bassel
07/24/2023, 10:51 PMbassel
07/24/2023, 10:57 PMMaking request for OperationModel(name=DeleteObjects) with params: {'url_path': '?delete', 'query_string': {}, 'method': 'POST', 'headers': {'User-Agent': 'Boto3/1.26.147 Python/3.9.16 Linux/4.14.318-241.531.amzn2.x86_64 exec-env/AWS_ECS_EC2 Botocore/1.29.147', 'Content-MD5': 'FdYOXo6N5+HngpLwbQKDSg=='}, 'body': b'<Delete xmlns="<http://s3.amazonaws.com/doc/2006-03-01/>"><Object><Key>us-east-1/metadata/mysql-to-s3/lock</Key></Object></Delete>', 'auth_path': '/etl-pipeline?delete', 'url': '<https://etl-pipeline.s3.amazonaws.com/?delete>', 'context': {'client_region': 'us-east-1', 'client_config': <botocore.config.Config object at 0x7fa605f21e50>, 'has_streaming_input': False, 'auth_type': 'v4', 's3_redirect': {'redirected': False, 'bucket': 'etl-pipeline', 'params': {'Bucket': 'etl-pipeline', 'Delete': {'Objects': [{'Key': 'us-east-1/metadata/mysql-to-s3/lock'}]}}}, 'signing': {'region': 'us-east-1', 'signing_name': 's3', 'disableDoubleEncoding': True}, 'checksum': {'request_algorithm': 'conditional-md5'}}}
bassel
07/24/2023, 10:59 PM<s3://etl-pipeline/us-east-1/metadata>
and the etl command end with: --state-id=mysql-to-s3
edgar_ramirez_mondragon
07/25/2023, 5:57 PMbassel
07/25/2023, 5:59 PM• Were you seeing the state lock before but not anymore?Yes, i was seeing it before
• what version of Meltano you using?Latest - from the docker image
• did you upgrade recently?No
bassel
07/25/2023, 5:59 PMedgar_ramirez_mondragon
07/25/2023, 6:02 PMbassel
07/25/2023, 6:07 PMedgar_ramirez_mondragon
07/25/2023, 6:47 PM<s3://some_bucket/some/path>
, the backend is trying to acquire the lock in <s3://some_bucket/some/path/some/path/my-state-id/lock>
.cody_hanson
07/25/2023, 10:19 PMMELTANO_STATE_BACKEND_URI
setting?
Also: are you seeing the same path duplication as described in the bug report that @edgar_ramirez_mondragon shared? And are you running into errors when running the pipeline or are you just not seeing the lock file as you would expect?bassel
07/26/2023, 1:53 PMCould you share the specific value of yoursettingMELTANO_STATE_BACKEND_URI
state_backend:
uri: $STATE_BACKEND_URI
s3:
aws_access_key_id: $STATE_BACKEND_AWS_ACCESS_KEY_ID
aws_secret_access_key: $STATE_BACKEND_AWS_SECRET_ACCESS_KEY
the uri field was: <s3://etl-pipeline/us-east-1/provider/metadata>
and the meltano's elt command ended with this flag:
--state-id=mysql-to-s3
So, what I was expecting: to have a folder named mysql-to-s3
with two files in it: lock & state.json
(that's what I had before when the uri
was only a bucket) and now I'm only having the state.json
I guess its not an issue - since out case will not have two or more pipelines running at the same time from the same source to the same destination .
Also: are you seeing the same path duplication as described in the bug report that @edgar_ramirez_mondragon shared?Had it before .. it was an issue in the uri itself .. then not anymore, but with only
state.json
no lock
And are you running into errors when running the pipeline or are you just not seeing the lock file as you would expect?Getting an error where it complains about the state and lock files, when running for the first time - if they don't exist, then they will get created.