Recovery DB

[Jared Miller]

Last night we had one of our ADT processes freeze up, and had to reboot and I believe the issue is being caused by an issue related to the recovery DB.

Our master ADT system is Star.  They are also sending DFT^P03 messages on to that thread, but they are not routed to go to any other systems.  However, it seems all those DFTs are remaining in the recovery DB in a message 1 state.

Why would those messages that are not routed to go anywhere be sitting in the recovery DB?  

Thanks

Topic

[Jared Miller]

In additiion to that, what’s the best way to get those messages out of the recovery DB?  I would like to prevent another freeze up.  Thanks again!

[Jim Kosloskey]

Jared,

You need to do something with ALL messages coming into the engine.

If there is no destination for the DFT Messages to go, just KILL them.

Jim Kosloskey

email: jim.kosloskey@jim-kosloskey.com 30+ years Cloverleaf, 60 years IT – old fart.

[Robert Milfajt]

Jim is right, you need to kill them.  I remember back in Cloverleaf training hearing about this, all messages need a destination.  If I recall, you need to set up a “dummy route” with hcitpsmsgkill executed to wipe out the message.

Hope this helps,

Robert Milfajt
Northwestern Medicine
Chicago, IL

[Jim Kosloskey]

You can also route the unwanted messages back to the inbound (be sure to have the hcitpsmsgkill proc on the route). That way there is no need for a ‘dummy’ thread and route.

Jim Kosloskey

email: jim.kosloskey@jim-kosloskey.com 30+ years Cloverleaf, 60 years IT – old fart.

[Author]

Replies