We have two RHEL 6.2 servers (175-15-1 and 175-15-3) which have TimesTen Release 126.96.36.199.2 installed. We have a bunch of tables in TimesTen which are replicated("Transmit Durable") in both ways. Both 175-15-1 and 175-15-3 are subscriber and master for each other. Each host has multiple network interfaces. To make sure TimesTen uses a specific network interface, there is route configuration with the replication scheme.
Replication Route Information
Replication Scheme ROOT.175-15-1
Route Master I167 on 175-15-1 -> Subscriber I167 on 175-15-3
Replication Scheme ROOT.175-15-3
Route Master I167 on 175-15-3 -> Subscriber I167 on 175-15-1
The output of "ttrepadmin -showstatus" says that replication peers are connected.
Process ID : 7585 (Started)
Replication Agent Policy : always
Host : 175-15-1
RepListener Port : 55034 (AUTO)
Last write LSN : 1.40464632
Last LSN forced to disk : 1.40464384
Replication hold LSN : 1.40452344
Name : I167
Host : 175-15-3
Port : 56973 (AUTO) (Connected)
Replication State : STARTED
Communication Protocol : 29
For some reason, replication does not seem to work. The following are visible in the TimesTen logs. The same logs are printed on both hosts.
2012-10-05 15:44:11.04 Warn: REP: 1272: I167:receiver.c(2875): TT16060: Failed to read data from the network. TimesTen replication agent is stopping
2012-10-05 15:45:00.62 Warn: REP: 7151: I167:receiver.c(2875): TT16060: Failed to read data from the network. select() timed out
2012-10-05 15:45:02.35 Warn: REP: 7151: I167:transmitter.c(6004): TT16060: Failed to read data from the network. select() timed out
2012-10-05 15:45:02.35 Err : REP: 7151: I167:transmitter.c(3637): TT16121: Failed to flush transaction queue. Restarting log read loop
We thought firewall could be a problem. However, even after disabling the firewall, same errors are thrown.
What could be the cause of the problem? Thanks for an answer.
If there are any debug logs that can be enabled to diagnose the problem, please let us know.