[leo_manager] Inconsistent storage node status

[mocchira]

[E]     leofs12_0@192.168.100.12        2015-02-25 17:10:11.406757 +0900        1424851811      null:null       0       gen_server leo_manager_cluster_monitor terminated with reason: {timeout,{gen_server,call,[leo_redundant_manager,{update_member_by_node,'leofs14@192.168.100.14',1424851781405057,attached},30000]}} in gen_server:call/3 line 190
[E]     leofs12_0@192.168.100.12        2015-02-25 17:10:11.407673 +0900        1424851811      null:null       0       ["CRASH REPORT ",[80,114,111,99,101,115,115,32,"leo_manager_cluster_monitor",32,119,105,116,104,32,"0",32,110,101,105,103,104,98,111,117,114,115,32,"exited",32,119,105,116,104,32,114,101,97,115,111,110,58,32,[[123,["timeout",44,[123,["gen_server",44,"call",44,[91,["leo_redundant_manager",44,[123,["update_member_by_node",44,"'leofs14@192.168.100.14'",44,"1424851781405057",44,"attached"],125],44,"30000"],93]],125]],125]," in ",[["gen_server",58,"terminate",47,"7"],[32,108,105,110,101,32,"804"]]]]]
[E]     leofs12_0@192.168.100.12        2015-02-25 17:10:11.408403 +0900        1424851811      null:null       0       Supervisor leo_manager_sup had child leo_manager_cluster_monitor started with leo_manager_cluster_monitor:start_link() at <0.638.0> exit with reason {timeout,{gen_server,call,[leo_redundant_manager,{update_member_by_node,'leofs14@192.168.100.14',1424851781405057,attached},30000]}} in context child_terminated
[E]     leofs12_0@192.168.100.12        2015-02-25 17:10:27.962202 +0900        1424851827      leo_membership_cluster_local:notify_error_to_manager/3  432     {'leofs12_0@192.168.100.12',{badrpc,timeout}}

In this case,
The storage node which coudn't update its status to attached will be not shown as running but as attached after starting LeoFS cluster.

[mocchira]

There are two solutions.

1. Retry update_member_by_node
2. Reduce accesses to leo_redundant_manager(gen_server)

Applying both of them would be best.

[mocchira]

The root cause of #317 could affect this issue.