Replications fail with the following error message (or another referencing 'keep alive'):
Error: 2181 - RETRY timedout operation timed out [at timeout_thread:424] FATAL repl_keepalive_fail unable to send keep alive message to replslave [at ping_slaves:742]
Since replications tend to happen across a WAN connection at times this might be a difficult error to rectify, especially in the case of having a limited connection. Regardless of the speed of the line however, you can limit the amount of jobs that Ranger tries to concurrently send across the WAN. Since the replication (in an ESXi environment) will be using the VAs you can, through Ranger:
Click on Tools > Options > My Jobs Configuration
Here you have the option to choose how many tasks the VAs can process at a time. Across the WAN you might want to consider dropping this setting to 1 to validate that it works/completes, and from that a point forward at your discretion try to increase this setting accordingly. If your jobs are already failing with this error however, it is our belief you are overutilizing your WAN connection in this situation and our link to the VA is dropping packets.
