terraform-provider-confluent: Postgres CDC Source Connector never finishes provisioning
Description
I am attempting to provision a new Postgres CDC Source Connector in Confluent Cloud. However, when applying the plan that implements the new connector resource, the Connector stays in the provisioning state and never errors out or completes. I tried to apply it earlier this morning and let it run for 2 hours with no error or completion. Terraform then timed out and cancelled the run. I deleted the Connector that was in the provisioning state in Confluent cloud. I then kicked off another plan apply and it still stays in this “provisioning” state longer than I’d anticipate. Lastly, we’ve checked our database logs and find no evidence of the Connector connecting to our database or trying to do anything within the database (i.e. no new publication, nothing in the replication slot, etc.).
Details
- Confluent Provider Version: 1.28.0
- Connection Type: We are using an AWS Peering Connection for our database that lives in AWS. We verified that the VPCs can connect to each other.
- Resource Type:
confluent_connector - Connector Class:
PostgresCdcSource - Example used as reference: https://github.com/confluentinc/terraform-provider-confluent/tree/master/examples/configurations/connectors/postgre-sql-cdc-debezium-source-connector
- Since we are still working on our staging environment, all of our resources have lifecycle
prevent_destroyexplicitly set to false.
If there are any other details that I can provide to help troubleshoot, please let me know. Thanks in advance!
About this issue
- Original URL
- State: closed
- Created a year ago
- Comments: 20 (1 by maintainers)
Looks like there’s an incident today that “All Connector provisioning/deprovisioning has been delayed.”. Can you please try again? https://status.confluent.cloud/incidents/y72ql4p0db3q
@linouk23 yup, understood. Makes sense to switch over to that route at this point. I really appreciate all your guidance here.
Hi @linouk23 , just following up here, confirming that the connector is in fact still provisioning. I assume that it’d be best to leave the connector alone until we isolate the issue?
@linouk23 I have a UI-initiated connector still provisioning at the moment. I’ll let it continue to provision for a bit and see if anything further happens. After an hour or two, I’ll provide an update on whether we were successful or not.
I’ll also email the connector and env IDs. Thanks!
@linouk23 I just sent the email with the requested configs. Please let me know if you need anything else from me!