CentOS 7 Redis 설치와 설정

Redis 구성

image-20210528165507979

Redis 사전 작업

1. 시스템 업데이트

# sudo yum -y update

2. 패키지 설치

# sudo yum -y install gcc-c++

# sudo yum install wget

# sudo yum install tcl tk

3. 메모리 설정

org.springframework.data.redis.RedisSystemException: 
Error in execution; nested exception is io.lettuce.core.RedisCommandExecutionException: MISCONF Redis is configured to save RDB snapshots,
 but it is currently not able to persist on disk. Commands that may modify the data set are disabled, 
 because this instance is configured to report errors during writes if RDB snapshotting fails (stop-writes-on-bgsave-error option).
  Please check the Redis logs for details about the RDB error.

Redis는 프로세스가 장애로 인해 종료되더라도 이전의 상태를 동일하게 복구하기 위해서 자동으로 .rdb 라는 확장자의 파일에 인메모리 데이터를 저장합니다. rdb 파일을 저장하려 할때 Redis는 하위 프로세스를 포킹하여 데이터를 디스크에 저장합니다. 이때 포크를 제대로 하지 못해 발생할 수 있습니다. 이를 해결 하기 위해 아래와 같은 설정을 합니다.

메모리 사용량이 허용량을 초과할 경우, overcommit을 처리하는 방식 결정하는 값을 “항상”으로 변경하고, 기본 값은 “0” 입니다.

0 : 커널 기본값, Heuristic 하게 Overcommit을 허용 (Page Cache + Swap Memory + Slab Reclaimable 값이 요청한 메모리 수 보다 클 경우 허용) 1 : 항상 Overcommit을 허용 2 : 제한적 Overcommit 허용

적용

# sudo sysctl vm.overcommit_memory=1

# sudo echo “vm.overcommit_memory=1” » /etc/sysctl.conf

적용 확인

# sudo sysctl -a grep vm.overcommit

sysctl: reading key “net.ipv6.conf.all.stable_secret” sysctl: reading key “net.ipv6.conf.default.stable_secret” sysctl: reading key “net.ipv6.conf.ens32.stable_secret” sysctl: reading key “net.ipv6.conf.lo.stable_secret” vm.overcommit_kbytes = 0 vm.overcommit_memory = 1 vm.overcommit_ratio = 50

단순히 이 오류 메시지를 무시하려면 /etc/redis.conf을 열어서 아래와 같이 설정합니다.

#vim /etc/sysctl.conf

stop-writes-on-bgsave-error no

4. TCP BackLog 설정

WARNING: The TCP backlog setting of 511 cannot be enforced because /proc/sys/net/core/somaxconn is set to the lower value of 128.

기본적인 소켓 accept limit 값입니다. 보통 리눅스 배포판에는 128로 되어있고 이 때문에 tcp backlog 의 셋팅이 511 로 되어있지만 강제로 128로 적용 된다는 뜻입니다.서버 소켓에 Accept를 대기하는 소켓 개수 파라미터를 변경해줍니다. 기본 Accept limit은 128이고 1024 이상으로 변경합니다. (최대 65535)

설정

# sudo sysctl -w net.core.somaxconn=1024

# sudo echo “net.core.somaxconn=1024” » /etc/sysctl.conf

확인

# sudo sysctl -a grep somaxconn

net.core.somaxconn = 1024 sysctl: reading key “net.ipv6.conf.all.stable_secret” sysctl: reading key “net.ipv6.conf.default.stable_secret” sysctl: reading key “net.ipv6.conf.ens32.stable_secret” sysctl: reading key “net.ipv6.conf.lo.stable_secret”

5. THP 설정

THP(Transparent Huge Pages) 기능이 Enable 되어 있는 경우 Redis에서는 이를 Disable 시킬 것을 권장합니다. 재부팅 시 설정을 변경하기 위해 /etc/rc.local 에도 명령어를 추가해 줍니다.

# echo 'never' >/sys/kernel/mm/transparent_hugepage/enabled

재부팅시 적용을 위해 rc.local에도 적용합니다.

# vim /etc/rc.local
//맨 아래줄에 추가합니다.
echo 'never' >/sys/kernel/mm/transparent_hugepage/enabled
exit 0

# reboot

Redis 설치

rpm으로 설치한게 있다면 삭제

#sudo yum remove redis

Redis 최신 버전 다운로드

#wget https://download.redis.io/releases/redis-6.2.3.tar.gz

압축 풀기

#tar -xvf tar -xvf redis-6.2.3.tar.gz

폴더명 변경

#mv redis-6.2.3 ./redis

컴파일

# cd redis

# ls

00-RELEASENOTES CONTRIBUTING MANIFESTO TLS.md runtest runtest-sentinel tests

BUGS COPYING Makefile deps runtest-cluster sentinel.conf utils

CONDUCT INSTALL README.md redis.conf runtest-moduleapi src

# sudo make

cd src && make all

make[1]: Entering directory /usr/local/src/redis/src

​ CC Makefile.dep

make[1]: Leaving directory /usr/local/src/redis/src

make[1]: Entering directory /usr/local/src/redis/src

…

jemalloc version : 5.1.0-0-g0 library revision : 2

…

Hint: It’s a good idea to run ‘make test’ ;)

Redis 설치 (redis 실행파일이 /usr/local/bin로 복사됨.)

# sudo make install

CC Makefile.de

Hint: It’s a good idea to run ‘make test’ ;)

INSTALL redis-server

INSTALL redis-benchmark

INSTALL redis-cli

/usr/local/bin 확인

# ls /usr/local/bin grep redis

redis-benchmark

redis-check-aof

redis-check-rdb

redis-cli

redis-sentinel

redis-server

Redis Master-Slave 구성

# cd /usr/local/src/redis/utils

# ./install_server.sh

Welcome to the redis service installer

This script will help you easily set up a running redis server

This systems seems to use systemd.

Please take a look at the provided example service unit files in this directory, and adapt and install them. Sorry

# vim install_server.sh

// 해당 라인 주석처리.

image-20210526142149545

Master Redis 서버 설치

#./install_server.sh

Welcome to the redis service installer This script will help you easily set up a running redis server

Please select the redis port for this instance: [6379] 6379 // Redis 실행 포트설정

Please select the redis config file name [/etc/redis/6379.conf] /usr/local/src/redis/conf/6379.conf // 설정파일 위치 지정

Please select the redis log file name [/var/log/redis_6379.log] /usr/local/src/redis/logs/6379.log // 로그저장 위치 지정

Please select the data directory for this instance [/var/lib/redis/6379] /usr/local/src/redis/6379 // 데이터 디렉터리 위치 지정

Please select the redis executable path [/usr/local/bin/redis-server] //엔터, 실행 위치는 디폴트값 사용

Selected config:

Port : 6379

Config file : /usr/local/src/redis/conf/6379.conf

Log file : /usr/local/src/redis/logs/6379.log

Data dir : /usr/local/src/redis/6379

Executable : /usr/local/bin/redis-server

Cli Executable : /usr/local/bin/redis-cli

Is this ok? Then press ENTER to go on or Ctrl-C to abort. //엔터

Copied /tmp/6379.conf => /etc/init.d/redis_6379

Installing service…

Successfully added to chkconfig!

Successfully added to runlevels 345!

Starting Redis server…

Installation successful!

Slave Redis #1 서버 설치

# ./install_server.sh

Welcome to the redis service installer

This script will help you easily set up a running redis server

Please select the redis port for this instance: [6379] 6380

Please select the redis config file name [/etc/redis/6380.conf] /usr/local/src/redis/conf/6380.conf

Please select the redis log file name [/var/log/redis_6380.log] /usr/local/src/redis/logs/6380.log

Please select the data directory for this instance [/var/lib/redis/6380] /usr/local/src/redis/6380

Please select the redis executable path [/usr/local/bin/redis-server]

Selected config:

Port : 6380

Config file : /usr/local/src/redis/conf/6380.conf

Log file : /usr/local/src/redis/logs/6380.log

Data dir : /usr/local/src/redis/6380

Executable : /usr/local/bin/redis-server

Cli Executable : /usr/local/bin/redis-cli

Is this ok? Then press ENTER to go on or Ctrl-C to abort.

