fairgraph `File` query fails with `code=500 'Internal Server Error'`

[mih]

Report is here https://github.com/datalad/datalad-ebrains/pull/35#issuecomment-1344125357

I just reconfirmed that with my account things are still working as documented in that PR.

[adswa]

It looks like a failure in the middle of a successful process to me. The dataset did get quite a bit of history

adina@muninn in /tmp/kgdstryout on git:master!
❱ git log --oneline     
5a86cfa (HEAD -> master, tag: v2.6) In this version, the maximum probability map is comprised of 119 brain regions in each hemisphere and available for the following coordinate spaces: + MNI ICBM 152 (2009c Nonlinear Asymmet[...]
9ebd888 (tag: v2.5) In this version, the maximum probability map is comprised of 119 brain regions in each hemisphere and available for the following coordinate spaces: + MNI ICBM 152 (2009c Nonlinear Asymmetric) + Colin [...]
adef09b (tag: v2.4) In this version, the maximum probability map is comprised of 120 brain regions in each hemisphere and available for the following coordinate spaces: + MNI ICBM 152 (2009c Nonlinear Asymmetric) + Colin [...]
a898d5f (tag: v2.2) In this version, the maximum probability map is comprised of 120 brain regions in each hemisphere and available for the following coordinate spaces: + MNI ICBM 152 (2009c Nonlinear Asymmetric) + Colin [...]
ae29844 (tag: v1.18) In this version, the maximum probability map is comprised of 106 brain regions in each hemisphere and available for the following coordinate spaces: + MNI ICBM 152 (2009c Nonlinear Asymmetric) + Colin [...]
a326f45 (tag: v1.13) This is the first version of this dataset. In this version, the maximum probability map is comprised of 77 brain regions in each hemisphere and available for the following coordinate spaces: + Colin 27[...]
f473170 [DATALAD] new dataset

Its current state is this, it seems all contained files were deleted

❱ git status
On branch master
Changes not staged for commit:
  (use "git add/rm <file>..." to update what will be committed)
  (use "git restore <file>..." to discard changes in working directory)
    deleted:    MPM-collections/26/Licence%20%28CCBY-NC-SA%29.pdf
    deleted:    MPM-collections/26/ListOfPMFiles_icbm152casym.csv
    deleted:    MPM-collections/26/MPM/JulichBrain_MPMAtlas_l_N10_nlin2Stdicbm152asym2009c_publicDOI_3f5ec6016bc2242769c41befdbc1b2e0.nii.gz
    deleted:    MPM-collections/26/MPM/JulichBrain_MPMAtlas_l_N10_nlin2Stdicbm152asym2009c_publicDOI_3f5ec6016bc2242769c41befdbc1b2e0.xml
    deleted:    MPM-collections/26/MPM/JulichBrain_MPMAtlas_r_N10_nlin2Stdicbm152asym2009c_publicDOI_453c4c95827e38dcce5372b200065cba.nii.gz
    deleted:    MPM-collections/26/MPM/JulichBrain_MPMAtlas_r_N10_nlin2Stdicbm152asym2009c_publicDOI_453c4c95827e38dcce5372b200065cba.xml
    deleted:    MPM-collections/26/jubrain-ontology_26.json
    deleted:    version

If I restore those deletions, I can get files fine:

(ebrains) adina@muninn in /tmp/kgdstryout on git:master
❱ datalad get MPM-collections/26/jubrain-ontology_26.json
get(ok): MPM-collections/26/jubrain-ontology_26.json (file) [from web...]                                                                                                                                                        
(ebrains) adina@muninn in /tmp/kgdstryout on git:master
❱ datalad get MPM-collections/26/Licence%20%28CCBY-NC-SA%29.pdf
get(ok): MPM-collections/26/Licence%20%28CCBY-NC-SA%29.pdf (file) [from web...]    

As a small side note, the file names have some occasional encoding issues (e.g., Licence%20%28CCBY-NC-SA%29.pdf).

[mih]

Today it happened again. There was no code change since the last successful run on the very same dataset.

[mih]

The error seems to be specific to the particular query being made. Here is the line where it fails

>>> ds = omcore.Dataset.list(self.client, versions=dv)
-> crash
>>> type(dv)
<class 'fairgraph.openminds.core.products.dataset_version.DatasetVersion'>
# no obvious issues with that instance
>>> self.client.user_info()
User(alternate_name='mih', name='...', email=None, given_name='...', family_name='...', identifiers=['fc583e67-...'])

So the general connection and basic authentication seems to be working still in this error case.

[mih]

One day later, things are back working. No change in code or account status.

[mih]

A fix is coming as discussed in #58