Looks like path to VABSLib.so is hardcoded

rdrclrbh commented 2 years ago

I tried it on my local Linux and that resolved the python / scipy issue, but I am still having trouble with this case (see below, see the prevabs error, the path that it is looking for VABSLib.so looks hardcoded):

duhd1993 commented 2 years ago

It should be fixed with most recent update. Can you have a try?

rdrclrbh commented 2 years ago

It looks like this fixed that error, but I still am not getting a .sg file created. So I am not sure what is causing that failure.

duhd1993 commented 2 years ago

Can you copy the error message? what happens if you try running VABS with generated .sg file manually?

rdrclrbh commented 2 years ago

How do I run it manually?

rdrclrbh commented 2 years ago

Here is the error message:

(py37) [rdcrlrbh@crlrbh-ln-r1902 try_0]$ python3 run.py uh60_blade.yml 1 input.in output.out

INFO     | 2022-06-22 10:13:46 | dakota.initialize | initializing DakotaDesignAnalysis...
CRITICAL | 2022-06-22 10:13:46 | dakota.analysis | eval 5 start
CRITICAL | 2022-06-22 10:13:46 | dakota.analysis | running step: generate cs design
INFO     | 2022-06-22 10:13:46 | blade.run | running design analysis for uh60_blade_1...
INFO     | 2022-06-22 10:13:46 | design_analysis.preprocess | pre-processing...
INFO     | 2022-06-22 10:13:46 | blade.readBladeDesignInputYaml | generating blade design...
INFO     | 2022-06-22 10:13:46 | blade.readBladeParamDistr | reading distribution of parameter oml...
INFO     | 2022-06-22 10:13:46 | blade.readBladeParamDistr | reading distribution of parameter geo...
INFO     | 2022-06-22 10:13:46 | blade.readBladeParamDistr | reading distribution of parameter layup (spar)...
INFO     | 2022-06-22 10:13:46 | blade.readBladeParamDistr | reading distribution of parameter layup (others)...
INFO     | 2022-06-22 10:13:46 | blade.readBladeParamDistr | reading distribution of parameter lamina...
INFO     | 2022-06-22 10:13:46 | blade.generateCSDesign | generating cross-sectional designs...
INFO     | 2022-06-22 10:13:46 | blade.genCrossSectionDesignR | generating cs design at 0.2: uh60_blade_1_cs1
INFO     | 2022-06-22 10:13:46 | blade.genCrossSectionDesignR | generating cs design at 0.8: uh60_blade_1_cs2
INFO     | 2022-06-22 10:13:46 | design_analysis.postprocess | post-processing...
CRITICAL | 2022-06-22 10:13:46 | dakota.analysis | running step: cs analysis
INFO     | 2022-06-22 10:13:46 | cs.run | running design analysis for uh60_blade_1_cs1...
INFO     | 2022-06-22 10:13:46 | design_analysis.preprocess | pre-processing...
INFO     | 2022-06-22 10:13:46 | design_analysis.preprocess | cwd: /home/rdcrlrbh/galaxy/runs/UH60_opt_strength_constraint_soga_0.7/2022-06-21-10_49_27.429_-0400-256/attempts/stage_0/point_5/try_0
INFO     | 2022-06-22 10:13:46 | design_analysis.preprocess | import data_proc_funcs as user_mod
/home/rdcrlrbh/galaxy/runs/UH60_opt_strength_constraint_soga_0.7/2022-06-21-10_49_27.429_-0400-256/starting_files/data_proc_funcs.py
INFO     | 2022-06-22 10:13:46 | design_analysis.preprocess | evaluating user function: user_mod.materialId2Name
INFO     | 2022-06-22 10:13:46 | cs.runH | substituting airfoil_gbox_uni.xml.tmp -> uh60_blade_1_cs1.xml...
INFO     | 2022-06-22 10:13:46 | execu.solve | preprocessing...
INFO     | 2022-06-22 10:13:46 | presg.preSG | preprocessing structure gene...
CRITICAL | 2022-06-22 10:13:46 | execu.run | prevabs -i uh60_blade_1_cs1.xml -h
2022-06-22 10:13:46 [info] prevabs start

