Redis/Redis-sentinel环境建立和验证---接着昨天的验证流程，再按照binding物理IP的方式进行
背景：
12.26日的验证，采用redis直接绑定127.0.0.1，导致sentinel切换/监控存在问题；
本次验证条件：
各个redis node都在配置文件中指定binding 物理IP(之前是采用默认，绑定127.0.0.1）；
环境：
192.68.0.11 ： master：6379/slave：6380/sentinel：26379
192.68.0.12 ： slave：6379
过程：
1. 正常启动各个节点，设置master/slave状态，节点显示正常；
启动sentinel，显示状态正常：
[29777] 27 Nov 10:16:01.526 * +slave slave 192.168.0.12:6379 192.168.0.12 6379 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:16:01.526 * +slave slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.11 6379
2. shutdown 11 6379master：
sentinel console显示正常，选择192.68.0.12 ： slave：6379为master：
[29777] 27 Nov 10:18:04.906 # +sdown master mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:04.906 # +odown master mymaster 192.168.0.11 6379 #quorum 1/1
[29777] 27 Nov 10:18:04.906 # +failover-triggered master mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:04.906 # +failover-state-wait-start master mymaster 192.168.0.11 6379 #starting in 12572 milliseconds
[29777] 27 Nov 10:18:17.554 # +failover-state-select-slave master mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:17.654 # +selected-slave slave 192.168.0.12:6379 192.168.0.12 6379 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:17.654 * +failover-state-send-slaveof-noone slave 192.168.0.12:6379 192.168.0.12 6379 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:17.754 * +failover-state-wait-promotion slave 192.168.0.12:6379 192.168.0.12 6379 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:17.957 # +promoted-slave slave 192.168.0.12:6379 192.168.0.12 6379 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:17.957 # +failover-state-reconf-slaves master mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:18.056 * +slave-reconf-sent slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:18.959 * +slave-reconf-inprog slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:18.960 * +slave-reconf-done slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:19.060 # +failover-end master mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:19.060 # +switch-master mymaster 192.168.0.11 6379 192.168.0.12 6379
[29777] 27 Nov 10:18:19.161 * +slave slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.12 6379
[29777] 27 Nov 10:18:49.141 # +sdown slave 192.168.0.11:6379 192.168.0.11 6379 @ mymaster 192.168.0.12 6379
查询两个节点，显示：
11的6380为12：6379的slave，正常
12的6379显示正常，一个slave：
role:master
connected_slaves:1
slave0:192.168.0.11,6380,online
sentinel masters查询显示12为主机，正常；
sentinel slaves mymaster查询：
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6380" --正常
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6380"
7) "runid"
8) "e02b196ce500fa2c4aaecb8143270ad2a290c50a"
9) "flags"
10) "slave"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "895"
15) "last-ping-reply"
16) "895"
17) "info-refresh"
18) "3302"
19) "master-link-down-time"
20) "0"
21) "master-link-status"
22) "ok"
23) "master-host"
24) "192.168.0.12"
25) "master-port"
26) "6379"
27) "slave-priority"
28) "100"
2) 1) "name"
2) "192.168.0.11:6379" ---原来11变成slave
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid" --没有了runid
8) ""
9) "flags"
10) "s_down,slave,disconnected,demote" --状态已经down
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "484937"
15) "last-ping-reply"
16) "484937"
17) "s-down-time"
18) "454856"
19) "info-refresh"
20) "1385519183997"
21) "master-link-down-time"
22) "0"
23) "master-link-status" ---下边状态都不正常
24) "err"
25) "master-host"
26) "?"
27) "master-port"
28) "0"
29) "slave-priority"
30) "100"
------shutdown 11的6380从机：
sentinel console：显示11的6380状态为下线：
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6380"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6380"
7) "runid"
8) "e02b196ce500fa2c4aaecb8143270ad2a290c50a"
9) "flags"
10) "s_down,slave,disconnected"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "52727"
15) "last-ping-reply"
16) "52727"
17) "s-down-time"
18) "22714"
19) "info-refresh"
20) "52325"
21) "master-link-down-time"
22) "0"
23) "master-link-status"
24) "ok"
25) "master-host"
26) "192.168.0.12"
27) "master-port"
28) "6379"
29) "slave-priority"
30) "100"
2) 1) "name"
2) "192.168.0.11:6379"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid"
8) ""
9) "flags"
10) "s_down,slave,disconnected,demote"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "654327"
15) "last-ping-reply"
16) "654327"
17) "s-down-time"
18) "624246"
19) "info-refresh"
20) "1385519353387"
21) "master-link-down-time"
22) "0"
23) "master-link-status"
24) "err"
25) "master-host"
26) "?"
