计算化学公社

 找回密码 Forget password
 注册 Register
Views: 8677|回复 Reply: 8
打印 Print 上一主题 Last thread 下一主题 Next thread

[ORCA] 计算含有Pr的原子簇ORCA报错signal 6 (Aborted).

[复制链接 Copy URL]

126

帖子

1

威望

416

eV
积分
562

Level 4 (黑子)

跳转到指定楼层 Go to specific reply
楼主
各位老师好!
在ORCA计算含有重元素Pr的原子簇时总是崩溃退出,已经调整了输入文件的基组和收敛标准但没什么用。。。
ORCA4.1.1和4.1.2都有这个问题,对应用的都是OpenMPI3.1.3

输入文件:
  1. ! HF

  2. ! NODIIS SlowConv

  3. %scf
  4.     maxIter 1000
  5.     MaxIntMem 81920
  6.     MaxDisk 1024000
  7. end

  8. %basis
  9.     basis "def2-TZVP"
  10. end

  11. %maxCore 5120

  12. %pal
  13.     nProcs 16
  14. end

  15. %output
  16.     XYZFile true
  17. end

  18. * xyz -1 5
  19. Pr                -0.95184600   -0.00034200    0.00000000    newGTO "SARC-DKH-TZVP" end
  20. ...
  21. *
复制代码
输出文件:
  1. ...
  2. ------------------------------
  3. INITIAL GUESS: MODEL POTENTIAL
  4. ------------------------------
  5. Loading Hartree-Fock densities                     ... done
  6.   calling /work/chem/Software/ORCA_411/orca HF-1-5-PrB8_atom59.inp > HF-1-5-PrB8_atom59.out in order to generate an atomic fitting density for atom 0 (Pr with ECP) on-the-fly...
  7.     atom 0 (Pr), assumed electronic state with S=4: 1s2 2s2 2p6 3s2 3p6 4s2 3d10 4p6 5s2 4d10 5p6 6s2 4f3           ... [file orca_scf/scfguess.cpp, line 4720, Process 0]:
  8. Error encountered when trying to calculate the atomic fitting density!

  9. ORCA finished by error termination in SCF
  10. Calling Command: mpirun -np 16  /work/chem/Software/ORCA_411/orca_scf_mpi HF-1-5-PrB8.gbw b HF-1-5-PrB8
  11. [file orca_tools/qcmsg.cpp, line 458]:
  12.   .... aborting the run
复制代码
报错信息:
  1. [file orca_scf/scfguess.cpp, line 4720, Process 0]:
  2. Error encountered when trying to calculate the atomic fitting density!
  3. [r13n40:58358] *** Process received signal ***
  4. [r13n40:58358] Signal: Aborted (6)
  5. [r13n40:58358] Signal code:  (-6)
  6. [r13n40:58358] [ 0] /lib64/libc.so.6(+0x36280)[0x2b7e0a9e2280]
  7. [r13n40:58358] [ 1] /lib64/libc.so.6(gsignal+0x37)[0x2b7e0a9e2207]
  8. [r13n40:58358] [ 2] /lib64/libc.so.6(abort+0x148)[0x2b7e0a9e38f8]
  9. [r13n40:58358] [ 3] /work/chem/Software/ORCA_411/orca_scf_mpi[0x7954d2]
  10. [r13n40:58358] [ 4] /work/chem/Software/ORCA_411/orca_scf_mpi[0x50f23c]
  11. [r13n40:58358] [ 5] /work/chem/Software/ORCA_411/orca_scf_mpi[0x462a60]
  12. [r13n40:58358] [ 6] /work/chem/Software/ORCA_411/orca_scf_mpi[0x46e65a]
  13. [r13n40:58358] [ 7] /work/chem/Software/ORCA_411/orca_scf_mpi[0x47570c]
  14. [r13n40:58358] [ 8] /lib64/libc.so.6(__libc_start_main+0xf5)[0x2b7e0a9ce3d5]
  15. [r13n40:58358] [ 9] /work/chem/Software/ORCA_411/orca_scf_mpi[0x461e11]
  16. [r13n40:58358] *** End of error message ***
  17. --------------------------------------------------------------------------
  18. Primary job  terminated normally, but 1 process returned
  19. a non-zero exit code. Per user-direction, the job has been aborted.
  20. --------------------------------------------------------------------------
  21. --------------------------------------------------------------------------
  22. mpirun noticed that process rank 0 with PID 0 on node r13n40 exited on signal 6 (Aborted).
  23. --------------------------------------------------------------------------
  24. [file orca_tools/qcmsg.cpp, line 458]:
  25.   .... aborting the run
复制代码
还请各位老师能不吝赐教!非常感谢!


