MDC Troubleshooting Checklist

• Are there any old patches in basho-patches?

• Has each node in the source/sink cluster(s) been tuned?

• Network connectivity verified

  • iperf etc
  • netstat -a (included in riak debug)
  • Using EC2? Suggest this doc: http://docs.basho.com/riak/latest/ops/tuning/aws/

• Static data

  • priority (urgent/high/low)
  • version of Riak EE
    • from sources AND sinks
  • version of replication: v2 | v3 | AAE
  • topology
    • how many clusters
    • bidirectional?
  • app.config config
    • from sources AND sinks

• Command line status

  • colllected from the cluster_leaders on source AND sink
    • riak-repl status
    • riak-repl connections
    • riak-admin status

• Erlang state
  • cluster_mgr state rp(sys:get_status(riak_core_cluster_manager)).
  • connection_mgr state rp(sys:get_status(riak_core_connection_manager)).
  • service_mgr state rp(sys:get_status(riak_core_service_manager)).
  • riak_repl2_leader state rp(sys:get_status(riak_repl2_leader_gs)).
  • riak_repl_leader_gs state rp(sys:get_status(riak_repl_leader_gs)).

Fullsync issues

• riak_repl2_fscoordinator rp(sys:get_status(riak_repl2_fscoordinator)).
• riak_repl2_fscoordinator_serv rp(sys:get_status(riak_repl2_fscoordinator_serv)).

Realtime issues

• riak_repl2_rtq rp(sys:get_status(riak_repl2_rtq))