27) "master-port"
28) "0"
29) "slave-priority"
30) "100"
12的info：显示slave已经没有了；
-----重启11的6380：
sentinel console：侦测到6380启动，但是由于默认为master，所以无法挂载：并从slave中去除；
[29777] 27 Nov 10:30:17.397 * +reboot slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.12 6379
[29777] 27 Nov 10:30:17.397 # -slave-restart-as-master slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.12 6379 #removing it from the attached slaves
sentinel slaves查询显示只有一个slave： 11：6379：
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6379"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid"
8) ""
9) "flags"
10) "s_down,slave,disconnected,demote"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "805280"
15) "last-ping-reply"
16) "805280"
17) "s-down-time"
18) "775199"
19) "info-refresh"
20) "1385519504340"
21) "master-link-down-time"
22) "0"
23) "master-link-status"
24) "err"
25) "master-host"
26) "?"
27) "master-port"
28) "0"
29) "slave-priority"
30) "100"
----重启11：6379：
11：6379info显示自己变成了slave！
12：6379master info显示自己有一个slave，是11：6379，正常；
sentinel console显示：11：6379被移除然后又被重新添加；
[29777] 27 Nov 10:32:42.236 # -sdown slave 192.168.0.11:6379 192.168.0.11 6379 @ mymaster 192.168.0.12 6379
[29777] 27 Nov 10:32:52.054 * +slave slave 192.168.0.11:6379 192.168.0.11 6379 @ mymaster 192.168.0.12 6379
sentinel slaves查询：一个slave，状态正常；
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6379"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid"
8) "2a6a7e441f40b2ce803080273dae656382a8076f"
9) "flags"
10) "slave"
11) "pending-commands"
12) "-1"
13) "last-ok-ping-reply"
14) "678"
15) "last-ping-reply"
16) "678"
17) "info-refresh"
18) "6488"
19) "master-link-down-time"
20) "0"
21) "master-link-status"
22) "ok"
23) "master-host"
24) "192.168.0.12"
25) "master-port"
26) "6379"
27) "slave-priority"
28) "100"
redis 127.0.0.1:26379>
在主上做set，验证从机：主机设置，从机可以查到，功能正常；
--------设置11：6380为12的从机：
设置成功，状态正常，从该从机可以查到刚才新设置的变量，功能正常；
sentinel console显示添加新slave：
[29777] 27 Nov 10:37:32.337 * +slave slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.12 6379
查询slaves正常：
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6380"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6380"
7) "runid"
8) "ee58ff9f4e565b04d0f6fe1b5c7877229ef8b509"
9) "flags"
10) "slave"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "332"
15) "last-ping-reply"
16) "332"
17) "info-refresh"
18) "32"
19) "master-link-down-time"
20) "0"
21) "master-link-status"
22) "ok"
23) "master-host"
24) "192.168.0.12"
25) "master-port"
26) "6379"
27) "slave-priority"
28) "100"
2) 1) "name"
2) "192.168.0.11:6379"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid"
8) "2a6a7e441f40b2ce803080273dae656382a8076f"
9) "flags"
10) "slave"
11) "pending-commands"
12) "-1"
13) "last-ok-ping-reply"
14) "332"
15) "last-ping-reply"
16) "332"
17) "info-refresh"
18) "9949"
19) "master-link-down-time"
20) "0"
21) "master-link-status"
22) "ok"
23) "master-host"
24) "192.168.0.12"
25) "master-port"
26) "6379"
27) "slave-priority"
28) "100"
redis 127.0.0.1:26379>
------------shutdown 11：6380 slave（刚添加的):
12显示只有一个从机，正常；
sentinel slave显示：两个从机，11：6380状态为down，正常：
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6380"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6380"
7) "runid"
8) "ee58ff9f4e565b04d0f6fe1b5c7877229ef8b509"
9) "flags"
10) "s_down,slave,disconnected"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "31882"
15) "last-ping-reply"
16) "31882"
17) "s-down-time"
18) "1795"
19) "info-refresh"
20) "40099"
21) "master-link-down-time"
22) "0"
23) "master-link-status"
24) "ok"
25) "master-host"
26) "192.168.0.12"
27) "master-port"
28) "6379"
29) "slave-priority"
30) "100"
2) 1) "name"
2) "192.168.0.11:6379"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid"
8) "2a6a7e441f40b2ce803080273dae656382a8076f"
9) "flags"
10) "slave"
11) "pending-commands"
12) "-1"
13) "last-ok-ping-reply"
14) "792"
15) "last-ping-reply"
16) "792"
17) "info-refresh"
18) "9922"
19) "master-link-down-time"
20) "0"
21) "master-link-status"
22) "ok"