========================================

  PreVABS 1.5.0

  (For VABS 4.0 and SwiftComp 2.1)

========================================

2022-06-22 10:13:46 [info] - reading input files
2022-06-22 10:13:46 [info]     - reading main input file: uh60_blade_1_cs1.xml
2022-06-22 10:13:46 [info]       - reading airfoil data file: SC1095.dat
2022-06-22 10:13:46 [info]       - reading materials file: /home/rdcrlrbh/bin/ivabs/bin/MaterialDB.xml
2022-06-22 10:13:46 [info]       - reading materials file: material_database_us_ft.xml
2022-06-22 10:13:46 [info] - reading input files -- done

2022-06-22 10:13:46 [info] - building the shape
2022-06-22 10:13:46 [info]   - building the cross section, step 1
2022-06-22 10:13:46 [info]     - building component: spar
2022-06-22 10:13:46 [info]     - building component: le
2022-06-22 10:13:46 [info]     - building component: te
TimeoutExpired
Process killed
INFO     | 2022-06-22 10:13:56 | execu.solve | running analysis...
CRITICAL | 2022-06-22 10:13:56 | execu.run | VABS uh60_blade_1_cs1.sg
 Cannot open the file uh60_blade_1_cs1.sg for reading!                                                                                                                                                                                                                                                       
 Cannot open the file for writing error message
INFO     | 2022-06-22 10:13:56 | execu.solve | reading results...
INFO     | 2022-06-22 10:13:56 | dakota.writeOutput | writing output files...
INFO     | 2022-06-22 10:13:56 | dakota.writeOutput | eval 5 failed
Traceback (most recent call last):
  File "/home/rdcrlrbh/bin/ivabs/scripts/msgd/ivabs.py", line 184, in runDakotaEval
    dda.analysis()
  File "/home/rdcrlrbh/bin/ivabs/scripts/msgd/da/dakota.py", line 605, in analysis
    design_analysis.run()
  File "/home/rdcrlrbh/bin/ivabs/scripts/msgd/da/cs.py", line 60, in run
    self.runH()
  File "/home/rdcrlrbh/bin/ivabs/scripts/msgd/da/cs.py", line 134, in runH
    logger=self.logger
  File "/home/rdcrlrbh/bin/ivabs/scripts/msgpi/sg/execu.py", line 98, in solve
    results = msi.readVABSOut(sg_in, analysis, scrnout)
  File "/home/rdcrlrbh/bin/ivabs/scripts/msgpi/sg/io.py", line 615, in readVABSOut
    return readVABSOutHomo(fn_in, scrnout)
  File "/home/rdcrlrbh/bin/ivabs/scripts/msgpi/sg/io.py", line 273, in readVABSOutHomo
    with open(fn, 'r') as fin:
FileNotFoundError: [Errno 2] No such file or directory: 'uh60_blade_1_cs1.sg.K'

wenbinyu-PU commented 2 years ago

Is the sg file generated? Looks time you can increase your timeout threshold.

From: rdrclrbh @.> Sent: Wednesday, June 22, 2022 10:32 AM To: wenbinyugroup/ivabs @.> Cc: Subscribed @.***> Subject: Re: [wenbinyugroup/ivabs] Looks like path to VABSLib.so is hardcoded (Issue #52)

---- External Email: Use caution with attachments, links, or sharing data ----

Here is the error message:

(py37) @.*** try_0]$ python3 run.py uh60_blade.yml 1 input.in output.out

INFO | 2022-06-22 10:13:46 | dakota.initialize | initializing DakotaDesignAnalysis...