Copied /tmp/6380.conf => /etc/init.d/redis_6380

Installing service…

Successfully added to chkconfig!

Successfully added to runlevels 345!

Starting Redis server…

Installation successful!

Slave Redis #2 서버 설치

# ./install_server.sh

Welcome to the redis service installer

This script will help you easily set up a running redis server

Please select the redis port for this instance: [6379] 6381

Please select the redis config file name [/etc/redis/6381.conf] /usr/local/src/redis/conf/6381.conf

Please select the redis log file name [/var/log/redis_6381.log] /usr/local/src/redis/logs/6381.log

Please select the data directory for this instance [/var/lib/redis/6381] /usr/local/src/redis/6381

Please select the redis executable path [/usr/local/bin/redis-server]

Selected config:

Port : 6381

Config file : /usr/local/src/redis/conf/6381.conf

Log file : /usr/local/src/redis/logs/6381.log

Data dir : /usr/local/src/redis/6381

Executable : /usr/local/bin/redis-server

Cli Executable : /usr/local/bin/redis-cli

Is this ok? Then press ENTER to go on or Ctrl-C to abort.

Copied /tmp/6381.conf => /etc/init.d/redis_6381

Installing service…

Successfully added to chkconfig!

Successfully added to runlevels 345!

Starting Redis server…

Installation successful!

Redis 서버 확인

# ps aux grep redis

root 13717 0.1 0.1 165060 9852 ? Ssl 17:53 0:00 /usr/local/bin/redis-server 0.0.0.0:6379

root 13727 0.1 0.1 240840 9800 ? Ssl 17:54 0:00 /usr/local/bin/redis-server 0.0.0.0:6380

root 13737 0.1 0.1 240840 9980 ? Ssl 17:54 0:00 /usr/local/bin/redis-server 0.0.0.0:6381

Redis 데몬 복사

#/usr/bin/cp -f /tmp/6379.conf /etc/init.d/redis_6379

#/usr/bin/cp -f /tmp/6380.conf /etc/init.d/redis_6380

#/usr/bin/cp -f /tmp/6381.conf /etc/init.d/redis_6381

#ls /etc/init.d

README functions netconsole network redis_6379 redis_6380 redis_6381

Redis 데몬 중지

#sudo /etc/init.d/redis_6379 stop

#sudo /etc/init.d/redis_6380 stop

#sudo /etc/init.d/redis_6381 stop

중지 확인

#ps aux grep redis

root 17163 0.0 0.0 116972 1016 pts/0 S+ 15:40 0:00 grep –color=auto redis

Redis 인스턴스별 상태, 실행중이지 않으면 /var/run/redis_6381.pid가 생성되지 않기 때문에 아래와 같이 출력된다.

#sudo /etc/init.d/redis_6381 status

cat: /var/run/redis_6381.pid: 그런 파일이나 디렉터리가 없습니다

Redis is running ()

실행 중이면 /var/run/redis_6381.pid가 생성되어 정상 출력된다.

#sudo /etc/init.d/redis_6381 start

Starting Redis server…

#sudo /etc/init.d/redis_6381 status

Redis is running (17310)

#sudo /etc/init.d/redis_6381 stop

사용하는 사용자 그룹이 있다면 redis 하위폴더 포함 owner를 사용자계정으로 변경

#chown -R 사용자:그룹 redis

Master 파일설정

# vim /usr/local/src/redis/conf/6379.conf

bind 0.0.0.0

daemonize yes

requirepass test1234

masterauth test1234

//로그 레벨

//debug (많은 정보, 개발/개발에 유용)

//verbose(Debug 수준 같은 엉망진창은 아니지만 거의 유용하지 않은 많은 정보)

//notice (자세한 내용, 제작 시 원하는 내용)

//warning(매우 중요한/중요한 메시지만 기록됨)

loglevel notice

logfile /usr/local/src/redis/logs/6379.log

slave #1 파일설정

# vim /usr/local/src/redis/conf/6380.conf

bind 0.0.0.0

daemonize yes

requirepass test1234

replicaof 211.xxx.xxx.xxx 6379

masterauth test1234

repl-ping-slave-period 10 //서버 동기화 주기, 주석제거 repl-timeout 60 주석제거

slave #2 파일설정

# vim /usr/local/src/redis/conf/6381.conf

bind 0.0.0.0

daemonize yes

replicaof 211.xxx.xxx.xxx 6379

masterauth test1234

repl-ping-slave-period 10 //서버 동기화 주기, 주석제거

repl-timeout 60 주석제거

Redis 데몬 패스워드 설정

설정안하면 데몬 stop시 Waiting for Redis to shutdown … 발생하기 때문에 아래 파일을 편집하여 패스워드를 설정합니다.

# vim /etc/init.d/redis_6379

# vim /etc/init.d/redis_6380

# vim /etc/init.d/redis_6381

#!/bin/sh

#Configurations injected by install_server below….

EXEC=/usr/local/bin/redis-server

CLIEXEC=/usr/local/bin/redis-cli

PIDFILE=/var/run/redis_6381.pid

CONF=”/usr/local/src/redis/conf/6381.conf”

REDISPORT=”6381”

REDISPW=test1234 //추가

…

stop)

​ if [ ! -f $PIDFILE ]

​ then

​ echo “$PIDFILE does not exist, process is not running” ​ else

​ PID=$(cat $PIDFILE)

​ echo “Stopping …”

​ $CLIEXEC -p $REDISPORT -a $REDISPW shutdown //이부분에 -a $REDISPW추가

….

:wq

실행

# sudo /etc/init.d/redis_6379 start

# sudo /etc/init.d/redis_6380 start

# sudo /etc/init.d/redis_6381 start

실행 확인

# ps aux grep redis

root 8288 0.9 0.1 165060 9908 ? Ssl 11:00 0:05 /usr/local/bin/redis-server 0.0.0.0:6379

root 8532 0.1 0.1 240840 10068 ? Ssl 11:08 0:00 /usr/local/bin/redis-server 0.0.0.0:6380

root 8562 0.1 0.1 240840 10072 ? Ssl 11:08 0:00 /usr/local/bin/redis-server 0.0.0.0:6381

root 8598 0.0 0.0 116972 1020 pts/1 S+ 11:10 0:00 grep –color=auto redis

마스터 로그 확인

# vim /usr/local/src/redis/logs/6379.log

13717:M 27 May 2021 17:54:22.972 * Replication backlog created, my new replication IDs are ‘818b4519e78e443ee4f9bde17b13496e22b352dc’ and

‘0000000000000000000000000000000000000000’

13717:M 27 May 2021 17:54:22.972 * Starting BGSAVE for SYNC with target: disk

13717:M 27 May 2021 17:54:22.973 * Background saving started by pid 13732

13732:C 27 May 2021 17:54:22.976 * DB saved on disk

13732:C 27 May 2021 17:54:22.977 * RDB: 0 MB of memory used by copy-on-write

13717:M 27 May 2021 17:54:23.032 * Background saving terminated with success

13717:M 27 May 2021 17:54:23.032 * Synchronization with replica 211.xxx.xxx.xxx:6380 succeeded

13717:M 27 May 2021 17:54:42.335 * Replica 211.xxx.xxx.xxx:6381 asks for synchronization

13717:M 27 May 2021 17:54:42.335 * Partial resynchronization not accepted: Replication ID mismatch (Replica asked for

‘c6ba9a5157368f7d2c7b1771b887c8c2dcee9974’, my replication IDs are ‘818b4519e78e443ee4f9bde17b13496e22b352dc’ and

‘0000000000000000000000000000000000000000’)

13717:M 27 May 2021 17:54:42.335 * Starting BGSAVE for SYNC with target: disk

13717:M 27 May 2021 17:54:42.335 * Background saving started by pid 13742

13742:C 27 May 2021 17:54:42.338 * DB saved on disk

13742:C 27 May 2021 17:54:42.339 * RDB: 0 MB of memory used by copy-on-write

13717:M 27 May 2021 17:54:42.387 * Background saving terminated with success

13717:M 27 May 2021 17:54:42.387 * Synchronization with replica 211.xxx.xxx.xxx:6381 succeeded

Slave 로그 확인

# vim /usr/local/src/redis/logs/6380.log

# vim /usr/local/src/redis/logs/6381.log

88532:C 27 May 2021 11:08:13.547 # oO0OoO0OoO0Oo Redis is starting oO0OoO0OoO0Oo

8532:C 27 May 2021 11:08:13.547 # Redis version=6.2.3, bits=64, commit=00000000, modified=0, pid=8532, just started

8532:C 27 May 2021 11:08:13.547 # Configuration loaded

8532:S 27 May 2021 11:08:13.548 * Increased maximum number of open files to 10032 (it was originally set to 1024).

