[BUG] - QCG Terminates with Backtrace #357
Comments
|
Also, I can confirm the same error when running example 2 on the documentation website found here. I'm using |
|
Also can confirm this isn't a |
|
Hi, which version of xtb are you using for the CREST runs? In version 6.7.1, there was a bug that caused a segmentation fault in the docking module, which might be the reason for the error you encountered. If you're using this version, I recommend switching to 6.7.1 or to the bleeding edge version of xtb to avoid this issue. |
|
I'm not using an xtb binary. I'm using the CREST |
|
QCG still uses xtb, there is a disclaimer in the README.md. It will have uses whatever binary it has found as there should be a error stop if none is present. |
|
I see! That could be the issue! Really appreciate the feedback here to know what's going on. It would be excellent if the program communicated these types of messages ( Much thanks! Lemme try to run in an environment with an |
|
It actually already does that, here's the relevant part in QCG: Lines 56 to 71 in a12cb48 Are you sure there wasn't an xtb somewhere in path? |
|
Ah. I see. And yes, I have Is there some reason you'd expect this executable to not work? |
|
In the interests of not making yet another thread to report this, I'll comment here to report that I am also having the same issue. In contrast to @matsumuraf in the other related thread, I can't run QCG example 1 either though and get the following output: The error output with example 2 is similar: And with my own attempt: This last calculation was run with methanol rather than water as the solvent, I can provide structures if desired but it doesn't seem to be specific to that. Following the advice above I tried using the bleeding-edge binary of xtb rather than 6.7.1 but that doesn't change anything. This is with the pre-built binary of CREST 3.0.2, not the conda version, for reference, and all on Linux. |
|
Hi, I am having the same issue and - sorry - I don't get if there is a solution from the posts above. xtb version 6.7.1 crest 3.0.2 (pre-built binary). |
|
No, there is no de facto solution yet. Still waiting for some input of @cplett I think it's probably a combination of different things. For example, in the last comment by @matterhorn103 the path to the xtb executable is set as relative path. That won't work at all because QCG seems to switch working directories internally and there is no relative to absolute path conversion. Could at least explain the failures in the "Preoptimization" part. |
Ok well if I use absolute paths and run my final example with: /home/matt/.local/share/easyxtb/bin/crest input.xyz --qcg aux1.xyz --nsolv 2 --T 2 --xnam /home/matt/.local/share/easyxtb/bin/xtb-bleed/bin/xtbit works, but only when I use the bleeding-edge version of xtb, with xtb 6.7.1 I still get an error similar to the first one I posted: Example 1 still fails though even before the cluster generation begins: On the other hand, running example 1 with absolute paths and bleeding-edge xtb it seems to do the initial cluster growing and first MTD iteration fine, but crashes during the multi-level optimization: |
Can I open an issue for this? |
It's fine, I think I can take care of that and refer to this issue here. Regarding the bleeding edge vs 6.7.1 issue, it might be because of this: grimme-lab/xtb#1089 which, as I understand, is not in 6.7.1 |
|
Hi, any news about this issue? |
|
Hi, I am still trying to make QCG work. I tried everything posted in this thread before with no success. "For any installation make sure that you have correctly installed and sourced the xtb program before attempting any calculations with CREST. While xtb is technically not needed for the primary runtypes of CREST versions >3.0 thanks to an integration of tblite, some functionalities, like QCG, still require it!" Of course I have xtb installed, but is it possibly it is not correctly linked to crest ("sourced")? Thanks |
|
Hi all,
@matterhorn103, regarding the crashed multilevel optimizations: this might be a problem of the QCG and the new conformer-search routines. For me, this didn't occur, but perhaps using the legacy algorithm originally tested with the QCG is more stable. You can enable it by providing the |
|
it works, but unfortunately not always. sometimes I get "error while reading inout coordinates" :( |
|
Thank you for trying. |
|
Here there are (called "input" but also includes the output). The command was |
Describe the bug
Running a quantum cluster growth calculation terminates with what appears to be a an issue reading structures.
