalker
commented
3 years ago Hello,
Please try the latest commit from Friday, we observed some issues from very recent changes.
As for the deleting of opt.log, this typically only happens if the coordinates from molecule-opt.xyz (cartesian xyz directly from QM log file) file and the coordinates from molecule.xyz (tinker xyz) are not the same. This can happen if user changed opt method (then everything else needs to be removed after). As an example, the SymmetryWater and SymmetryMethane both have this issue if you inspect their coordinate files, probably because when these files were originally generated we had a different opt method (so user might thing to remove all xyz, key and qm log files log, however the tinker xyz file comes from GDMA output, so more files actually need to be removed. You can see in this case the Methane and Water examples online are using DFT (we had gone through a period we were using this for OPT), but now we use MP2 as our default.
On Sat, Nov 13, 2021 at 3:32 AM Jacek Kozuch @.***> wrote:
Hi,
I am observing that poltype2 gets stuck directly after the first optimization step for a few larger molecules than methanol (e.g. penicillin G, p-coumaric acid). The opt_1.log file shows normal termination (last few lines):
Job cpu time: 0 days 5 hours 53 minutes 58.0 seconds. Elapsed time: 0 days 0 hours 56 minutes 22.6 seconds. File lengths (MBytes): RWF= 82 Int= 0 D2E= 0 Chk= 13 Scr= 1 Normal termination of Gaussian 16 at Sat Nov 13 01:40:57 2021.
poltype.log shows:
Sat Nov 13 00:44:29 2021 Running on host: c142.curta.zedat.fu-berlin.de Sat Nov 13 00:44:29 2021 NEED QM Density Matrix: Executing Gaussian Opt and SP Sat Nov 13 00:44:29 2021 Percent of jobs finished 0.0 Sat Nov 13 00:44:29 2021 Calling: GAUSS_SCRDIR=/scratch/jkozuch/Gau-PCC /trinity/shared/software/gaussian/g16_A03/g16/g16 PCC-opt_1.com path = /home/jkozuch/Param/PCC
At this point nothing happens for hours. I then restart poltype to continue from here, but the normally terminated *_opt_1.log is not recognized and is overwritten, as a new OPT_1 calculation starts.
My poltype.ini looks as follows: structure=PCC.sdf totalcharge=-1 numproc=8 maxmem=64GB maxdisk=500GB atmidx=455 use_gaus rotalltors
The MeOH example with essentially the same input doesn't have an issue at this point. However, it seems to get stuck in the torsion section (unless there's something happening in the background). I also just checked methyl propionate as a smaller molecule with two torsions, instead of the one in MeOH) and I have the same issue - opt_1 is finished; nothing else happens after that; restart starts from scratch and overwrites opt_1.log.
Thanks, Jacek
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/TinkerTools/poltype2/issues/5, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKNB26NJZ45QKTWBYQECAV3ULYWDPANCNFSM5H6N467Q . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
jacekkozuch
Hi,
I am observing that poltype2 gets stuck directly after the first optimization step for a few larger molecules than methanol (e.g. penicillin G, p-coumaric acid). The opt_1.log file shows normal termination (last few lines):
poltype.log shows:
At this point nothing happens for hours. I then restart poltype to continue from here, but the normally terminated *_opt_1.log is not recognized and is overwritten, as a new OPT_1 calculation starts.
My poltype.ini looks as follows: structure=PCC.sdf totalcharge=-1 numproc=8 maxmem=64GB maxdisk=500GB atmidx=455 use_gaus rotalltors
The MeOH example with essentially the same input doesn't have an issue at this point. However, it seems to get stuck in the torsion section (unless there's something happening in the background). I also just checked methyl propionate as a smaller molecule with two torsions, instead of the one in MeOH) and I have the same issue - opt_1 is finished; nothing else happens after that; restart starts from scratch and overwrites opt_1.log.
Addition after I wrote this. I tried the series MeOH, EtOH, PrOH, and BuOH, making the molecule longer by one CH3 and I observe the issue described above with BuOH, where the first three molecules work. Everything is similar as stated above.
Best, Jacek