MultiBackendJobManager: don't count "created" jobs as "running"

Note that this behavior was present since beginning of MultiBackendJobManager (b0d100e),
but is unnecessary as the resource consumption of "created" jobs is
negligible compared to queued or running jobs.

Author: soxofaan

CHANGELOG.md

@@ -16,6 +16,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - Internal reorganization of `openeo.extra.job_management` submodule to ease future development ([#741](https://github.com/Open-EO/openeo-python-client/issues/741))
 - `openeo.Connection`: add some more HTTP error codes to the default retry list: `502 Bad Gateway`, `503 Service Unavailable` and `504 Gateway Timeout` ([#835](https://github.com/Open-EO/openeo-python-client/issues/835))
+- `MultiBackendJobManager`: don't count "created" jobs as "running"
 
 ### Removed
 

openeo/extra/job_management/_manager.py

@@ -541,8 +541,7 @@ def _job_update_loop(
         not_started = job_db.get_by_status(statuses=["not_started"], max=200).copy()
         if len(not_started) > 0:
             # Check number of jobs running at each backend
-            # TODO: should "created" be included in here? Calling this "running" is quite misleading then.
-            running = job_db.get_by_status(statuses=["created", "queued", "queued_for_start", "running"])
+            running = job_db.get_by_status(statuses=["queued", "queued_for_start", "running"])
             stats["job_db get_by_status"] += 1
             per_backend = running.groupby("backend_name").size().to_dict()
             _log.info(f"Running per backend: {per_backend}")