23) "master-host"
24) "192.168.0.12"
25) "master-port"
26) "6379"
27) "slave-priority"
28) "100"
sentinel console显示从机移出：
[29777] 27 Nov 10:40:00.926 # +sdown slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.12 6379
--------shutdown 11：6379 从机：
不再直接被接管为slave，而是为master，需要slaveof；不知道为什么
----------------结论：
相比较与昨天的测试：
1. 必须要在redis节点上明确binding 物理IP，不能使用localhost/127.0.0.1，否则sentinel将会出现问题；
2. 从机down/up正常，但是up后需要明确执行slaveof；
3. 主机down后，可以被正常接管，其余从机也正确转移；但是原有主机up后，马上就被接管为slave，和上述从机不同，不知到什么原因；再次shutdown该节点，行为和一般从机一样，先是master，明确slaveof以后正常；
结论：sentinel功能基本可用，可以完成master-slave切换，但是原有从机下线后，需要手工操作并做slaveof指定！
12.26日的验证，采用redis直接绑定127.0.0.1，导致sentinel切换/监控存在问题；
本次验证条件：
各个redis node都在配置文件中指定binding 物理IP(之前是采用默认，绑定127.0.0.1）；
环境：
192.68.0.11 ： master：6379/slave：6380/sentinel：26379
192.68.0.12 ： slave：6379
过程：
1. 正常启动各个节点，设置master/slave状态，节点显示正常；
启动sentinel，显示状态正常：
[29777] 27 Nov 10:16:01.526 * +slave slave 192.168.0.12:6379 192.168.0.12 6379 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:16:01.526 * +slave slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.11 6379
2. shutdown 11 6379master：
sentinel console显示正常，选择192.68.0.12 ： slave：6379为master：
[29777] 27 Nov 10:18:04.906 # +sdown master mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:04.906 # +odown master mymaster 192.168.0.11 6379 #quorum 1/1
[29777] 27 Nov 10:18:04.906 # +failover-triggered master mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:04.906 # +failover-state-wait-start master mymaster 192.168.0.11 6379 #starting in 12572 milliseconds
[29777] 27 Nov 10:18:17.554 # +failover-state-select-slave master mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:17.654 # +selected-slave slave 192.168.0.12:6379 192.168.0.12 6379 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:17.654 * +failover-state-send-slaveof-noone slave 192.168.0.12:6379 192.168.0.12 6379 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:17.754 * +failover-state-wait-promotion slave 192.168.0.12:6379 192.168.0.12 6379 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:17.957 # +promoted-slave slave 192.168.0.12:6379 192.168.0.12 6379 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:17.957 # +failover-state-reconf-slaves master mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:18.056 * +slave-reconf-sent slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:18.959 * +slave-reconf-inprog slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:18.960 * +slave-reconf-done slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:19.060 # +failover-end master mymaster 192.168.0.11 6379
[29777] 27 Nov 10:18:19.060 # +switch-master mymaster 192.168.0.11 6379 192.168.0.12 6379
[29777] 27 Nov 10:18:19.161 * +slave slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.12 6379
[29777] 27 Nov 10:18:49.141 # +sdown slave 192.168.0.11:6379 192.168.0.11 6379 @ mymaster 192.168.0.12 6379
查询两个节点，显示：
11的6380为12：6379的slave，正常
12的6379显示正常，一个slave：
role:master
connected_slaves:1
slave0:192.168.0.11,6380,online
sentinel masters查询显示12为主机，正常；
sentinel slaves mymaster查询：
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6380" --正常
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6380"
7) "runid"
8) "e02b196ce500fa2c4aaecb8143270ad2a290c50a"
9) "flags"
10) "slave"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "895"
15) "last-ping-reply"
16) "895"
17) "info-refresh"
18) "3302"
19) "master-link-down-time"
20) "0"
21) "master-link-status"
22) "ok"
23) "master-host"
24) "192.168.0.12"
25) "master-port"
26) "6379"
27) "slave-priority"
28) "100"
2) 1) "name"
2) "192.168.0.11:6379" ---原来11变成slave
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid" --没有了runid
8) ""
9) "flags"
10) "s_down,slave,disconnected,demote" --状态已经down
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "484937"
15) "last-ping-reply"
16) "484937"
17) "s-down-time"
18) "454856"
19) "info-refresh"
20) "1385519183997"
21) "master-link-down-time"
22) "0"
23) "master-link-status" ---下边状态都不正常
24) "err"
25) "master-host"
26) "?"
