Repository not updating?

[lemorgan]

I can't seem to access some runs that were collected over the last day. I can open the blank runs, but the standards I ran are appearing but giving the attached error when trying to open the file.

Of note is that the lab computer had a forced restart overnight due to a security update. I had tried to install/restart yesterday, but apparently I didn't do that correctly and so it went ahead. Is there a procedure for recovering from this? I've tried updating repositories through Pipeline and through GitHub desktop already.

[lemorgan]

pychron.current.log

[jirhiker]

Did you try pushing the analyses from the Data collection computer?

Id need to see the log from the computer that is giving the "Failed to the requested analyses..." to give any more insight. It appears the log file you shared is from the data collection computer and does not show any instances of failing to recall an analysis

[stephen-e-cox]

I think I might have seen this exact or similar messages on a data reduction machine when there was a local git repo screwup preventing it from updating new runs in just one repo.

[jirhiker]

That could be the case. In those situations one approach is to archive the offending repository and let pychron re clone it.

Note. This "deletes" any of you unshared local changes, but typically this is not an issue

[lemorgan]

I get the same error message on both data redux and data collection computers. I did try pushing from dev/dc but that didn't seem to help. Then I archived the repository but recloned it using github desktop, but still get the same message when I try to open any of the files in that repo.

[stephen-e-cox]

I just saw that at least some run data has been written to the repo called "laboratory," and I guess everything else must be hidden so I can't see them. So, can you find the analyses you're looking for in the Irradiation-DT-47 folder on the computer or the corresponding repo on Github?

[lemorgan]

ok, try this log file
pychron.current.log

[jirhiker]

Looks like you have a typo in your database/Metadata repo. The Irradiation is named DT-47 in the database but the irradiation is named DT-047 in the MetaData repo. I suggest editing the name of the Irradiation in the Database to DT-047. You can use something like MySQLWorkbench to edit the database directly.

The other option is to rename the irradiation in the local repository. This would require some git command line interaction to make permanent. However as a simple test to see if this typo is truely the issue, open Users/lemorgan/Documents/Pychron/WorkEnv/data/.dvc/MetaData and change the name of the folder DT-047 to DT-47

here is the relevant error in the log

git.repo.base                           : 2022-04-20 13:20:15,153 DEBUG     (MainThread) Cmd(['git', 'clone', '-v', 'https://[email protected]/USGSDenverData/Irradiation-DT-47.git', '/Users/lemorgan/Documents/Pychron/WorkEnv/data/.dvc/repositories/Irradiation-DT-47'])'s unused stdout: 
Pipeline Data Processing                : 2022-04-20 13:20:15,181 DEBUG     (MainThread) Traceback (most recent call last):
  File "/Users/lemorgan/Documents/Pychron/Pychron/pychron/envisage/browser/browser_task.py", line 217, in recall
    records, pull_frequency=pf, use_progress=False
  File "/Users/lemorgan/Documents/Pychron/Pychron/pychron/dvc/dvc.py", line 1144, in make_analyses
    prod_levels[level] = meta_repo.get_production(irrad, level)
  File "/Users/lemorgan/Documents/Pychron/Pychron/pychron/dvc/meta_repo.py", line 698, in get_production
    ip = Production(p, allow_null=allow_null)
  File "/Users/lemorgan/Documents/Pychron/Pychron/pychron/dvc/meta_object.py", line 43, in __init__
    raise MetaObjectException(msg)
pychron.dvc.meta_object.MetaObjectException: failed loading /Users/lemorgan/Documents/Pychron/WorkEnv/data/.dvc/MetaData/DT-47/productions/.json False```

[cmmercer]

Aha, thanks! We'll test and work to correct that!

[lemorgan]

Cam helped me solve this. Jake was right - we had changed DT-047 to DT-47 in the database but not in the metadata repo. We changed it there and pushed changes to github, which solved the problem.