redis 5.0 cluster install

cxm.cm

1、下載部署
wget  -b http://download.redis.io/releases/redis-5.0.5.tar.gz
tar xzf redis-5.0.5.tar.gz
cd redis-5.0.5   
make && make install 編譯安裝
2、集群創(chuàng)建
Redis 集群最小配置為 3主 3 從，先進(jìn)行簡(jiǎn)單配置文件設(shè)置

cd  /data/redis
mkdir  conf logs 7000 7001
修改配置文件，主要修改以下參數(shù)，詳細(xì)配置文件參考附件

cluster-enabled yes
port 7000
cluster-config-file nodes-7000.conf
pidfile /data/redis/redis_7000.pid
dbfilename 7000-dump.rdb
dir /data/redis/7000
logfile /data/redis/logs/redis-7000.log
include ./redis-conf
使用命令快速創(chuàng)建7001文件 sed 's/7000/7001/g' 7000.conf > 7001.conf
啟動(dòng)單機(jī)服務(wù)

./redis-server /data/redis/conf/7000.conf
140688:C 08 Nov 2019 16:03:44.932 # oO0OoO0OoO0Oo Redis is starting oO0OoO0OoO0Oo
140688:C 08 Nov 2019 16:03:44.932 # Redis version=5.0.5, bits=64, commit=00000000, modified=0, pid=140688, just started
140688:C 08 Nov 2019 16:03:44.932 # Configuration loaded
p此時(shí)在/data/redis/6379 文件夾下將看到如下文件

-rw-r--r-- 1 root root   114 Nov  8 15:37 nodes-7000.conf
-rw-r--r-- 1 root root     6 Nov  8 15:37 redis_7000.pid
-rw-r--r-- 1 root root 10275 Nov  8 15:36 redis.conf
參照如上配置分別啟動(dòng)7000 7001 7002 7003 7004 7005六個(gè)節(jié)點(diǎn)
使用命令進(jìn)行集群創(chuàng)建

./redis-cli --cluster create 192.168.70.4:7000 192.168.70.4:7001 192.168.41.14:7002 192.168.41.14:7003 192.168.41.18:7004  192.168.41.18:7005  --cluster-replicas 1
--cluster-replicas 1 命令的意思： 一主一從配置，六個(gè)節(jié)點(diǎn)就是 三主三從

[root@bjxg-elk2-kafka-02 src]# ./redis-cli --cluster create 192.168.70.4:7000 192.168.70.4:7001 192.168.41.14:7002 192.168.41.14:7003 192.168.41.18:7004  192.168.41.18:7005  --cluster-replicas 1
>>> Performing hash slots allocation on 6 nodes...
Master[0] -> Slots 0 - 5460
Master[1] -> Slots 5461 - 10922
Master[2] -> Slots 10923 - 16383
Adding replica 192.168.41.14:7003 to 192.168.70.4:7000
Adding replica 192.168.41.18:7005 to 192.168.41.14:7002
Adding replica 192.168.70.4:7001 to 192.168.41.18:7004
M: 0f96b6e007c27255d90f50737073c34741e7aa32 192.168.70.4:7000
   slots:[0-5460] (5461 slots) master
S: 4b0d525711e0704266e78acb15c3053c94dc9dbd 192.168.70.4:7001
   replicates fbadfccf8e5228d35afa93a17a7e8aef70a3da43
M: fccbefd52d4bb9ea348a9601d83844a81d3d59ee 192.168.41.14:7002
   slots:[5461-10922] (5462 slots) master
S: ae2cb8a7821d16fe3770f71ffcc47522d72a24d2 192.168.41.14:7003
   replicates 0f96b6e007c27255d90f50737073c34741e7aa32
M: fbadfccf8e5228d35afa93a17a7e8aef70a3da43 192.168.41.18:7004
   slots:[10923-16383] (5461 slots) master
S: a3b9b3edd2a6c69666d2ca7964a1186b2da063a4 192.168.41.18:7005
   replicates fccbefd52d4bb9ea348a9601d83844a81d3d59ee
Can I set the **ve configuration? (type 'yes' to accept): yes
>>> Nodes configuration updated
>>> Assign a different config epoch to each node
>>> Sending CLUSTER MEET messages to join the cluster
Waiting for the cluster to join
...
>>> Performing Cluster Check (using node 192.168.70.4:7000)
M: 0f96b6e007c27255d90f50737073c34741e7aa32 192.168.70.4:7000
   slots:[0-5460] (5461 slots) master
   1 additional replica(s)
S: a3b9b3edd2a6c69666d2ca7964a1186b2da063a4 192.168.41.18:7005
   slots: (0 slots) slave
   replicates fccbefd52d4bb9ea348a9601d83844a81d3d59ee
S: 4b0d525711e0704266e78acb15c3053c94dc9dbd 192.168.70.4:7001
   slots: (0 slots) slave
   replicates fbadfccf8e5228d35afa93a17a7e8aef70a3da43
M: fccbefd52d4bb9ea348a9601d83844a81d3d59ee 192.168.41.14:7002
   slots:[5461-10922] (5462 slots) master
   1 additional replica(s)
S: ae2cb8a7821d16fe3770f71ffcc47522d72a24d2 192.168.41.14:7003
   slots: (0 slots) slave
   replicates 0f96b6e007c27255d90f50737073c34741e7aa32
M: fbadfccf8e5228d35afa93a17a7e8aef70a3da43 192.168.41.18:7004
   slots:[10923-16383] (5461 slots) master
   1 additional replica(s)
[OK] All nodes agree **ut slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All 16384 slots covered.
檢查集群狀態(tài)
redis-cli --cluster check 192.168.41.14:7002
[root@bjxg-elk2-kafka-02 src]# ./redis-cli --cluster check 192.168.41.14:7002
192.168.41.14:7002 (fccbefd5...) -> 0 keys | 5462 slots | 1 slaves.
192.168.41.18:7004 (fbadfccf...) -> 0 keys | 5461 slots | 1 slaves.
192.168.70.4:7000 (0f96b6e0...) -> 0 keys | 5461 slots | 1 slaves.
[OK] 0 keys in 3 masters.
0.00 keys per slot on average.
>>> Performing Cluster Check (using node 192.168.41.14:7002)
M: fccbefd52d4bb9ea348a9601d83844a81d3d59ee 192.168.41.14:7002
   slots:[5461-10922] (5462 slots) master
   1 additional replica(s)
M: fbadfccf8e5228d35afa93a17a7e8aef70a3da43 192.168.41.18:7004
   slots:[10923-16383] (5461 slots) master
   1 additional replica(s)
S: 4b0d525711e0704266e78acb15c3053c94dc9dbd 192.168.70.4:7001
   slots: (0 slots) slave
   replicates fbadfccf8e5228d35afa93a17a7e8aef70a3da43
S: a3b9b3edd2a6c69666d2ca7964a1186b2da063a4 192.168.41.18:7005
   slots: (0 slots) slave
   replicates fccbefd52d4bb9ea348a9601d83844a81d3d59ee
S: ae2cb8a7821d16fe3770f71ffcc47522d72a24d2 192.168.41.14:7003
   slots: (0 slots) slave
   replicates 0f96b6e007c27255d90f50737073c34741e7aa32
M: 0f96b6e007c27255d90f50737073c34741e7aa32 192.168.70.4:7000
   slots:[0-5460] (5461 slots) master
   1 additional replica(s)
[OK] All nodes agree **ut slots configuration.
>>> Check for open slots...
>>> Check slots coverage...
[OK] All 16384 slots covered.
測(cè)試集群狀態(tài)