8532:S 27 May 2021 11:08:13.548 * monotonic clock: POSIX clock_gettime

8532:S 27 May 2021 11:08:13.549 * Running mode=standalone, port=6380.

8532:S 27 May 2021 11:08:13.549 # Server initialized

8532:S 27 May 2021 11:08:13.549 * Loading RDB produced by version 6.2.3

8532:S 27 May 2021 11:08:13.549 * RDB age 430 seconds

8532:S 27 May 2021 11:08:13.549 * RDB memory usage when created 0.83 Mb

8532:S 27 May 2021 11:08:13.550 * DB loaded from disk: 0.000 seconds

8532:S 27 May 2021 11:08:13.550 * Ready to accept connections

8532:S 27 May 2021 11:08:13.551 * Connecting to MASTER 211.xxx.xxx.xxx:6379

8532:S 27 May 2021 11:08:13.551 * MASTER <-> REPLICA sync started

8532:S 27 May 2021 11:08:13.551 * Non blocking connect for SYNC fired the event.

8532:S 27 May 2021 11:08:13.551 * Master replied to PING, replication can continue…

8532:S 27 May 2021 11:08:13.552 * Partial resynchronization not possible (no cached master)

8532:S 27 May 2021 11:08:13.553 * Full resync from master: d495eb6da5b58f21f7c3921199363305fe6b7da7:0

8532:S 27 May 2021 11:08:13.558 * MASTER <-> REPLICA sync: receiving 996 bytes from master to disk

8532:S 27 May 2021 11:08:13.558 * MASTER <-> REPLICA sync: Flushing old data

8532:S 27 May 2021 11:08:13.558 * MASTER <-> REPLICA sync: Loading DB in memory

8532:S 27 May 2021 11:08:13.559 * Loading RDB produced by version 6.2.3

8532:S 27 May 2021 11:08:13.559 * RDB age 0 seconds

8532:S 27 May 2021 11:08:13.559 * RDB memory usage when created 1.87 Mb

8532:S 27 May 2021 11:08:13.559 * MASTER <-> REPLICA sync: Finished with success

8532:S 27 May 2021 13:28:48.717 * 1 changes in 3600 seconds. Saving…

8532:S 27 May 2021 13:28:48.720 * Background saving started by pid 11249

11249:C 27 May 2021 13:28:48.724 * DB saved on disk

11249:C 27 May 2021 13:28:48.725 * RDB: 0 MB of memory used by copy-on-write

8532:S 27 May 2021 13:28:48.821 * Background saving terminated with success

Replication 확인

# redis-cli -p 6379

127.0.0.1:6379> AUTH test1234 OK 127.0.0.1:6379> info replication

Replication

role:master

connected_slaves:2

slave0:ip=211.xxx.xxx.xxx,port=6380,state=online,offset=938,lag=1

slave1:ip=211.xxx.xxx.xxx,port=6381,state=online,offset=938,lag=1

master_failover_state:no-failover

master_replid:d495eb6da5b58f21f7c3921199363305fe6b7da7

master_replid2:0000000000000000000000000000000000000000

master_repl_offset:938

second_repl_offset:-1

repl_backlog_active:1

repl_backlog_size:1048576

repl_backlog_first_byte_offset:1

repl_backlog_histlen:938

127.0.0.1:6379> quit

# redis-cli -p 6380

127.0.0.1:6380> AUTH test1234

OK

127.0.0.1:6380> info replication

Replication

role:slave

master_host:211.xxx.xxx.xxx

master_port:6379

master_link_status:up

master_last_io_seconds_ago:8

master_sync_in_progress:0

slave_repl_offset:11354

slave_priority:100

slave_read_only:1

replica_announced:1

connected_slaves:0

master_failover_state:no-failover

master_replid:d495eb6da5b58f21f7c3921199363305fe6b7da7

master_replid2:0000000000000000000000000000000000000000

master_repl_offset:11354

second_repl_offset:-1

repl_backlog_active:1

repl_backlog_size:1048576

repl_backlog_first_byte_offset:1

repl_backlog_histlen:11354

127.0.0.1:6380> quit

Replication 테스트

테스트1. Master 데이터가 Slave에 잘 복사 되는지 체크

접속후 AUTH 패스워드를 입력해도 되고, Redis 접속 시 -a 옵션으로 암호를 입력해도 됩니다.

# redis-cli -p 6379 -a test1234

Warning: Using a password with ‘-a’ or ‘-u’ option on the command line interface may not be safe.

127.0.0.1:6379> set replica_test helloworld

OK

127.0.0.1:6379> get replica_test

“helloworld”

127.0.0.1:6379> quit

# redis-cli -p 6380 -a test1234

Warning: Using a password with ‘-a’ or ‘-u’ option on the command line interface may not be safe.

127.0.0.1:6380> get replica_test

“helloworld”

127.0.0.1:6380> quit

Warning: Using a password with ‘-a’ or ‘-u’ option on the command line interface may not be safe.

command line에서 입력하므로 패스워드가 로그로 남아 유출될 수 있습니다.

-a 대신 –askpass 옵션을 써서 암호를 입력합니다.

#redis-cli -p 6381 –askpass

Please input password: *****

127.0.0.1:6381> info replication

Replication

role:slave

master_host:211.xxx.xxx.xxx

master_port:6379

master_link_status:up

master_last_io_seconds_ago:5

master_sync_in_progress:0

slave_repl_offset:12630

slave_priority:100

slave_read_only:1

replica_announced:1 connected_slaves:0

master_failover_state:no-failover

master_replid:d495eb6da5b58f21f7c3921199363305fe6b7da7

master_replid2:0000000000000000000000000000000000000000

master_repl_offset:12630

second_repl_offset:-1

repl_backlog_active:1

repl_backlog_size:1048576

repl_backlog_first_byte_offset:71

repl_backlog_histlen:12560

테스트2. 마스터를 shutdown 한다면?

# sudo /etc/init.d/redis_6379 stop

# vim /usr/local/src/redis/logs/6380.log

8532:S 27 May 2021 14:38:01.558 # Error condition on socket for SYNC: Connection refused

8532:S 27 May 2021 14:38:02.560 * Connecting to MASTER 211.xxx.xxx.xxx:6379

8532:S 27 May 2021 14:38:02.561 * MASTER <-> REPLICA sync started

8532:S 27 May 2021 14:38:02.561 # Error condition on socket for SYNC: Connection refused

8532:S 27 May 2021 14:38:03.563 * Connecting to MASTER 211.xxx.xxx.xxx:6379

8532:S 27 May 2021 14:38:03.564 * MASTER <-> REPLICA sync started

…. Master Redis가 정상 동작할 때까지 동일한 메시지가 반복해서 발생합니다.

이상태로 방치하면 로그파일 크기가 엄청나게 커져버리기 때문에 모니터링이 필수 입니다.

또한 마스터만 read/write가 되고 slave는 read만 되기 때문에 read만 되는 상태가 됩니다.

다시 Master Redis를 시작하고 로그와 정보를 확인합니다.

#sudo /etc/init.d/redis_6379 start

#vim /usr/local/src/redis/logs/6380.log

8532:S 27 May 2021 14:38:03.564 * Non blocking connect for SYNC fired the event.

8532:S 27 May 2021 14:38:03.564 * Master replied to PING, replication can continue…

8532:S 27 May 2021 14:38:03.564 * Trying a partial resynchronization (request d495eb6da5b58f21f7c3921199363305fe6b7da7:41942).

8532:S 27 May 2021 14:38:03.565 * Full resync from master: 2c728a8720bc19ea3bdfceed46f6d41dd78ecb9c:0

8532:S 27 May 2021 14:38:03.565 * Discarding previously cached master state.