27) "master-port"
28) "0"
29) "slave-priority"
30) "100"
------shutdown 11的6380从机：
sentinel console：显示11的6380状态为下线：
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6380"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6380"
7) "runid"
8) "e02b196ce500fa2c4aaecb8143270ad2a290c50a"
9) "flags"
10) "s_down,slave,disconnected"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "52727"
15) "last-ping-reply"
16) "52727"
17) "s-down-time"
18) "22714"
19) "info-refresh"
20) "52325"
21) "master-link-down-time"
22) "0"
23) "master-link-status"
24) "ok"
25) "master-host"
26) "192.168.0.12"
27) "master-port"
28) "6379"
29) "slave-priority"
30) "100"
2) 1) "name"
2) "192.168.0.11:6379"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid"
8) ""
9) "flags"
10) "s_down,slave,disconnected,demote"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "654327"
15) "last-ping-reply"
16) "654327"
17) "s-down-time"
18) "624246"
19) "info-refresh"
20) "1385519353387"
21) "master-link-down-time"
22) "0"
23) "master-link-status"
24) "err"
25) "master-host"
26) "?"
27) "master-port"
28) "0"
29) "slave-priority"
30) "100"
12的info：显示slave已经没有了；
-----重启11的6380：
sentinel console：侦测到6380启动，但是由于默认为master，所以无法挂载：并从slave中去除；
[29777] 27 Nov 10:30:17.397 * +reboot slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.12 6379
[29777] 27 Nov 10:30:17.397 # -slave-restart-as-master slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.12 6379 #removing it from the attached slaves
sentinel slaves查询显示只有一个slave： 11：6379：
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6379"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid"
8) ""
9) "flags"
10) "s_down,slave,disconnected,demote"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "805280"
15) "last-ping-reply"
16) "805280"
17) "s-down-time"
18) "775199"
19) "info-refresh"
20) "1385519504340"
21) "master-link-down-time"
22) "0"
23) "master-link-status"
24) "err"
25) "master-host"
26) "?"
