TERMINATE_LISTENER is recorded when a thread waits for a SQL Server Network Interface (SNI) listener to be destroyed. SNI is the protocol layer wrapping TDS packets over whatever transports are configured (TCP, named pipes, shared memory), and listeners get torn down during instance shutdown, endpoint reconfiguration, and network setting changes.
A teardown marker from the networking layer.
Is It a Problem?
No; listener destruction is an infrequent lifecycle event, and this wait’s occurrences map to those events. It carries no signal about connection performance or network health during normal operation.
If a shutdown or a network reconfiguration hangs, this wait showing on the relevant thread would simply confirm which phase is slow; the cause would live in whatever keeps the listener busy (draining connections, a wedged transport).
Common Causes
- Instance shutdown tearing down network listeners.
- Endpoint or protocol configuration changes recreating listeners.
- AG listener operations touching the SNI layer.
What To Do
- Filter it out of wait analysis.
- For hung shutdowns, read it as a phase marker alongside the error log’s shutdown sequence.
How To See It
Rank waits with Get-WaitStatistics; expect only traces around lifecycle events.
Part of the SQL Server Wait Types Library.
Related deep dive: ASYNC_NETWORK_IO Wait Type.
Leave a Reply