8532:S 27 May 2021 14:38:03.604 * MASTER <-> REPLICA sync: receiving 3180 bytes from master to disk

8532:S 27 May 2021 14:38:03.604 * MASTER <-> REPLICA sync: Flushing old data

8532:S 27 May 2021 14:38:03.605 * MASTER <-> REPLICA sync: Loading DB in memory

8532:S 27 May 2021 14:38:03.605 * Loading RDB produced by version 6.2.3

8532:S 27 May 2021 14:38:03.605 * RDB age 0 seconds

8532:S 27 May 2021 14:38:03.605 * RDB memory usage when created 1.84 Mb

8532:S 27 May 2021 14:38:03.606 * MASTER <-> REPLICA sync: Finished with success

8532:S 27 May 2021 14:38:28.033 * 100 changes in 300 seconds. Saving…

8532:S 27 May 2021 14:38:28.034 * Background saving started by pid 12162

12162:C 27 May 2021 14:38:28.037 * DB saved on disk

12162:C 27 May 2021 14:38:28.038 * RDB: 0 MB of memory used by copy-on-write

8532:S 27 May 2021 14:38:28.134 * Background saving terminated with success

#redis-cli -p 6379 –askpass

Please input password: *****

127.0.0.1:6379> keys *

1) “spring:session:expirations:1622095740000”

2) “replica_test”

3) “spring:session:index:org.springframework.session.FindByIndexNameSessionRepository.PRINCIPAL_NAME_INDEX_NAME:testco”

4) “menusCache::POSTAL_MENU”

5) “spring:session:sessions:expires:930ed11b-394d-419a-af57-8fab9dae1656”

6) “spring:session:sessions:930ed11b-394d-419a-af57-8fab9dae1656”

7) “spring:session:expirations:1622095440000”

127.0.0.1:6379>quit

#redis-cli -p 6380 –askpass

Please input password: *****

127.0.0.1:6380> keys *

1) “menusCache::POSTAL_MENU”

2) “spring:session:sessions:930ed11b-394d-419a-af57-8fab9dae1656”

3) “spring:session:sessions:expires:930ed11b-394d-419a-af57-8fab9dae1656”

4) “spring:session:expirations:1622095440000”

5) “replica_test”

6) “spring:session:index:org.springframework.session.FindByIndexNameSessionRepository.PRINCIPAL_NAME_INDEX_NAME:testco”

7) “spring:session:expirations:1622095740000”

하지만 현재 상태로는 Master 서버에 장애가 발생했을때 Slave를 자동으로 Master로 승격해주지 못합니다.

이러한 장애에 자동으로 failover 해주는 도구가 Sentinel(센티널)입니다.

Sentinel(센티널)

Sentinel는 Master Redis를 감시하고 있다가 비정상적인 중단이 발생하면 자동으로 Slave Redis 중 한 개를 채택해서 Master로 승격 시켜주는 FailOver 역할을 합니다. Sentinel API를 통해 관리자나 프로그램에 피드백을 줄 수 있으며, 보통 Sentinel은 3대로 구성합니다. 가능하면 홀 수개로 구성하는데 그 이유는 Master Redis가 중단되었는지 판단할 때 다수결로 결정하기 때문에 홀수로 한다고 합니다. Master Redis가 정상 동작중인데 Master Redis 서버와 Sentinel 서버간의 네트워크 문제가 발생할 수 도 있기 때문에 3대의 Sentinel 중 Master Redis와 통신이 2대 이상 안될 경우 Master Redis가 중단되었다고 판단합니다. 이를 객관적인 다운상태라고 합니다.

Sentinel 설정하기

# cd /usr/local/src/redis

# ls

00-RELEASENOTES 6381 CONTRIBUTING MANIFESTO TLS.md logs runtest-cluster sentinel.conf utils

6379 BUGS COPYING Makefile conf redis.conf runtest-moduleapi src

6380 CONDUCT INSTALL README.md deps runtest runtest-sentinel tests

원본 파일을 그대로 두고 복제 파일을 만듭니다.

Sentinel #1 설정

# cp sentinel.conf ./conf/sentinel_7000.conf

# vim ./conf/sentinel_7000.conf

//첫번째 Sentinel 서버 포트를 지정합니다.

port 7000

daemonize yes

pidfile /var/run/redis-sentinel_7000.pid

logfile “/usr/local/src/redis/logs/redis-sentinel_7000.log”

// Master Redis IP, PORT 그리고 제일 뒤에 숫자 2는 Master Redis가 중단되었다고 판단하는 수 이며, quorum(쿼럼)이라고 부릅니다.

// Sentinel 서버를 한대로 사용한다면 1, 3대면 2 이런씩으로 다수결시 결정될 수 있는 값으로 입력해야 됩니다.

sentinel monitor mymaster 211.xxx.xxx.xxx 6379 2

// 몇초동안 Master Redis로 연결이 실패 했을 때 중단이라고 판단내릴지의 시간입니다. 단위는 밀리세컨드이고 기본값은 30초입니다.

sentinel down-after-milliseconds mymaster 50000

// Master Redis가 중단되었을때 Master에서 Slave로 sync(동기화)해야되는데 동시에 몇개의 Slave를 동기화 시킬지의 개수입니다.

sentinel parallel-syncs mymaster 1

// failOver의 타임아웃으로 기본값은 3분입니다.

sentinel failover-timeout mymaster 180000

// 주석을 풀고 Master Redis 비밀번호를 입력합니다. Master Redis와 Slave Redis 비밀번호를 동일하게 맞춰줘야 합니다.

sentinel auth-pass mymaster test1234

Sentinel #2 설정

# cp sentinel_7000.conf ./conf/sentinel_7001.conf

# vim ./conf/sentinel_7001.conf

port 7001

pidfile /var/run/redis-sentinel_7001.pid

logfile “/usr/local/src/redis/logs/redis-sentinel_7001.log”

Sentinel #3 설정

# cp sentinel_7000.conf ./conf/sentinel_7002.conf

# vim ./conf/sentinel_7002.conf

port 7002

pidfile /var/run/redis-sentinel_7002.pid

logfile “/usr/local/src/redis/logs/redis-sentinel_7002.log”

Sentinel 데몬 작성

# cd /etc/init.d/

# ls README functions netconsole network redis_6379 redis_6380 redis_6381

Sentinel #1

# cp redis_6379 ./sentinel_7000

# vim sentinel_7000

EXEC=/usr/local/bin/redis-sentinel

CLIEXEC=/usr/local/bin/redis-cli

PIDFILE=/var/run/redis-sentinel_7000.pid

CONF=”/usr/local/src/redis/conf/sentinel_7000.conf”

REDISPORT=”7000”

REDISPW=test1234

Sentinel #2

# cp sentinel_7000 ./sentinel_7001

# vim sentinel_7001

EXEC=/usr/local/bin/redis-sentinel

CLIEXEC=/usr/local/bin/redis-cli

PIDFILE=/var/run/redis-sentinel_7001.pid

CONF=”/usr/local/src/redis/conf/sentinel_7001.conf”

REDISPORT=”7001”

REDISPW=test1234

Sentinel #3

# cp sentinel_7000 ./sentinel_7002

# vim sentinel_7002

EXEC=/usr/local/bin/redis-sentinel

CLIEXEC=/usr/local/bin/redis-cli

PIDFILE=/var/run/redis-sentinel_7002.pid

CONF=”/usr/local/src/redis/conf/sentinel_7002.conf”

REDISPORT=”7002”

