for anyone else running into an issue like this:

meltano run

had been running fine for a while, then suddenly started failing, with this error message

2023-06-05T14:48:29.839317Z [error    ] Extractor failed
2023-06-05T14:48:29.839773Z [error    ] Block run completed.           block_type=ExtractLoadBlocks err=RunnerError('Extractor failed') exit_codes={<PluginType.EXTRACTORS: 'extractors'>: -9} set_number=0 success=False

running with debug logging didn’t reveal anything additionally helpful eventually i realized that that

-9

wasn’t a meltano or singer-specific error code, it appears to be the signal the child process terminated with. in this case, it was SIGKILLed by the oom killer due to using too much RAM. normally i’d expect to see

137

in that case, since signal values are added to 128, but apparently python’s subprocess module will return a negative value for the signal that was used https://docs.python.org/3/library/subprocess.html#subprocess.CompletedProcess.returncode increasing the RAM solved the problem