Shunning not works on network cut off

[laszlo.suto]

There is two nodes: A, B. They are in one group, then we cut off the network cable of A then we notice from the log that two of them are in a single group. After plug in back the network cable into the A then they merged together but none of them get the actual state of the other.

Our configuration:
<UDP send_on_all_interfaces="true" receive_on_all_interfaces="true" mcast_addr="228.1.2.3" mcast_port="48866"
ip_ttl="64" ip_mcast="true"
mcast_send_buf_size="150000" mcast_recv_buf_size="80000"
ucast_send_buf_size="150000" ucast_recv_buf_size="80000"
use_incoming_packet_handler="false" use_outgoing_packet_handler="false"
loopback="false"/>
<PING timeout="2000" num_initial_members="3"
up_thread="true" down_thread="true"/>
<MERGE2 min_interval="10000" max_interval="20000"/>
<FD shun="true" up_thread="true" down_thread="true"
timeout="2500" max_tries="5"/>
<VERIFY_SUSPECT timeout="1500"
up_thread="false" down_thread="false"/>
<pbcast.NAKACK gc_lag="50" retransmit_timeout="600,1200,2400,4800"
max_xmit_size="8192" up_thread="false" down_thread="false"/>
<UNICAST timeout="600,1200,2400" window_size="100" min_threshold="10"
down_thread="false"/>
<pbcast.STABLE desired_avg_gossip="20000"

up_thread="false" down_thread="false"/>
<FRAG frag_size="8192"
down_thread="false" up_thread="false"/>
<pbcast.GMS join_timeout="5000" join_retry_timeout="2000"
shun="true" print_local_addr="true"/>

[brian.stansberry]

Do you mean their state does synchronize after the merge? This is the responsibility of the application (JGroups has no way of knowing what the state is, whether it should be merged, or how to merge it). If you implement the JGroups MembershipListener interface you can be notified of changes in the group membership.

[belaban]

You need to
- add pbcast.STATE_TRANSFER to the top and
- enable automatic fetching of state after reconnect: Channel.setOpt(Channel.AUTO_GETSTATE, Boolean.TRUE)