REDISPW=test1234

Sentinel 시작하기

# sudo /etc/init.d/sentinel_7000 start

# sudo /etc/init.d/sentinel_7001 start

# sudo /etc/init.d/sentinel_7002 start

# ps aux grep redis

root 8532 0.1 0.1 240840 10288 ? Ssl 11:08 0:22 /usr/local/bin/redis-server 0.0.0.0:6380

root 8562 0.1 0.1 240840 10288 ? Ssl 11:08 0:22 /usr/local/bin/redis-server 0.0.0.0:6381

root 12154 0.1 0.1 165060 9928 ? Ssl 14:38 0:08 /usr/local/bin/redis-server 0.0.0.0:6379

root 12906 0.6 0.1 162500 10256 ? Ssl 16:28 0:01 /usr/local/bin/redis-sentinel *:7000 [sentinel]

root 12918 0.7 0.1 162500 10260 ? Ssl 16:28 0:01 /usr/local/bin/redis-sentinel *:7001 [sentinel]

root 12927 0.6 0.1 162628 10296 ? Ssl 16:28 0:01 /usr/local/bin/redis-sentinel *:7002 [sentinel]

root 12948 0.0 0.0 116972 1020 pts/3 S+ 16:31 0:00 grep –color=auto redis

#vim /usr/local/src/redis/logs/redis-sentinel_7000.log

#vim /usr/local/src/redis/logs/redis-sentinel_7001.log

#vim /usr/local/src/redis/logs/redis-sentinel_7002.log

28037:X 28 May 2021 10:42:10.216 # oO0OoO0OoO0Oo Redis is starting oO0OoO0OoO0Oo

28037:X 28 May 2021 10:42:10.217 # Redis version=6.2.3, bits=64, commit=00000000, modified=0, pid=28037, just started

28037:X 28 May 2021 10:42:10.217 # Configuration loaded

28037:X 28 May 2021 10:42:10.218 * Increased maximum number of open files to 10032 (it was originally set to 1024).

28037:X 28 May 2021 10:42:10.218 * monotonic clock: POSIX clock_gettime

28037:X 28 May 2021 10:42:10.219 * Running mode=sentinel, port=7000.

28037:X 28 May 2021 10:42:10.222 # Sentinel ID is bc3291c7c38dd85f2fd58c4453aa194ac21a8c44

28037:X 28 May 2021 10:42:10.222 # +monitor master mymaster 211.xxx.xxx.xxx 6379 quorum 2

28037:X 28 May 2021 10:42:10.223 * +slave slave 211.xxx.xxx.xxx:6380 211.xxx.xxx.xxx 6380 @ mymaster 211.xxx.xxx.xxx 6379

28037:X 28 May 2021 10:42:10.224 * +slave slave 211.xxx.xxx.xxx:6381 211.xxx.xxx.xxx 6381 @ mymaster 211.xxx.xxx.xxx 6379

28037:X 28 May 2021 10:42:15.937 * +sentinel sentinel f6ea7414b5e3daf9292fb9497875af7bba37dcee 211.xxx.xxx.xxx 7001 @ mymaster

211.xxx.xxx.xxx 6379

28037:X 28 May 2021 10:42:18.667 * +sentinel sentinel 9e50008fedfd4b884704a4ee6ad54584f598fb87 211.xxx.xxx.xxx 7002 @ mymaster

211.xxx.xxx.xxx 6379

Sentinel Conf파일 자동 생성되는 설정 정보들

sentinel_7000 ~ 7002 서버를 시작하면 sentinel_7000.conf, sentinel_7001.conf, sentinel_7002.conf 파일 맨 아래에 부분에 자동으로 설정되는 부분이 생깁니다.

Sentinel conf 파일 설정에서는 Master Redis 정보만 입력했지만 Sentinel이 redis ino 명령어를 통해 slave 정보들을 읽어서 slave 정보들을 자동으로 conf 파일에 추가하고 Sentinel 서버간 내부 통신으로 redis의 Pub/Sub 기능을 통해 다른 Sentinel 서버정보를 conf 파일에 자동으로 추가합니다.

# vim sentinel_7000.conf

…..

//맨아래에 자동으로 생성된 설정 정보

#Generated by CONFIG REWRITE

protected-mode no

user default on nopass ~* &* +@all

sentinel myid bc3291c7c38dd85f2fd58c4453aa194ac21a8c44

sentinel config-epoch mymaster 0

sentinel leader-epoch mymaster 0

sentinel current-epoch 0

sentinel known-replica mymaster 211.xxx.xxx.xxx 6380 // Slave Redis 정보

sentinel known-replica mymaster 211.xxx.xxx.xxx 6381 // Slave Redis 정보

sentinel known-sentinel mymaster 211.xxx.xxx.xxx 7001 f6ea7414b5e3daf9292fb9497875af7bba37dcee // sentinel 서버정보

sentinel known-sentinel mymaster 211.xxx.xxx.xxx 7002 9e50008fedfd4b884704a4ee6ad54584f598fb87 // sentinel 서버 정보

#vim sentinel_7001.conf

#Generated by CONFIG REWRITE

protected-mode no

user default on nopass ~* &* +@all

sentinel myid f6ea7414b5e3daf9292fb9497875af7bba37dcee

sentinel config-epoch mymaster 0

sentinel leader-epoch mymaster 0

sentinel current-epoch 0

sentinel known-replica mymaster 211.xxx.xxx.xxx 6381

sentinel known-replica mymaster 211.xxx.xxx.xxx 6380

sentinel known-sentinel mymaster 211.xxx.xxx.xxx 7000 bc3291c7c38dd85f2fd58c4453aa194ac21a8c44

sentinel known-sentinel mymaster 211.xxx.xxx.xxx 7002 9e50008fedfd4b884704a4ee6ad54584f598fb87

Sentinel 검증

현재 Master Redis는 6379이고 Slave Redis는 6380과 6381입니다.

# redis-cli -p 6379 –askpass

Please input password: *****

127.0.0.1:6379> info replication

role:master

connected_slaves:2

slave0:ip=127.0.0.1,port=6380,state=online,offset=160771,lag=1

slave1:ip=127.0.0.1,port=6381,state=online,offset=160771,lag=1

master_failover_state:no-failover

master_replid:2c728a8720bc19ea3bdfceed46f6d41dd78ecb9c

master_replid2:0000000000000000000000000000000000000000

master_repl_offset:161035

second_repl_offset:-1

repl_backlog_active:1

repl_backlog_size:1048576

repl_backlog_first_byte_offset:1

repl_backlog_histlen:161035

127.0.0.1:6379> quit

# redis-cli -p 6380 –askpass

Please input password: *****

127.0.0.1:6380> info replication

role:slave

master_host:127.0.0.1

master_port:6379

master_link_status:up

master_last_io_seconds_ago:0

master_sync_in_progress:0

slave_repl_offset:183101

slave_priority:100

slave_read_only:1

replica_announced:1

connected_slaves:0

master_failover_state:no-failover

master_replid:2c728a8720bc19ea3bdfceed46f6d41dd78ecb9c

master_replid2:0000000000000000000000000000000000000000

master_repl_offset:183101

second_repl_offset:-1

repl_backlog_active:1

repl_backlog_size:1048576

repl_backlog_first_byte_offset:1

repl_backlog_histlen:183101

127.0.0.1:6380> quit

Master Redis를 중단 시키고 상태를 확인 합니다.

# /etc/init.d/redis_6379 stop

Stopping …

Warning: Using a password with ‘-a’ or ‘-u’ option on the command line interface may not be safe.

Waiting for Redis to shutdown …

Redis stopped

Master Redis(6379) log - 중단됨

# vim /usr/local/src/redis/logs/6379.log

27921:M 28 May 2021 10:38:51.950 * Replica 211.xxx.xxx.xxx:6381 asks for synchronization

27921:M 28 May 2021 10:38:51.950 * Partial resynchronization not accepted: Replication ID mismatch (Replica asked for

‘818b4519e78e443ee4f9bde17b13496e22b352dc’, my replication IDs are ‘9461bcf55a4c832424c37d610b0edb44da16432b’ and

‘0000000000000000000000000000000000000000’)

27921:M 28 May 2021 10:38:51.950 * Starting BGSAVE for SYNC with target: disk

27921:M 28 May 2021 10:38:51.951 * Background saving started by pid 27942

27942:C 28 May 2021 10:38:51.955 * DB saved on disk

27942:C 28 May 2021 10:38:51.955 * RDB: 0 MB of memory used by copy-on-write

27921:M 28 May 2021 10:38:52.016 * Background saving terminated with success

27921:M 28 May 2021 10:38:52.017 * Synchronization with replica 211.xxx.xxx.xxx:6381 succeeded

27921:M 28 May 2021 10:55:17.228 # User requested shutdown…

27921:M 28 May 2021 10:55:17.228 * Saving the final RDB snapshot before exiting.

