Filed by: ncsoregi [at] mozilla.com
Parsed log: https://treeherder.mozilla.org/logviewer?job_id=353812285&repo=mozilla-central
Full log: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/NLavZi_5R7-DhqjMPonmtQ/runs/0/artifacts/public/logs/live_backing.log

[task 2021-10-06T04:22:16.659Z] Generating target_task_set @ 92641110e5c9
[task 2021-10-06T04:28:33.078Z] Looking for existing index gecko.v2.mozilla-central.revision.66f9b72b87297adf712b14be58df13c2333bb3a9.taskgraph.decision-nightly-desktop to avoid triggering multiple nightlies off the same revision...
[task 2021-10-06T04:28:33.079Z] Index gecko.v2.mozilla-central.revision.66f9b72b87297adf712b14be58df13c2333bb3a9.taskgraph.decision-nightly-desktop doesn't exist.
[task 2021-10-06T04:28:37.332Z] Generating target_task_set @ 73f23b9a5cf6
[task 2021-10-06T04:28:37.351Z] concurrent.futures.process._RemoteTraceback: 
[task 2021-10-06T04:28:37.351Z] """
[task 2021-10-06T04:28:37.351Z] Traceback (most recent call last):
[task 2021-10-06T04:28:37.351Z]   File "/usr/lib/python3.6/concurrent/futures/process.py", line 175, in _process_worker
[task 2021-10-06T04:28:37.351Z]     r = call_item.fn(*call_item.args, **call_item.kwargs)
[task 2021-10-06T04:28:37.351Z]   File "/builds/worker/checkouts/gecko/taskcluster/gecko_taskgraph/main.py", line 129, in format_taskgraph
[task 2021-10-06T04:28:37.351Z]   File "/builds/worker/checkouts/gecko/taskcluster/gecko_taskgraph/main.py", line 102, in get_taskgraph_generator
[task 2021-10-06T04:28:37.351Z] ModuleNotFoundError: No module named 'gecko_taskgraph.generator'
[task 2021-10-06T04:28:37.351Z] """
[task 2021-10-06T04:28:37.351Z] 
[task 2021-10-06T04:28:37.351Z] The above exception was the direct cause of the following exception:
[task 2021-10-06T04:28:37.351Z] 
[task 2021-10-06T04:28:37.351Z] ModuleNotFoundError: No module named 'gecko_taskgraph.generator'
[task 2021-10-06T04:28:37.352Z] 
[task 2021-10-06T04:28:37.352Z] 
[task 2021-10-06T04:28:37.352Z] concurrent.futures.process._RemoteTraceback: 
[task 2021-10-06T04:28:37.352Z] """
[task 2021-10-06T04:28:37.352Z] Traceback (most recent call last):
[task 2021-10-06T04:28:37.352Z]   File "/usr/lib/python3.6/concurrent/futures/process.py", line 175, in _process_worker
[task 2021-10-06T04:28:37.352Z]     r = call_item.fn(*call_item.args, **call_item.kwargs)
[task 2021-10-06T04:28:37.352Z]   File "/builds/worker/checkouts/gecko/taskcluster/gecko_taskgraph/main.py", line 129, in format_taskgraph
[task 2021-10-06T04:28:37.352Z]   File "/builds/worker/checkouts/gecko/taskcluster/gecko_taskgraph/main.py", line 102, in get_taskgraph_generator
[task 2021-10-06T04:28:37.352Z] ModuleNotFoundError: No module named 'gecko_taskgraph.generator'
[task 2021-10-06T04:28:37.352Z] """
[task 2021-10-06T04:28:37.352Z] 
[task 2021-10-06T04:28:37.352Z] The above exception was the direct cause of the following exception:
[task 2021-10-06T04:28:37.352Z] 
[task 2021-10-06T04:28:37.352Z] ModuleNotFoundError: No module named 'gecko_taskgraph.generator'
[task 2021-10-06T04:28:37.352Z] 
[task 2021-10-06T04:28:37.352Z] 
[task 2021-10-06T04:28:37.352Z] concurrent.futures.process._RemoteTraceback: 
[task 2021-10-06T04:28:37.352Z] """
[task 2021-10-06T04:28:37.352Z] Traceback (most recent call last):
[task 2021-10-06T04:28:37.352Z]   File "/usr/lib/python3.6/concurrent/futures/process.py", line 175, in _process_worker
[task 2021-10-06T04:28:37.352Z]     r = call_item.fn(*call_item.args, **call_item.kwargs)
[task 2021-10-06T04:28:37.352Z]   File "/builds/worker/checkouts/gecko/taskcluster/gecko_taskgraph/main.py", line 129, in format_taskgraph
[task 2021-10-06T04:28:37.352Z]   File "/builds/worker/checkouts/gecko/taskcluster/gecko_taskgraph/main.py", line 102, in get_taskgraph_generator
[task 2021-10-06T04:28:37.352Z] ModuleNotFoundError: No module named 'gecko_taskgraph.generator'
[task 2021-10-06T04:28:37.352Z] """
[task 2021-10-06T04:28:37.352Z] 

:ahal, can you take a look at this please?
thank you!

Ah, this is expected because the tgdiff task updates to the base revision of the push (in this case the module was back to the old name). Moving forward this task should start passing on new pushes again. I filed bug 1734388 to look into ways of avoiding confusion here.. though I think we won't see these types of failures too frequently going forward.