In Redshift loader (wise variant) I encounter this...
# troubleshooting
k
In Redshift loader (wise variant) I encounter this error.
Copy code
2024-08-28T17:51:34.077848Z [info     ]     if stream in state.get('bookmarks', {}): cmd_type=elb consumer=True job_name=prod:tap-hubspot-to-target-redshift name=target-redshift producer=False run_id=ae9839d1-47f1-4a5e-8cc8-07ddd90c36df stdio=stderr string_id=target-redshift
2024-08-28T17:51:34.078202Z [info     ] AttributeError: 'NoneType' object has no attribute 'get' cmd_type=elb consumer=True job_name=prod:tap-hubspot-to-target-redshift name=target-redshift producer=False run_id=ae9839d1-47f1-4a5e-8cc8-07ddd90c36df stdio=stderr string_id=target-redshift
And based on the code, this 'state' thing is the one that's managed by Meltano. On the tap side, I do see this message.
No state was found, complete import.
I assume this is all related to this state (https://sdk.meltano.com/en/latest/implementation/state.html) but it does say I shouldn't worry about it. Is there an easy fix for this? This is happening on the first/initial data pull, using the scheduler. It loads the first set of records, then throws this error. Do I need to specify
replication_key
somehow myself?
v
if it's the wise variant then they don't use the sdk so it's not related to that. Looks like it's expecting state to be sent over and maybe the tap isn't sending state over
k
Yes, I think tap is outputting
No state was found, complete import.
I also see
Copy code
| INFO     | tap-airbyte          | {'level': 'INFO', 'message': 'Setting state of SourceHubspot stream to {}'} cmd_type=elb consumer=False job_name=prod:tap-hubspot-to-target-redshift name=tap-hubspot producer=True run_id=ae9839d1-47f1-4a5e-8cc8-07ddd90c36df stdio=stderr string_id=tap-hubspot
This setting worked on another staging environment so it's odd that it's not working now.
v
you could try a different target, or a different tap, or modify the tap or the target to work for this. I think the target has a bug requiring the state to be there 🤷
k
I guess it may have worked on staging, primarily because that particular stream didn't have any data. Let me see if I can try a different target.
👍 1