27921:M 28 May 2021 10:55:17.230 * DB saved on disk

27921:M 28 May 2021 10:55:17.230 * Removing the pid file.

27921:M 28 May 2021 10:55:17.230 # Redis is now ready to exit, bye bye…

Slave Redis(6380) log - Master Redis와 Sync를 시도하다가 Sentinel이 승격시킴

# vim /usr/local/src/redis/logs/6380.log

27928:S 28 May 2021 10:55:45.586 * MASTER <-> REPLICA sync started

27928:S 28 May 2021 10:55:45.587 # Error condition on socket for SYNC: Connection refused

27928:S 28 May 2021 10:55:46.593 * Connecting to MASTER 211.xxx.xxx.xxx:6379

27928:S 28 May 2021 10:55:46.593 * MASTER <-> REPLICA sync started

27928:S 28 May 2021 10:55:46.593 # Error condition on socket for SYNC: Connection refused

27928:S 28 May 2021 10:55:47.598 * Connecting to MASTER 211.xxx.xxx.xxx:6379

27928:S 28 May 2021 10:55:47.599 * MASTER <-> REPLICA sync started

27928:S 28 May 2021 10:55:47.599 # Error condition on socket for SYNC: Connection refused

…

sentinel conf에서 설정한 sentinel down-after-milliseconds mymaster 50000 정보를 토대로 5초동안 master가 끊겼다고 판단할때까지 위 메시지 반복됨.sentinel 이 master가 중단되었다고 판단하면 slave 중 하나를 선택하여 마스터로 승격시킴.

27928:M 28 May 2021 10:55:47.669 * Discarding previously cached master state.

27928:M 28 May 2021 10:55:47.669 # Setting secondary replication ID to 9461bcf55a4c832424c37d610b0edb44da16432b, valid up to offset: 158082.

New replication ID is 7cd2df002f40d6f4f5522fefc15e79cd72240a9a

27928:M 28 May 2021 10:55:47.669 * MASTER MODE enabled (user request from ‘id=10 addr=211.xxx.xxx.xxx:41215 laddr=211.xxx.xxx.xxx:6380 fd=12

name=sentinel-9e50008f-cmd age=811 idle=0 flags=x db=0 sub=0 psub=0 multi=4 qbuf=188 qbuf-free=40766 argv-mem=4 obl=45 oll=0 omem=0 tot-

mem=61468 events=r cmd=exec user=default redir=-1’)

27928:M 28 May 2021 10:55:47.674 # CONFIG REWRITE executed with success.

27928:M 28 May 2021 10:55:48.108 * Replica 211.xxx.xxx.xxx:6381 asks for synchronization

27928:M 28 May 2021 10:55:48.108 * Partial resynchronization request from 211.xxx.xxx.xxx:6381 accepted. Sending 295 bytes of backlog starting from offset 158082.

Sentinel(7000) log - switch-master를 통해 6379에서 6380 Redis로 Master를 변경함.

# vim /usr/local/src/redis/logs/redis-sentinel_7000.log

28037:X 28 May 2021 10:55:47.325 # +sdown master mymaster 211.xxx.xxx.xxx 6379

28037:X 28 May 2021 10:55:47.429 # +new-epoch 1

28037:X 28 May 2021 10:55:47.430 # +vote-for-leader 9e50008fedfd4b884704a4ee6ad54584f598fb87 1

28037:X 28 May 2021 10:55:48.104 # +config-update-from sentinel 9e50008fedfd4b884704a4ee6ad54584f598fb87 211.xxx.xxx.xxx 7002 @ mymaster

211.xxx.xxx.xxx 6379

28037:X 28 May 2021 10:55:48.104 # +switch-master mymaster 211.xxx.xxx.xxx 6379 211.xxx.xxx.xxx 6380

28037:X 28 May 2021 10:55:48.104 * +slave slave 211.xxx.xxx.xxx:6381 211.xxx.xxx.xxx 6381 @ mymaster 211.xxx.xxx.xxx 6380

28037:X 28 May 2021 10:55:48.104 * +slave slave 211.xxx.xxx.xxx:6379 211.xxx.xxx.xxx 6379 @ mymaster 211.xxx.xxx.xxx 6380

28037:X 28 May 2021 10:56:18.138 # +sdown slave 211.xxx.xxx.xxx:6379 211.xxx.xxx.xxx 6379 @ mymaster 211.xxx.xxx.xxx 6380

자동으로 변경된 Conf파일 정보

6379.conf(구 Master Redis) - 현재 중단된 상태이고, 변경된 정보는 없음

# vim /usr/local/src/redis/conf/6380.conf(현 Master Redis) - Sentinel에 의해 Master로 승격되었으며, conf파일 맨 아래 중에 자동으로 설정정보가 추가됨.

…

#Generated by CONFIG REWRITE

save 3600 1

save 300 100

save 60 10000

user default on #49de38df71bfb36054c27bf55182374a3604cae480fc7132be78c2c505ade710 ~* &* +@all

# vim /usr/local/src/redis/conf/6380.conf(현 Slave Redis)

…

#Generated by CONFIG REWRITE

save 3600 1 save 300 100 save 60 10000 user default on #49de38df71bfb36054c27bf55182374a3604cae480fc7132be78c2c505ade710 ~* &* +@all

# vim /usr/local/src/redis/conf/sentinel_7000.conf (Sentinel #1)

…

#Generated by CONFIG REWRITE

protected-mode no

user default on nopass ~* &* +@all

sentinel myid bc3291c7c38dd85f2fd58c4453aa194ac21a8c44

sentinel config-epoch mymaster 1

sentinel leader-epoch mymaster 1

sentinel current-epoch 1

sentinel known-replica mymaster 211.xxx.xxx.xxx 6379

sentinel known-replica mymaster 211.xxx.xxx.xxx 6381

sentinel known-sentinel mymaster 211.xxx.xxx.xxx 7001 f6ea7414b5e3daf9292fb9497875af7bba37dcee

sentinel known-sentinel mymaster 211.xxx.xxx.xxx 7002 9e50008fedfd4b884704a4ee6ad54584f598fb87

# vim /usr/local/src/redis/conf/sentinel_7001.conf (Sentinel #2)

…

#Generated by CONFIG REWRITE

protected-mode no

user default on nopass ~* &* +@all

sentinel myid f6ea7414b5e3daf9292fb9497875af7bba37dcee

sentinel config-epoch mymaster 1

sentinel leader-epoch mymaster 1

sentinel current-epoch 1

sentinel known-replica mymaster 211.xxx.xxx.xxx 6379

sentinel known-replica mymaster 211.xxx.xxx.xxx 6381

sentinel known-sentinel mymaster 211.xxx.xxx.xxx 7000 bc3291c7c38dd85f2fd58c4453aa194ac21a8c44

sentinel known-sentinel mymaster 211.xxx.xxx.xxx 7002 9e50008fedfd4b884704a4ee6ad54584f598fb87

# vim /usr/local/src/redis/conf/sentinel_7002.conf (Sentinel #3)

…

#Generated by CONFIG REWRITE

protected-mode no

user default on nopass ~* &* +@all

sentinel myid 9e50008fedfd4b884704a4ee6ad54584f598fb87

sentinel config-epoch mymaster 1

sentinel leader-epoch mymaster 1

sentinel current-epoch 1

sentinel known-replica mymaster 211.xxx.xxx.xxx 6379

sentinel known-replica mymaster 211.xxx.xxx.xxx 6381

sentinel known-sentinel mymaster 211.xxx.xxx.xxx 7001 f6ea7414b5e3daf9292fb9497875af7bba37dcee

sentinel known-sentinel mymaster 211.xxx.xxx.xxx 7000 bc3291c7c38dd85f2fd58c4453aa194ac21a8c44

만약 정상 동작하지 않는다면 Conf파일에 자동으로 생성된 설정 값들을 지우고 데몬들을 순차적으로 다시 시작 해볼것.

