Service Exchange - Outbound Status = Down Recovery Requires Manual Queue Record Deletion

[MattCD]

We've noticed that when our instance, or a consumer instance, transitions to Outbound Status = Down (most commonly in non-production environments), the documented out-of-box recovery methods do not appear to restore the connection.

Method 1: Validate and Resolve from Provider Center

1. Open Provider Center.
2. Open the task for the downed connection.
3. Click Validate and Resolve.

Result: The task is closed and marked as resolved, but the connection remains down. A new "connection down" task is then generated.

---

Method 2: Validate and Activate Remote System

1. Open the affected Process Sync Remote System (ih_sync_remote_system) record.
2. Click Validate and Activate Remote System.

Result: The outbound state changes back to Active, but almost immediately returns to Error.

---

Method 3: Delete Error Queue Record and Reactivate (Consistently Works)

1. Open the affected Process Sync Remote System (ih_sync_remote_system) record.
2. Under Outbound Queue States, locate and delete the ih_sync_outbound_record that is in an error state.
3. Click Validate and Activate Remote System.

Result: The outbound state returns to Active and remains healthy (assuming the remote endpoint is available).

Afterward, we can return to Provider Center and successfully resolve the connection down task.

---

Question

Why is it necessary to manually delete the errored ih_sync_outbound_record before the connection can recover?

It seems that once the remote system becomes available again, the platform should either:

• Automatically retry and recover the connection, or
• Allow the Validate and Activate Remote System action to clear the error condition without manual intervention.

Has anyone else encountered this behavior, and is this expected product design or a known issue? 
@Kenny Caldwell I'm sure you already know the answer.

[MattCD]

I did find an article that suggest updating the state of the outbound record but again, this seems like design flaw that requires too much manual intervention to address common outage challenges.
Service Exchange outbound connection is down due to record stuck on "ERROR" state on Outbound Queue ...