CRITICAL | 2022-06-22 10:13:46 | dakota.analysis | eval 5 start

CRITICAL | 2022-06-22 10:13:46 | dakota.analysis | running step: generate cs design

INFO | 2022-06-22 10:13:46 | blade.run | running design analysis for uh60_blade_1...

INFO | 2022-06-22 10:13:46 | design_analysis.preprocess | pre-processing...

INFO | 2022-06-22 10:13:46 | blade.readBladeDesignInputYaml | generating blade design...

INFO | 2022-06-22 10:13:46 | blade.readBladeParamDistr | reading distribution of parameter oml...

INFO | 2022-06-22 10:13:46 | blade.readBladeParamDistr | reading distribution of parameter geo...

INFO | 2022-06-22 10:13:46 | blade.readBladeParamDistr | reading distribution of parameter layup (spar)...

INFO | 2022-06-22 10:13:46 | blade.readBladeParamDistr | reading distribution of parameter layup (others)...

INFO | 2022-06-22 10:13:46 | blade.readBladeParamDistr | reading distribution of parameter lamina...

INFO | 2022-06-22 10:13:46 | blade.generateCSDesign | generating cross-sectional designs...

INFO | 2022-06-22 10:13:46 | blade.genCrossSectionDesignR | generating cs design at 0.2: uh60_blade_1_cs1

INFO | 2022-06-22 10:13:46 | blade.genCrossSectionDesignR | generating cs design at 0.8: uh60_blade_1_cs2

INFO | 2022-06-22 10:13:46 | design_analysis.postprocess | post-processing...

CRITICAL | 2022-06-22 10:13:46 | dakota.analysis | running step: cs analysis

INFO | 2022-06-22 10:13:46 | cs.run | running design analysis for uh60_blade_1_cs1...

INFO | 2022-06-22 10:13:46 | design_analysis.preprocess | pre-processing...

INFO | 2022-06-22 10:13:46 | design_analysis.preprocess | cwd: /home/rdcrlrbh/galaxy/runs/UH60_opt_strength_constraint_soga_0.7/2022-06-21-10_4927.429-0400-256/attempts/stage_0/point_5/try_0

INFO | 2022-06-22 10:13:46 | design_analysis.preprocess | import data_proc_funcs as user_mod

/home/rdcrlrbh/galaxy/runs/UH60_opt_strength_constraint_soga_0.7/2022-06-21-10_4927.429-0400-256/starting_files/data_proc_funcs.py

INFO | 2022-06-22 10:13:46 | design_analysis.preprocess | evaluating user function: user_mod.materialId2Name

INFO | 2022-06-22 10:13:46 | cs.runH | substituting airfoil_gbox_uni.xml.tmp -> uh60_blade_1_cs1.xml...

INFO | 2022-06-22 10:13:46 | execu.solve | preprocessing...

INFO | 2022-06-22 10:13:46 | presg.preSG | preprocessing structure gene...

CRITICAL | 2022-06-22 10:13:46 | execu.run | prevabs -i uh60_blade_1_cs1.xml -h

2022-06-22 10:13:46 [info] prevabs start

========================================

PreVABS 1.5.0

(For VABS 4.0 and SwiftComp 2.1)

========================================

2022-06-22 10:13:46 [info] - reading input files

2022-06-22 10:13:46 [info] - reading main input file: uh60_blade_1_cs1.xml

2022-06-22 10:13:46 [info] - reading airfoil data file: SC1095.dat

2022-06-22 10:13:46 [info] - reading materials file: /home/rdcrlrbh/bin/ivabs/bin/MaterialDB.xml

2022-06-22 10:13:46 [info] - reading materials file: material_database_us_ft.xml

2022-06-22 10:13:46 [info] - reading input files -- done

2022-06-22 10:13:46 [info] - building the shape

2022-06-22 10:13:46 [info] - building the cross section, step 1

2022-06-22 10:13:46 [info] - building component: spar

2022-06-22 10:13:46 [info] - building component: le

2022-06-22 10:13:46 [info] - building component: te

TimeoutExpired

Process killed

INFO | 2022-06-22 10:13:56 | execu.solve | running analysis...

