cancel
Showing results for 
Search instead for 
Did you mean: 

Waited for incoming connection for 60 seconds...

Breck_Carter
Participant
0 Kudos
1,999

How should I approach this symptom in an arbiter-partner1-partner2-copy setup?

16.0.0.2419 read-only scale-out server is failing to connect over the WAN, and it produces the following I messages every 60 seconds.

The E message appears only very occasionally.

I. 04/24 12:44:38. Database "XXX" mirroring: Waited for incoming connection for 60 seconds before requesting log page, now dropping connections
I. 04/24 12:44:38. Database "XXX" mirroring:  disconnected from parent "DONORCENTRAL_primary"
I. 04/24 12:44:38. Database "XXX" mirroring:  parent reconnection process starting
E. 04/24 12:44:38. Database "XXX" mirroring:  mirror connection to server "DONORCENTRAL_primary" failed: connection already exists
I. 04/24 12:44:40. Database "XXX" mirroring:  connected to parent "DONORCENTRAL_primary"
I. 04/24 12:44:40. Database "XXX" mirroring:  synchronizing ...
I. 04/24 12:44:41. Database "XXX" mirroring:  parent reconnection process stopping

Accepted Solutions (1)

Accepted Solutions (1)

ian_mchardy
Product and Topic Expert
Product and Topic Expert

So your main problem appears to be that the outgoing connection from this server to the primary can be established, but the connection going the other way cannot (either due to misconfiguration or a firewall or something else in your network blocking it). If you don't know why this is occurring, you can start by enabling -z or sa_server_option('DebuggingInformation') on both the copy node and the primary and look at the messages in the server console. They should give you a hint about what the outgoing and incoming connections are attempting to do.

I've seen the "E. 04/24 12:44:38. Database "XXX" mirroring: mirror connection to server "DONORCENTRAL_primary" failed: connection already exists" message sporadically occur when the server attempts to create a mirroring connection immediately after the last one was dropped, but internally we haven't fully cleaned up the old connection. In the cases I've seen it, this condition should clear with in a few seconds and isn't anything to worry about given what is happening when it is occurring.

I hope this helps,

Ian

Breck_Carter
Participant

In this case the client reports "firewall" was both problem and solution... similar to "beer" 🙂

Answers (0)