To Reproduce
crest solute.xyz --qcg solvent.xyz --nsolv 10 --keepdir --gfnff --chrg 0 --uhf 0solute.xyz
solvent.xyz
CREST Stdout
╔════════════════════════════════════════════╗ ║ ___ ___ ___ ___ _____ ║ ║ / __| _ \ __/ __|_ _| ║ ║ | (__| / _|\__ \ | | ║ ║ \___|_|_\___|___/ |_| ║ ║ ║ ║ Conformer-Rotamer Ensemble Sampling Tool ║ ║ based on the xTB methods ║ ║ ║ ╚════════════════════════════════════════════╝ Version 3.0.2, Sun, 25 August 20:02:44, 08/25/2024 commit (af7eb99) compiled by 'usr@fv-az732-492' Cite work conducted with this code as • P.Pracht, F.Bohle, S.Grimme, PCCP, 2020, 22, 7169-7192. • S.Grimme, JCTC, 2019, 15, 2847-2862. • P.Pracht, S.Grimme, C.Bannwarth, F.Bohle, S.Ehlert, G.Feldmann, J.Gorges, M.Müller, T.Neudecker, C.Plett, S.Spicher, P.Steinbach, P.Wesołowski, F.Zeller, J. Chem. Phys., 2024, 160, 114110. for works involving QCG cite • S.Spicher, C.Plett, P.Pracht, A.Hansen, S.Grimme, JCTC, 2022, 18 (5), 3174-3189. • C.Plett, S. Grimme, Angew. Chem. Int. Ed. 2023, 62, e202214477. for works involving MECP screening cite • P.Pracht, C.Bannwarth, JCTC, 2022, 18 (10), 6370-6385. Original code P.Pracht, S.Grimme, Universität Bonn, MCTC with help from (alphabetical order): C.Bannwarth, F.Bohle, S.Ehlert, G.Feldmann, J.Gorges, S.Grimme, C.Plett, P.Pracht, S.Spicher, P.Steinbach, P.Wesolowski, F.Zeller Online documentation is available at https://crest-lab.github.io/crest-docs/ This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License (LGPL) for more details. Command line input: $ crest solute.xyz -qcg solvent.xyz --nsolv 10 --keepdir --gfnff --chrg 0 --uhf 0 --gfnff : Use of GFN-FF requested. --chrg 0 --uhf 0 > Setting up backup calculator ... done. ---------------- Calculation info ---------------- > User-defined calculation level: : GFN-FF calculation via GFNFF lib : Molecular charge : 0 ======================================== | ---------------- | | Q C G | | ---------------- | | Quantum Cluster Growth | | University of Bonn, MCTC | ======================================== S. Grimme, S. Spicher, C. Plett. Cite work conducted with this code as S. Spicher, C. Plett, P. Pracht, A. Hansen, S. Grimme, JCTC, 2022, 18, 3174-3189. The use of the aISS algorithm is requested (recommend). This requires xtb version 6.6.0 or newer. xTB-IFF can still be used with the --xtbiff flag. ========================================= | quantum cluster growth: INPUT | ========================================= QCG: Only Cluster Generation input parameters solute : solute.xyz charge : 0 uhf : 0 solvent : solvent.xyz # of solvents to add : 10 Cluster generated that are above 10 % populated # of CPUs used : 1 Solvation model : xtb opt level : normal System temperature [K] : 298.1 RRHO scaling factor : 0.75 Solute geometry molecular radius (Bohr**1): 10.19 molecular area (Bohr**2): 2013.72 molecular volume (Bohr**3): 4427.45 Solvent geometry molecular radius (Bohr**1): 5.59 molecular area (Bohr**2): 429.00 molecular volume (Bohr**3): 732.83 radius of solute : 16.42 radius of solvent : 9.02 ========================================= | Preoptimization | ========================================= Total Energy of solute not found Total Energy of solvent not found ________________________________________________________________________ __________________ Solute Cluster Generation _____________________ ________________________________________________________________________ ========================================= | quantum cluster growth: GROW | ========================================= Solute: unit ellipsoid axis a,b,c : 0.333 0.333 0.333 Solvent: unit ellipsoid axis a,b,c : 0.333 0.333 0.333 solvent anisotropy : 1.000 solute anisotropy : 1.000 roff inner wall : 0.000 solute max dist : 23.878 solvent max dist : 5.525 inner unit axis : 0.333 0.333 0.333 inner ellipsoid/Bohr : 2.388 2.388 2.388 scaling factor outer ellipsoid: 1.000 outer ellipsoid/Bohr : 2.763 2.763 2.763 Size E De Detot Density Eatom av. R Rlast Volume Opt [Eh] [kcal] [kcal] [u/Å^3] [kcal] [bohr] [bohr] [bohr^3] Note: The following floating-point exceptions are signalling: IEEE_UNDERFLOW_FLAG IEEE_DENORMAL ERROR STOP error while reading input coordinates Error termination. Backtrace: #0 0xced1d0 in ??? #1 0xced479 in ??? #2 0xcee587 in ??? #3 0x845de4 in __strucrd_MOD_rdxmol at /home/runner/work/crest/crest/src/strucreader.f90:1028 #4 0x846c83 in __strucrd_MOD_rdcoord at /home/runner/work/crest/crest/src/strucreader.f90:936 #5 0x6a94d3 in qcg_grow_ at /home/runner/work/crest/crest/src/qcg/solvtool.f90:661 #6 0x6b3e68 in crest_solvtool_ at /home/runner/work/crest/crest/src/qcg/solvtool.f90:87 #7 0x40e52b in crest at /home/runner/work/crest/crest/src/crest_main.f90:261 #8 0x408f7e in main at /home/runner/work/crest/crest/src/crest_main.f90:26Expected behavior
The QCG algorithm would execute correctly.
The text was updated successfully, but these errors were encountered: