No problem:
vito-b5c1d087-de52-4ec5-8311-dd53f00412f6
Could it be that this one had a different error, about not finding the collection referenced in load_result:
File “/data3/hadoop/yarn/local/usercache/jaapel/appcache/application_1645173896198_5801/container_e5014_1645173896198_5801_01_000002/venv/lib/python3.8/site-packages/openeo_driver/ProcessGraphDeserializer.py”, line 1511, in load_result
return env.backend_implementation.load_result(job_id=job_id, user=user, load_params=load_params, env=env)
File “/data3/hadoop/yarn/local/usercache/jaapel/appcache/application_1645173896198_5801/container_e5014_1645173896198_5801_01_000002/venv/lib/python3.8/site-packages/openeogeotrellis/backend.py”, line 500, in load_result
for _, asset in self.batch_jobs.get_results(job_id=job_id, user_id=user.user_id).items()
File “/data3/hadoop/yarn/local/usercache/jaapel/appcache/application_1645173896198_5801/container_e5014_1645173896198_5801_01_000002/venv/lib/python3.8/site-packages/openeogeotrellis/backend.py”, line 1381, in get_results
job_info = self._get_job_info(job_id=job_id, user_id=user_id)
File “/data3/hadoop/yarn/local/usercache/jaapel/appcache/application_1645173896198_5801/container_e5014_1645173896198_5801_01_000002/venv/lib/python3.8/site-packages/openeogeotrellis/backend.py”, line 720, in _get_job_info
job_info = registry.get_job(job_id, user_id)
File “/data3/hadoop/yarn/local/usercache/jaapel/appcache/application_1645173896198_5801/container_e5014_1645173896198_5801_01_000002/venv/lib/python3.8/site-packages/openeogeotrellis/job_registry.py”, line 186, in get_job
job_info, _ = self._read(job_id, user_id, include_done=True)
File “/data3/hadoop/yarn/local/usercache/jaapel/appcache/application_1645173896198_5801/container_e5014_1645173896198_5801_01_000002/venv/lib/python3.8/site-packages/openeogeotrellis/job_registry.py”, line 275, in _read
raise JobNotFoundException(job_id)
openeo_driver.errors.JobNotFoundException: The batch job ‘4cfc3e73-4847-483f-a580-34d765e86f30’ does not exist.
The same synchronous query gives me the aforementioned 500 error. Is this again because of the stripping of the vito- prefix?
The batch job id that is in the error that you shared is the one i use for the load_result step
FYI It is now also deployed in production, so no need anymore to do manual stripping, or connecting to dev instance
1 Like
Also to be sure we are not chasing ghosts: there two jobs in play here: a first one to generate what you call your cached result, and a second one that loads this result with load_result.
Are both jobs submitted on the same openeo backend (e.g. both on “openeo.cloud”) with exactly the same user account?
Yes correct, both submitted in the samebook!
I pinpointed the problem that load_result is not working in a batch job context:
root cause: batch job context only knows proxyuser and uses that as user_id, while job lookup for load_result requires standard (e.g. OIDC based) user id
Thanks for the investigation! I have subscribed to the GitHub thread for updates!
I also managed to reproduce the OpenEoApiError: [500] unknown: / by zero issue now and created another github issue: `load_result` failure · Issue #127 · Open-EO/openeo-geopyspark-driver · GitHub
I only get it when trying to download in NetCDF
when I try GeoTIFF, I can download a result succesfully, but it’s all nodata, so there is still something wrong I guess