Endless queue for process graph using Landsat Col2 TIR band

Hi all,
this process graph tests temporal aggregation for the Landsat Collection TIR bands, however the batch process is stuck in an endless queue and does not start. What’s the issue here?

job ID: vito-j-f119a3e74aad4bd899918ea49ccf07f5

{
“process_graph”: {
“load3”: {
“arguments”: {
“bands”: [
“B10”
],
“id”: “LANDSAT8-9_L2”,
“spatial_extent”: {
“east”: 12.857630310674619,
“north”: 42.21443746558316,
“south”: 41.66416158120455,
“west”: 12.083250049971543
},
“temporal_extent”: [
“2022-08-01T00:00:00Z”,
“2022-10-10T00:00:00Z”
]
},
“process_id”: “load_collection”
},
“reduce1”: {
“arguments”: {
“data”: {
“from_node”: “load3”
},
“dimension”: “t”,
“reducer”: {
“process_graph”: {
“max1”: {
“arguments”: {
“data”: {
“from_parameter”: “data”
}
},
“process_id”: “max”,
“result”: true
}
}
}
},
“description”: “Compute the max over the temporal dimension”,
“process_id”: “reduce_dimension”
},
“savere1”: {
“arguments”: {
“data”: {
“from_node”: “reduce1”
},
“format”: “GTIFF”
},
“description”: “Store as GTIFF”,
“process_id”: “save_result”,
“result”: true
}
}
}

At fist sight, it looks that something went wrong with creating the job in the first place, causing issues when trying to start the job.

Can you try creating a new job and starting that one?

(we are also investigating why the creation went wrong by the way)

Tried it again, but it again resulted in an error:
ID:vito-j-d9d5e619b3fa42be983bb1e08891b38c
Submitted:10/12/2023, 1:21:13 PM UTC
Updated:10/12/2023, 2:18:57 PM UTC
Status:error

FYI: Running that process graph synchronously seems to work fine.

e.g.

import openeo
con = openeo.connect("openeo.cloud").authenticate_oidc()
pg = """
{
  "process_graph": {
    "load3": {
      "arguments": {
        "bands": ["B10"],
        "id": "LANDSAT8-9_L2",
        "spatial_extent": {"east": 12.857630310674619, "north": 42.21443746558316, "south": 41.66416158120455, "west": 12.083250049971543},
        "temporal_extent": ["2022-08-01T00:00:00Z", "2022-10-10T00:00:00Z"]
      },
      "process_id": "load_collection"
    },
    "reduce1": {
      "arguments": {
        "data": {"from_node": "load3"},
        "dimension": "t",
        "reducer": {"process_graph": {"max1": {"arguments": {"data": {"from_parameter": "data"}}, "process_id": "max", "result": true}}}
      },
      "description": "Compute the max over the temporal dimension",
      "process_id": "reduce_dimension"
    },
    "savere1": {
      "arguments": {"data": {"from_node": "reduce1"}, "format": "GTIFF"},
      "description": "Store as GTIFF",
      "process_id": "save_result",
      "result": true
    }
  }
}
"""
con.download(pg, "forum633-orig-platform-sync.tiff")

But indeed, I can reproduce that running it as a batch job does not work at the moment.

escalating this to a bug in github tracker: Failing batch jobs with LANDSAT8-9_L2 collection · Issue #540 · Open-EO/openeo-geopyspark-driver · GitHub

I can confirm that this works now! thanks