126

帖子

1

威望

416

eV
积分
562

Level 4 (黑子)

9#
 楼主 Author| 发表于 Post on 2019-11-11 14:18:01 | 只看该作者 Only view this author
最近这个问题又出现了,经过仔细的观察,问题产生的原因是$orcadir/orca $jobName_atomXX.inp > $jobName_atomXX.out这一步没有成功运行,具体原因不详,但更改运行ORCA的命令(例如加入&或者用nohup等)有一定几率避免这个问题

339

帖子

0

威望

4999

eV
积分
5338

Level 6 (一方通行)

8#
发表于 Post on 2019-4-23 14:51:05 | 只看该作者 Only view this author
DoorBell 发表于 2019-4-23 11:24
就是最上面那个文件直接删掉SARC-DKH-TZVP那一句

坐标给的不全吧

126

帖子

1

威望

416

eV
积分
562

Level 4 (黑子)

7#
 楼主 Author| 发表于 Post on 2019-4-23 11:24:24 | 只看该作者 Only view this author
chrinide 发表于 2019-4-22 21:00
你发输入文件给我瞧瞧吧

就是最上面那个文件直接删掉SARC-DKH-TZVP那一句

339

帖子

0

威望

4999

eV
积分
5338

Level 6 (一方通行)

6#
发表于 Post on 2019-4-22 21:00:49 | 只看该作者 Only view this author
DoorBell 发表于 2019-4-22 20:52
呃,不过SARC-DKH-TZVP是我因为崩溃而加上的
不加SARC-DKH-TZVP的错误报告:

你发输入文件给我瞧瞧吧

126

帖子

1

威望

416

eV
积分
562

Level 4 (黑子)

5#
 楼主 Author| 发表于 Post on 2019-4-22 20:52:50 | 只看该作者 Only view this author
chrinide 发表于 2019-4-22 19:54
你输入文件有问题,用 def2-TZVP计算的时候 Pr元素是默认带 def2-ECP 的赝势计算,随后你将 Pr元素换用全电 ...

呃,不过SARC-DKH-TZVP是我因为崩溃而加上的
不加SARC-DKH-TZVP的错误报告:
  1. ------------------------------
  2. INITIAL GUESS: MODEL POTENTIAL
  3. ------------------------------
  4. Loading Hartree-Fock densities                     ... done
  5.   calling /work/chem/Software/ORCA_411/orca HF-1-5-PrB8_atom59.inp > HF-1-5-PrB8_atom59.out in order to generate an atomic fitting density for atom 0 (Pr with ECP) on-the-fly...
  6.     atom 0 (Pr), assumed electronic state with S=4: 1s2 2s2 2p6 3s2 3p6 4s2 3d10 4p6 5s2 4d10 5p6 6s2 4f3           ... [file orca_scf/scfguess.cpp, line 4720, Process 0]:
  7. Error encountered when trying to calculate the atomic fitting density!

  8. ORCA finished by error termination in SCF
  9. Calling Command: mpirun -np 16  /work/chem/Software/ORCA_411/orca_scf_mpi HF-1-5-PrB8.gbw b HF-1-5-PrB8
  10. [file orca_tools/qcmsg.cpp, line 458]:
  11.   .... aborting the run
复制代码
  1. [file orca_scf/scfguess.cpp, line 4720, Process 0]:
  2. Error encountered when trying to calculate the atomic fitting density!
  3. [r13n41:339870] *** Process received signal ***
  4. [r13n41:339870] Signal: Aborted (6)
  5. [r13n41:339870] Signal code:  (-6)
  6. [r13n41:339870] [ 0] /lib64/libc.so.6(+0x36280)[0x2b1d0b237280]
  7. [r13n41:339870] [ 1] /lib64/libc.so.6(gsignal+0x37)[0x2b1d0b237207]
  8. [r13n41:339870] [ 2] /lib64/libc.so.6(abort+0x148)[0x2b1d0b2388f8]
  9. [r13n41:339870] [ 3] /work/chem/Software/ORCA_411/orca_scf_mpi[0x7954d2]
  10. [r13n41:339870] [ 4] /work/chem/Software/ORCA_411/orca_scf_mpi[0x50f23c]
  11. [r13n41:339870] [ 5] /work/chem/Software/ORCA_411/orca_scf_mpi[0x462a60]
  12. [r13n41:339870] [ 6] /work/chem/Software/ORCA_411/orca_scf_mpi[0x46e65a]
  13. [r13n41:339870] [ 7] /work/chem/Software/ORCA_411/orca_scf_mpi[0x47570c]
  14. [r13n41:339870] [ 8] /lib64/libc.so.6(__libc_start_main+0xf5)[0x2b1d0b2233d5]
  15. [r13n41:339870] [ 9] /work/chem/Software/ORCA_411/orca_scf_mpi[0x461e11]
  16. [r13n41:339870] *** End of error message ***
  17. --------------------------------------------------------------------------
  18. Primary job  terminated normally, but 1 process returned
  19. a non-zero exit code. Per user-direction, the job has been aborted.
  20. --------------------------------------------------------------------------
  21. --------------------------------------------------------------------------
  22. mpirun noticed that process rank 0 with PID 0 on node r13n41 exited on signal 6 (Aborted).
  23. --------------------------------------------------------------------------
  24. [file orca_tools/qcmsg.cpp, line 458]:
  25.   .... aborting the run
复制代码

