DAC_INIT is recorded while the TDS listener for the Dedicated Admin Connection (DAC) initialises at instance startup. The DAC is the emergency side door: a reserved connection with its own scheduler and resources that lets an administrator connect to an otherwise unresponsive instance to diagnose and kill whatever is wedging it.
One initialisation per startup; that is the wait’s whole life.
Is It a Problem?
No; it is a startup marker on the benign list. The operationally useful thought it should trigger is not about the wait but about the feature: will the DAC work when you need it? By default it listens locally only; remote admin connections must be enabled (sp_configure) for remote DAC access, which is what you will want when the box is so unwell you cannot RDP to it comfortably.
A DAC you first test during a disaster is a DAC that fails during a disaster.
Common Causes
- Instance startup initialising the DAC listener. Nothing else.
What To Do
- Filter the wait; act on the feature instead.
- Enable remote DAC on production instances:
sp_configure 'remote admin connections', 1; RECONFIGURE;. - Test it once calmly:
sqlcmd -S admin:<server> -E, so the path is proven before an incident needs it.
How To See It
Rank waits with Get-WaitStatistics; one small entry per startup, and a reminder to keep the side door oiled.
Part of the SQL Server Wait Types Library.
Related deep dive: SOS_SCHEDULER_YIELD Wait Type.
Leave a Reply