Redis Info Replication 정보 확인

# redis-cli -p 6379 –askpass

Please input password: *****

Could not connect to Redis at 127.0.0.1:6379: Connection refused

not connected> quit

# redis-cli -p 6380 –askpass

Please input password: *****

127.0.0.1:6380> info replication

#Replication

role:master

connected_slaves:1

slave0:ip=211.xxx.xxx.xxx,port=6381,state=online,offset=478420,lag=1

master_failover_state:no-failover

master_replid:7cd2df002f40d6f4f5522fefc15e79cd72240a9a

master_replid2:9461bcf55a4c832424c37d610b0edb44da16432b

master_repl_offset:478692

second_repl_offset:158082

repl_backlog_active:1

repl_backlog_size:1048576

repl_backlog_first_byte_offset:1

repl_backlog_histlen:478692

# redis-cli -p 6381 –askpass

Please input password: *****

127.0.0.1:6381> info replication

#Replication

role:slave

master_host:211.xxx.xxx.xxx

master_port:6380

master_link_status:up

master_last_io_seconds_ago:1

master_sync_in_progress:0

slave_repl_offset:513746

slave_priority:100

slave_read_only:1

replica_announced:1

connected_slaves:0

master_failover_state:no-failover

master_replid:7cd2df002f40d6f4f5522fefc15e79cd72240a9a

master_replid2:9461bcf55a4c832424c37d610b0edb44da16432b

master_repl_offset:513746

second_repl_offset:158082

repl_backlog_active:1

repl_backlog_size:1048576

repl_backlog_first_byte_offset:1

repl_backlog_histlen:513746

이전 Master Redis(6379)를 다시 시작하면 Slave로 등록되고, 현 Master Redis(6380)과 Sync합니다.

테스트 key하나를 현 Master Redis(6380)에 추가 합니다.

#redis-cli -p 6380 –askpass

Please input password: *****

127.0.0.1:6380> set sentinel_test 12345

OK

127.0.0.1:6380> get sentinel_test

“12345”

Master Redis와 Snyc함

# /etc/init.d/redis_6379 start & tail -f /usr/local/src/redis/logs/6379.log

29827:C 28 May 2021 11:33:38.000 # oO0OoO0OoO0Oo Redis is starting oO0OoO0OoO0Oo

29827:C 28 May 2021 11:33:38.000 # Redis version=6.2.3, bits=64, commit=00000000, modified=0, pid=29827, just started

29827:C 28 May 2021 11:33:38.000 # Configuration loaded

29827:M 28 May 2021 11:33:38.002 * Increased maximum number of open files to 10032 (it was originally set to 1024).

29827:M 28 May 2021 11:33:38.002 * monotonic clock: POSIX clock_gettime

29827:M 28 May 2021 11:33:38.003 * Running mode=standalone, port=6379.

29827:M 28 May 2021 11:33:38.003 # Server initialized

29827:M 28 May 2021 11:33:38.004 * Loading RDB produced by version 6.2.3

29827:M 28 May 2021 11:33:38.004 * RDB age 2301 seconds

29827:M 28 May 2021 11:33:38.004 * RDB memory usage when created 2.03 Mb

29827:M 28 May 2021 11:33:38.004 * DB loaded from disk: 0.000 seconds

29827:M 28 May 2021 11:33:38.004 * Ready to accept connections

29827:S 28 May 2021 11:33:48.283 * Before turning into a replica, using my own master parameters to synthesize a cached master: I may be able to

synchronize with the new master with just a partial transfer.

29827:S 28 May 2021 11:33:48.283 * Connecting to MASTER 211.xxx.xxx.xxx:6380

29827:S 28 May 2021 11:33:48.283 * MASTER <-> REPLICA sync started

29827:S 28 May 2021 11:33:48.283 * REPLICAOF 211.xxx.xxx.xxx:6380 enabled (user request from ‘id=3 addr=211.xxx.xxx.xxx:51756

laddr=211.xxx.xxx.xxx:6379 fd=7 name=sentinel-9e50008f-cmd age=10 idle=0 flags=x db=0 sub=0 psub=0 multi=4 qbuf=199 qbuf-free=40755 argv-

mem=4 obl=45 oll=0 omem=0 tot-mem=61468 events=r cmd=exec user=default redir=-1’)

29827:S 28 May 2021 11:33:48.288 # CONFIG REWRITE executed with success.

29827:S 28 May 2021 11:33:48.288 * Non blocking connect for SYNC fired the event.

29827:S 28 May 2021 11:33:48.289 * Master replied to PING, replication can continue…

29827:S 28 May 2021 11:33:48.289 * Trying a partial resynchronization (request e77afac4a6263a356e8b8dc3ec002ac7d162515e:1).

29827:S 28 May 2021 11:33:48.291 * Full resync from master: 7cd2df002f40d6f4f5522fefc15e79cd72240a9a:618150

29827:S 28 May 2021 11:33:48.291 * Discarding previously cached master state.

29827:S 28 May 2021 11:33:48.322 * MASTER <-> REPLICA sync: receiving 467 bytes from master to disk

29827:S 28 May 2021 11:33:48.322 * MASTER <-> REPLICA sync: Flushing old data

29827:S 28 May 2021 11:33:48.322 * MASTER <-> REPLICA sync: Loading DB in memory

29827:S 28 May 2021 11:33:48.323 * Loading RDB produced by version 6.2.3

29827:S 28 May 2021 11:33:48.323 * RDB age 0 seconds

29827:S 28 May 2021 11:33:48.323 * RDB memory usage when created 1.97 Mb

29827:S 28 May 2021 11:33:48.323 * MASTER <-> REPLICA sync: Finished with success

다시 시작된 Slave Redis와 Snyc됨

# vim /usr/local/src/redis/logs/6380.log

27928:M 28 May 2021 11:33:48.289 * Replica 211.xxx.xxx.xxx:6379 asks for synchronization

27928:M 28 May 2021 11:33:48.289 * Partial resynchronization not accepted: Replication ID mismatch (Replica asked for

‘e77afac4a6263a356e8b8dc3ec002ac7d162515e’, my replication IDs are ‘7cd2df002f40d6f4f5522fefc15e79cd72240a9a’ and

‘9461bcf55a4c832424c37d610b0edb44da16432b’)

27928:M 28 May 2021 11:33:48.289 * Starting BGSAVE for SYNC with target: disk

27928:M 28 May 2021 11:33:48.290 * Background saving started by pid 29836

29836:C 28 May 2021 11:33:48.293 * DB saved on disk

29836:C 28 May 2021 11:33:48.294 * RDB: 0 MB of memory used by copy-on-write

27928:M 28 May 2021 11:33:48.322 * Background saving terminated with success

27928:M 28 May 2021 11:33:48.322 * Synchronization with replica 211.xxx.xxx.xxx:6379 succeeded

6379 Slave Redis를 감지하고 6379 Redis Conf 파일 설정을 추가함

# vim /usr/local/src/redis/logs/redis-sentinel_7000.log

28037:X 28 May 2021 10:55:47.325 # +sdown master mymaster 211.xxx.xxx.xxx 6379

28037:X 28 May 2021 10:55:47.429 # +new-epoch 1

28037:X 28 May 2021 10:55:47.430 # +vote-for-leader 9e50008fedfd4b884704a4ee6ad54584f598fb87 1

28037:X 28 May 2021 10:55:48.104 # +config-update-from sentinel 9e50008fedfd4b884704a4ee6ad54584f598fb87 211.xxx.xxx.xxx 7002 @ mymaster

211.xxx.xxx.xxx 6379

28037:X 28 May 2021 10:55:48.104 # +switch-master mymaster 211.xxx.xxx.xxx 6379 211.xxx.xxx.xxx 6380

28037:X 28 May 2021 10:55:48.104 * +slave slave 211.xxx.xxx.xxx:6381 211.xxx.xxx.xxx 6381 @ mymaster 211.xxx.xxx.xxx 6380

28037:X 28 May 2021 10:55:48.104 * +slave slave 211.xxx.xxx.xxx:6379 211.xxx.xxx.xxx 6379 @ mymaster 211.xxx.xxx.xxx 6380

28037:X 28 May 2021 10:56:18.138 # +sdown slave 211.xxx.xxx.xxx:6379 211.xxx.xxx.xxx 6379 @ mymaster 211.xxx.xxx.xxx 6380