redis-cli -c -h  192.168.41.14 -p 7002
[root@bjxg-elk2-kafka-02 src]# ./redis-cli -c -h  192.168.41.14 -p 7002
192.168.41.14:7002> cluster info
cluster_statek
cluster_slots_assigned:16384
cluster_slots_ok:16384
cluster_slots_pfail:0
cluster_slots_fail:0
cluster_known_nodes:6
cluster_size:3
cluster_current_epoch:6
cluster_my_epoch:3
cluster_stats_messages_ping_sent:213
cluster_stats_messages_pong_sent:231
cluster_stats_messages_meet_sent:5
cluster_stats_messages_sent:449
cluster_stats_messages_ping_received:231
cluster_stats_messages_pong_received:218
cluster_stats_messages_received:449
查看幫助命令

./redis-cli --help

測(cè)試性能

redis-benchmark -h192.168.41.14 -p 7002 -c 100 -n 100000
100個(gè)并發(fā)連接，100000個(gè)請(qǐng)求，檢測(cè)host為localhost 端口為7002的redis服務(wù)器性能
3、遇到的問(wèn)題
1、WARNING overcommit_memory is set to 0! Background save may fail
under low memory condition. To fix this issue add
'vm.overcommit_memory = 1' to /etc/sysctl.conf and then reboot or run
the command 'sysctl vm.overcommit_memory=1' for this to take effect.

解決：
如果“/proc/sys/vm/overcommit_memory”的值為0，則會(huì)表示開(kāi)啟了OOM。可以設(shè)置為1關(guān)閉OOM
需要在文件/etc/sysctl.conf中增加一行 vm.overcommit_memory = 1，然后執(zhí)行命令“sysctl -p”以生效。

2、WARNING you have Transparent Huge Pages (THP) support enabled in
your kernel. This will create latency and memory usage issues with
Redis. To fix this issue run the command 'echo never >
/sys/kernel/mm/transparent_hugepage/enabled' as root, and add it to
your /etc/rc.local in order to retain the setting after a reboot.
Redis must be restarted after THP is disabled.

解決：
你使用的是透明大頁(yè)，可能導(dǎo)致redis延遲和內(nèi)存使用問(wèn)題。執(zhí)行echo never > /sys/kernel/mm/transparent_hugepage/enabled修復(fù)該問(wèn)題。
永久解決方法：
將其寫入/etc/rc.local文件中
echo never > /sys/kernel/mm/redhat_transparent_hugepage/enabled

3、TCP監(jiān)聽(tīng)隊(duì)列大小
即TCP listen的backlog大小，“/proc/sys/net/core/somaxconn”的默認(rèn)值一般較小如128，需要修改大一點(diǎn)，比如改成32767。立即生效還可以使用命令：sysctl -w net.core.somaxconn=32767。
要想永久生效，需要在文件/etc/sysctl.conf中增加一行：net.core.somaxconn = 32767，然后執(zhí)行命令“sysctl -p”以生效。
Redis配置項(xiàng)tcp-backlog的值不能超過(guò)somaxconn的大小。