Error in tinydb Provider for Processes

[justb4]

Happens in pygeometa metadata transformation using latest thus pygeoapi master version as on date 250119, thus including recent fix: geopython/pygeoapi#2209.

Using the URL:
https://demo.pygeoapi.io/master/processes/pygeometa-metadata-transform, the log output shows:

[2026-01-19T22:38:56Z] {/pygeoapi/pygeoapi/api/__init__.py:591} ERROR - fd4738c2-eb61-11f0-8e03-6d85883e42fb
Traceback (most recent call last):
  File "/pygeoapi/pygeoapi/process/manager/tinydb_.py", line 182, in get_job
    return found[0]
           ~~~~~^^^
IndexError: list index out of range

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/pygeoapi/pygeoapi/api/processes.py", line 551, in get_job_result
    job = api.manager.get_job(job_id)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/pygeoapi/pygeoapi/process/manager/tinydb_.py", line 184, in get_job
    raise JobNotFoundError()
pygeoapi.process.base.JobNotFoundError

Could be a config issue. If not, may be a pygeoapi issue.

[justb4]

Observation: does not happen just after Container restart...

[justb4]

Most likely the cronfile: demo.pygeoapi.io.cron is not executed as cronfiles in /etc/cron.d for Ubuntu should not contain chars like dots . Just renaming may fix...

[justb4]

Also does not work, file was not removed. Think cron under Ubuntu (Debian) is even more picky, sanity checking: mounted files and root-ownership.

Best solution IMO is like with the updater: add the cronfile (only on production host) to the host's /etc/cron.d and use Docker commands to clean container's /tmp: docker exec -it pygeoapi_master rm /tmp/pygeoapi-process-manager.db. @tomkralidis