CRITICAL | 2022-06-22 10:13:56 | execu.run | VABS uh60_blade_1_cs1.sg

Cannot open the file uh60_blade_1_cs1.sg for reading!

Cannot open the file for writing error message

INFO | 2022-06-22 10:13:56 | execu.solve | reading results...

INFO | 2022-06-22 10:13:56 | dakota.writeOutput | writing output files...

INFO | 2022-06-22 10:13:56 | dakota.writeOutput | eval 5 failed

Traceback (most recent call last):

File "/home/rdcrlrbh/bin/ivabs/scripts/msgd/ivabs.py", line 184, in runDakotaEval

dda.analysis()

File "/home/rdcrlrbh/bin/ivabs/scripts/msgd/da/dakota.py", line 605, in analysis

design_analysis.run()

File "/home/rdcrlrbh/bin/ivabs/scripts/msgd/da/cs.py", line 60, in run

self.runH()

File "/home/rdcrlrbh/bin/ivabs/scripts/msgd/da/cs.py", line 134, in runH

logger=self.logger

File "/home/rdcrlrbh/bin/ivabs/scripts/msgpi/sg/execu.py", line 98, in solve

results = msi.readVABSOut(sg_in, analysis, scrnout)

File "/home/rdcrlrbh/bin/ivabs/scripts/msgpi/sg/io.py", line 615, in readVABSOut

return readVABSOutHomo(fn_in, scrnout)

File "/home/rdcrlrbh/bin/ivabs/scripts/msgpi/sg/io.py", line 273, in readVABSOutHomo

with open(fn, 'r') as fin:

FileNotFoundError: [Errno 2] No such file or directory: 'uh60_blade_1_cs1.sg.K'

— Reply to this email directly, view it on GitHubhttps://github.com/wenbinyugroup/ivabs/issues/52#issuecomment-1163179746, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ASUPJ6JUU5TWT474EOWYS2DVQMPVVANCNFSM5ZMXJAYA. You are receiving this because you are subscribed to this thread.Message ID: @.**@.>>

rdrclrbh commented 2 years ago

No the .sg file is not generated

wenbinyu-PU commented 2 years ago

Can you increase the timeout to be 300 (equivalent to 5 mins)? It looks like that it takes more than 60 seconds to generate the cross-section file (*.sg)

From: rdrclrbh @.> Sent: Wednesday, June 22, 2022 10:35 AM To: wenbinyugroup/ivabs @.> Cc: Yu, Wenbin @.>; Comment @.> Subject: Re: [wenbinyugroup/ivabs] Looks like path to VABSLib.so is hardcoded (Issue #52)

---- External Email: Use caution with attachments, links, or sharing data ----

No the .sg file is not generated

— Reply to this email directly, view it on GitHubhttps://github.com/wenbinyugroup/ivabs/issues/52#issuecomment-1163183774, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ASUPJ6J5BCFZFXUJPOKP3JDVQMQARANCNFSM5ZMXJAYA. You are receiving this because you commented.Message ID: @.**@.>>

duhd1993 commented 2 years ago

Do you have these in the input

file_save: on
      work_directory:
        directory_tag: on
        directory_save: on

? This way the generated files will not be deleted by dakota so we can have a try manually example

study:
  method:
    multidim_parameter_study:
      partitions: [2, 2]
  variables:
    data_form: "compact"
    data: |
      a2p1,    design, continuous,      0.8:0.9
      a2p3,    design, continuous,      0.5:0.6
  interface:
    fork:
      parameters_file: "input.in"
      results_file: "output.out"
      file_save: on
      work_directory:
        directory_tag: on
        directory_save: on

rdrclrbh commented 2 years ago

where is timeout set? in the .yml file? The only timeout I find there is set to 10.

wenbinyu-PU commented 2 years ago

Yes, in yml file, line 116.