339

帖子

0

威望

4999

eV
积分
5338

Level 6 (一方通行)

4#
发表于 Post on 2019-4-22 19:54:09 | 只看该作者 Only view this author
你输入文件有问题,用 def2-TZVP计算的时候 Pr元素是默认带 def2-ECP 的赝势计算,随后你将 Pr元素换用全电子 SARC-DKH-TZVP 基组,你让ORCA懵逼了,到底是做赝势ECP计算,还是全电子DKH标量相对论计算,知道不,所以ORCA挂逼了!!!!!!!

解决方法一:直接def2-TZVP的赝势计算,算含Pr的肯定没问题,别再瞎改用全电子基组

解决方法二:改用全电子标量相对论计算,加关键词DKH或者ZORA,用相关的全电子基组

126

帖子

1

威望

416

eV
积分
562

Level 4 (黑子)

3#
 楼主 Author| 发表于 Post on 2019-4-22 19:11:35 | 只看该作者 Only view this author
sobereva 发表于 2019-4-22 18:17
换个初猜方式再试

PS:你不会真想用HF算吧

非常感谢sob老师!

在%scf加入Guess HCore之后可以计算了,不过就是初猜超级烂,迭代一次能量能下降几十个Eh。
另外HF就是先做一次粗略的计算拿产生的.gbw做初猜用的。

5万

帖子

99

威望

5万

eV
积分
112544

管理员

公社社长

2#
发表于 Post on 2019-4-22 18:17:08 | 只看该作者 Only view this author
换个初猜方式再试

PS:你不会真想用HF算吧
北京科音自然科学研究中心http://www.keinsci.com)致力于计算化学的发展和传播,长期开办高质量的各种计算化学类培训:初级量子化学培训班中级量子化学培训班高级量子化学培训班量子化学波函数分析与Multiwfn程序培训班分子动力学与GROMACS培训班CP2K第一性原理计算培训班,内容介绍以及往届资料购买请点击相应链接查看。这些培训是计算化学从零快速入门以及进一步全面系统性提升研究水平的高速路!培训各种常见问题见《北京科音办的培训班FAQ》
欢迎加入“北京科音”微信公众号获取北京科音培训的最新消息、避免错过网上有价值的计算化学文章!
欢迎加入人气非常高、专业性特别强的综合性理论与计算化学交流QQ群“思想家公社QQ群”:1号:18616395,2号:466017436,3号:764390338,搜索群号能搜到哪个说明目前哪个能加,合计9000人。北京科音培训班的学员在群中可申请VIP头衔,提问将得到群主Sobereva的最优先解答。
思想家公社的门口Blog:http://sobereva.com(发布大量原创计算化学相关博文)
Multiwfn主页:http://sobereva.com/multiwfn(十分强大的量子化学波函数分析程序)
ResearchGate:https://www.researchgate.net/profile/Tian_Lu
Money and papers are rubbish, get a real life!

手机版 Mobile version|北京科音自然科学研究中心 Beijing Kein Research Center for Natural Sciences|京公网安备 11010502035419号|计算化学公社 — 北京科音旗下高水平计算化学交流论坛 ( 京ICP备14038949号-1 )|网站地图

GMT+8, 2024-11-27 18:22 , Processed in 0.157397 second(s), 22 queries , Gzip On.

快速回复 返回顶部 返回列表 Return to list