Details
-
Improvement
-
Status: Resolved
-
Major
-
Resolution: Fixed
-
5.10.0
-
None
-
None
Description
The following priority back up config does not work:
failover:(tcp://localhost:1234,tcp://primary1:1234,tcp://secondary1:1234)? nested.socket.tcpNoDelay=true& nested.soTimeout=60000& nested.closeAsync=false& randomize=false& priorityBackup=true& priorityURIs=tcp://localhost:1234
This is because the full connected URI is used when testing whether the current connection is the priority connection. In the above example the nested socket params are added to the actual connected URI which then does not match tcp://localhost:1234 as setup in the priority list.
This causes the transport to continually try to reconnect to the primary, even though it is already connected to it.
Can it either take nested parameters into account so the above works as expected (it looks like a sensible URL to me) and/or validate that each of the priorityBackupURIs are valid URIs in the failover list