27) "master-port"
28) "0"
29) "slave-priority"
30) "100"
----重启11：6379：
11：6379info显示自己变成了slave！
12：6379master info显示自己有一个slave，是11：6379，正常；
sentinel console显示：11：6379被移除然后又被重新添加；
[29777] 27 Nov 10:32:42.236 # -sdown slave 192.168.0.11:6379 192.168.0.11 6379 @ mymaster 192.168.0.12 6379
[29777] 27 Nov 10:32:52.054 * +slave slave 192.168.0.11:6379 192.168.0.11 6379 @ mymaster 192.168.0.12 6379
sentinel slaves查询：一个slave，状态正常；
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6379"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid"
8) "2a6a7e441f40b2ce803080273dae656382a8076f"
9) "flags"
10) "slave"
11) "pending-commands"
12) "-1"
13) "last-ok-ping-reply"
14) "678"
15) "last-ping-reply"
16) "678"
17) "info-refresh"
18) "6488"
19) "master-link-down-time"
20) "0"
21) "master-link-status"
22) "ok"
23) "master-host"
24) "192.168.0.12"
25) "master-port"
26) "6379"
27) "slave-priority"
28) "100"
redis 127.0.0.1:26379>
在主上做set，验证从机：主机设置，从机可以查到，功能正常；
--------设置11：6380为12的从机：
设置成功，状态正常，从该从机可以查到刚才新设置的变量，功能正常；
sentinel console显示添加新slave：
[29777] 27 Nov 10:37:32.337 * +slave slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.12 6379
查询slaves正常：
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6380"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6380"
7) "runid"
8) "ee58ff9f4e565b04d0f6fe1b5c7877229ef8b509"
9) "flags"
10) "slave"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "332"
15) "last-ping-reply"
16) "332"
17) "info-refresh"
18) "32"
19) "master-link-down-time"
20) "0"
21) "master-link-status"
22) "ok"
23) "master-host"
24) "192.168.0.12"
25) "master-port"
26) "6379"
27) "slave-priority"
28) "100"
2) 1) "name"
2) "192.168.0.11:6379"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid"
8) "2a6a7e441f40b2ce803080273dae656382a8076f"
9) "flags"
10) "slave"
11) "pending-commands"
12) "-1"
13) "last-ok-ping-reply"
14) "332"
15) "last-ping-reply"
16) "332"
17) "info-refresh"
18) "9949"
19) "master-link-down-time"
20) "0"
21) "master-link-status"
22) "ok"
23) "master-host"
24) "192.168.0.12"
25) "master-port"
26) "6379"
27) "slave-priority"
28) "100"
redis 127.0.0.1:26379>
------------shutdown 11：6380 slave（刚添加的):
12显示只有一个从机，正常；
sentinel slave显示：两个从机，11：6380状态为down，正常：
redis 127.0.0.1:26379> sentinel slaves mymaster
1) 1) "name"
2) "192.168.0.11:6380"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6380"
7) "runid"
8) "ee58ff9f4e565b04d0f6fe1b5c7877229ef8b509"
9) "flags"
10) "s_down,slave,disconnected"
11) "pending-commands"
12) "0"
13) "last-ok-ping-reply"
14) "31882"
15) "last-ping-reply"
16) "31882"
17) "s-down-time"
18) "1795"
19) "info-refresh"
20) "40099"
21) "master-link-down-time"
22) "0"
23) "master-link-status"
24) "ok"
25) "master-host"
26) "192.168.0.12"
27) "master-port"
28) "6379"
29) "slave-priority"
30) "100"
2) 1) "name"
2) "192.168.0.11:6379"
3) "ip"
4) "192.168.0.11"
5) "port"
6) "6379"
7) "runid"
8) "2a6a7e441f40b2ce803080273dae656382a8076f"
9) "flags"
10) "slave"
11) "pending-commands"
12) "-1"
13) "last-ok-ping-reply"
14) "792"
15) "last-ping-reply"
16) "792"
17) "info-refresh"
18) "9922"
19) "master-link-down-time"
20) "0"
21) "master-link-status"
22) "ok"
23) "master-host"
24) "192.168.0.12"
25) "master-port"
26) "6379"
27) "slave-priority"
28) "100"
sentinel console显示从机移出：
[29777] 27 Nov 10:40:00.926 # +sdown slave 192.168.0.11:6380 192.168.0.11 6380 @ mymaster 192.168.0.12 6379
--------shutdown 11：6379 从机：
不再直接被接管为slave，而是为master，需要slaveof；不知道为什么
----------------结论：
相比较与昨天的测试：
1. 必须要在redis节点上明确binding 物理IP，不能使用localhost/127.0.0.1，否则sentinel将会出现问题；
2. 从机down/up正常，但是up后需要明确执行slaveof；
3. 主机down后，可以被正常接管，其余从机也正确转移；但是原有主机up后，马上就被接管为slave，和上述从机不同，不知到什么原因；再次shutdown该节点，行为和一般从机一样，先是master，明确slaveof以后正常；
结论：sentinel功能基本可用，可以完成master-slave切换，但是原有从机下线后，需要手工操作并做slaveof指定！