From: rdrclrbh @.> Sent: Wednesday, June 22, 2022 11:46 AM To: wenbinyugroup/ivabs @.> Cc: Yu, Wenbin @.>; Comment @.> Subject: Re: [wenbinyugroup/ivabs] Looks like path to VABSLib.so is hardcoded (Issue #52)

---- External Email: Use caution with attachments, links, or sharing data ----

where is timeout set? in the .yml file? The only timeout I find there is set to 10.

— Reply to this email directly, view it on GitHubhttps://github.com/wenbinyugroup/ivabs/issues/52#issuecomment-1163285006, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ASUPJ6I7K4TZAZDFIHDCR4DVQMYM7ANCNFSM5ZMXJAYA. You are receiving this because you commented.Message ID: @.**@.>>

rdrclrbh commented 2 years ago

Yes, it mine it was set to 10. I increased it to 60 and that may not be long enough. I'll try 300.

rdrclrbh commented 2 years ago

timeout = 300 does not seem to work either. Still getting no generation of the .sg file

wenbinyu-PU commented 2 years ago

The error is deeper than what I thought. You might want to remote desktop with Su to find out. I have no problem to run it on my linux box. I did an independent run my self.

From: rdrclrbh @.> Sent: Wednesday, June 22, 2022 12:01 PM To: wenbinyugroup/ivabs @.> Cc: Yu, Wenbin @.>; Comment @.> Subject: Re: [wenbinyugroup/ivabs] Looks like path to VABSLib.so is hardcoded (Issue #52)

---- External Email: Use caution with attachments, links, or sharing data ----

timeout = 300 does not seem to work either. Still getting not generation of the .sg file

— Reply to this email directly, view it on GitHubhttps://github.com/wenbinyugroup/ivabs/issues/52#issuecomment-1163303509, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ASUPJ6O4RFPWAT4P6UIG53TVQM2CVANCNFSM5ZMXJAYA. You are receiving this because you commented.Message ID: @.**@.>>

unitasium commented 2 years ago

@rdrclrbh Is this case the one I recently uploaded to cdmhub, or an older one?

rdrclrbh commented 2 years ago

it is the recent case. with only two stations along the blade length.

unitasium commented 2 years ago

Can you download the attached file and try to run it using prevabs? This is the first cross-section of the first evaluation of this example generated on our linux server. This one can be executed without any issue on our machine.

The command is

prevabs -i uh60_blade_1_cs1.xml -h -e

test.tar.gz

rdrclrbh commented 2 years ago

Oh, maybe I don't have the most recent case. Is teh case you sent an email about yesterday newer than the on eI got late last week?

unitasium commented 2 years ago

Yes.

rdrclrbh commented 2 years ago

Grabbing the new case now. I'll give it a try

rdrclrbh commented 2 years ago

The test case seemed to work fine (test.tar.gz)

rdrclrbh commented 2 years ago

OK, the test case case_0.7_3 ... seems to be working too. Though I think the .yml settings are such that no eval files are being saved. So I don't see the .sg and .K files.

rdrclrbh commented 2 years ago

ok, I modified the .yml and now I see the eval dirs. Looks good now. It has created the .sg and .K files like it should

rdrclrbh commented 2 years ago

I would like to know what is being changed to make the new case work while the old case did not work. Obviously the ivabs install, but something changed with the case too. i.e., a setting in the .yml file, or the data_proc_funcs.py?

rdrclrbh commented 2 years ago

OK, I have the new case working in GSB now.

unitasium commented 2 years ago

Actually this issue also confuses me. Some examples failed after I fixed the library linking issue and stopped at exactly the same location as yours (building component: te). Input files seems no difference. However, some cases works. So I just copy one of those workable cases and redo the example.

unitasium commented 2 years ago

The issue is identified. This is still caused by the hidden CRLF (^M) character. After removing it, the case should run properly.

wenbinyugroup / ivabs

Looks like path to VABSLib.so is hardcoded #52