28037:X 28 May 2021 11:33:38.491 # -sdown slave 211.xxx.xxx.xxx:6379 211.xxx.xxx.xxx 6379 @ mymaster 211.xxx.xxx.xxx 6380

맨 아래 자동으로 추가됨

# vim /usr/local/src/redis/conf/6379.conf

…

#Generated by CONFIG REWRITE

save 3600 1

save 300 100

save 60 10000

user default on #49de38df71bfb36054c27bf55182374a3604cae480fc7132be78c2c505ade710 ~* &* +@all

replicaof 211.xxx.xxx.xxx 6380

Sentinel이 Slave Redis를 Master Redis로 승격시킬때 맨 아래 replicaof 부분이 삭제함

# vim /usr/local/src/redis/conf/6380.conf

…

#Generated by CONFIG REWRITE

save 3600 1

save 300 100

save 60 10000

user default on #49de38df71bfb36054c27bf55182374a3604cae480fc7132be78c2c505ade710 ~* &* +@all

# redis-cli -p 6379 –askpass

Please input password: *****

127.0.0.1:6379> info replication

role:slave

master_host:211.xxx.xxx.xxx

master_port:6380

master_link_status:up

master_last_io_seconds_ago:0

master_sync_in_progress:0

slave_repl_offset:752105

slave_priority:100

slave_read_only:1

replica_announced:1

connected_slaves:0

master_failover_state:no-failover

master_replid:7cd2df002f40d6f4f5522fefc15e79cd72240a9a

master_replid2:0000000000000000000000000000000000000000

master_repl_offset:752105

second_repl_offset:-1

repl_backlog_active:1

repl_backlog_size:1048576

repl_backlog_first_byte_offset:618151

repl_backlog_histlen:133955

// 테스트 key 값 확인

127.0.0.1:6379> get sentinel_test

“12345”

127.0.0.1:6379> quit

#redis-cli -p 6380 –askpass

Please input password: *****

127.0.0.1:6380> info replication

role:master

connected_slaves:2

slave0:ip=211.xxx.xxx.xxx,port=6381,state=online,offset=767027,lag=0

slave1:ip=211.xxx.xxx.xxx,port=6379,state=online,offset=767027,lag=1

master_failover_state:no-failover

master_replid:7cd2df002f40d6f4f5522fefc15e79cd72240a9a

master_replid2:9461bcf55a4c832424c37d610b0edb44da16432b

master_repl_offset:767027

second_repl_offset:158082

repl_backlog_active:1

repl_backlog_size:1048576

repl_backlog_first_byte_offset:1

repl_backlog_histlen:767027

Redis 서버 수신 상태 확인

# ss -an grep 6379

tcp LISTEN 0 511 :6379 *:

tcp TIME-WAIT 0 0 127.0.0.1:59480 127.0.0.1:6379

tcp ESTAB 0 0 211.xxx.xxx.xxx:6379 211.xxx.xxx.xxx:42279

tcp ESTAB 0 0 211.xxx.xxx.xxx:42279 211.xxx.xxx.xxx:6379

tcp ESTAB 0 0 211.xxx.xxx.xxx:42356 211.xxx.xxx.xxx:6379

tcp ESTAB 0 0 211.xxx.xxx.xxx:6379 211.xxx.xxx.xxx:51756

tcp ESTAB 0 0 211.xxx.xxx.xxx:55217 211.xxx.xxx.xxx:6379

tcp ESTAB 0 0 211.xxx.xxx.xxx:6379 211.xxx.xxx.xxx:60046

tcp ESTAB 0 0 211.xxx.xxx.xxx:6379 211.xxx.xxx.xxx:47594

tcp ESTAB 0 0 211.xxx.xxx.xxx:6379 211.xxx.xxx.xxx:55217

tcp ESTAB 0 0 211.xxx.xxx.xxx:60046 211.xxx.xxx.xxx:6379

tcp ESTAB 0 0 211.xxx.xxx.xxx:6379 222.xxx.xxx.214:50338

tcp ESTAB 0 0 211.xxx.xxx.xxx:6379 222.xxx.xxx.214:50339

tcp ESTAB 0 0 211.xxx.xxx.xxx:51756 211.xxx.xxx.xxx:6379

tcp ESTAB 0 0 211.xxx.xxx.xxx:47594 211.xxx.xxx.xxx:6379

tcp ESTAB 0 0 211.xxx.xxx.xxx:6379 211.xxx.xxx.xxx:42356

Redis Memory Size 변경

# /usr/local/src/redis/conf/6379.conf
첫번째 라인에 아래 부분 추가하고 저장
maxmemory 2gb

# sudo /etc/init.d/redis_6379 restart

# redis-cli -p 6379 --askpass
password: **********

127.0.0.1:6379> config get maxmemory
1) "maxmemory"
2) "2147483648"

Service 등록

기존 redis, sentinel 실행 파일 위치 변경

# mkdir /usr/local/src/redis/bin
# cp /etc/init.d/redis_* /usr/local/src/redis/bin/
# cp /etc/init.d/sentinel_* /usr/local/src/redis/bin/

redis, sentinel 서비스 파일 생성

# vim /etc/systemd/system/redis-server-6379.service
[Unit]                                               
Description=Redis Server 6379                        
Wants=network-online.target                          
After=network.target                                 
                                                     
[Service]                                            
Type=forking                                         
User=root                                            
Restart=on-failure                                   
ExecStart=/usr/local/src/redis/bin/redis_6379 start  
ExecStop=/usr/local/src/redis/bin/redis_6379 stop    
                                                     
[Install]                                            
WantedBy=multi-user.target

나머지 redis도 서비스 파일 생성.

# systemctl enable redis-server-6379.service
# sudo systemctl daemon-reload

실행중인 redis 서비스 확인

# ps aux | grep redis

실행 중인 redis 하나를 내리고 서비스 하나를 실행한다.

# /etc/init.d/redis_6379 stop

# systemctl start redis-server-6379.service
# systemctl status redis-server-6379.service
● redis-server-6379.service - Redis Server 6379
   Loaded: loaded (/etc/systemd/system/redis-server-6379.service; enabled; vendor preset: disabled)
   Active: active (running) since Mon 2023-11-20 11:07:08 KST; 1min 13s ago
  Process: 9146 ExecStart=/usr/local/src/redis/bin/redis_6379 start (code=exited, status=0/SUCCESS)
 Main PID: 9148 (redis-server)
   CGroup: /system.slice/redis-server-6379.service
           └─9148 /usr/local/bin/redis-server 0.0.0.0:6379
           
           
# /etc/init.d/redis_6380 stop
# systemctl start redis-server-6380
# systemctl status redis-server-6380

sentinel service 작성

[Unit]
Description=Redis Sentinel 7000
After=network.target redis-server-6379 redis-server-6380
Wants=network-online.target

[Service]
Type=forking
User=root
Restart=on-failure
ExecStart=/usr/local/src/redis/bin/sentinel_7000 start
ExecStop=/usr/local/src/redis/bin/sentinel_7000 stop

[Install]
WantedBy=multi-user.target

# systemctl enable redis-sentinel-7000.service
# sudo systemctl daemon-reload

실행 중인 sentinel 종료 후 service로 실행

# /etc/init.d/sentinel_7000 stop
# systemctl start redis-sentinel-7000
# systemctl status redis-sentinel-7000

나머지 sentinel도 동일하게 작업

/etc/init.d 경로 redis, sentinel 파일 삭제

# rm -f /etc/init.d/redis_*
# rm -f /etc/init.d/sentinel_*

재부팅 후 redis, sentinel 데몬 확인

# shutdown -r now

외부접속을 위한 방화벽 작업

방화벽 실행 여부 확인

firewall-cmd –state

방화벽 작동중인지 확인

firewall-cmd –state

현재 적용중인 zone확인

firewall-cmd –get-default-zone

서비스 추가 [ 선택사항, 아래에서 특정 port에 대한 ip 제한안하고 서비스로 구분하여 방화벽 열때 ]

firewall-cmd –zone=public –add-service=redis –permanent

서비스 삭제 [ 선택사항, 아래에서 특정 port에 대한 ip 제한 할려고 기존 추가한 service 삭제시] firewall-cmd –permanent –zone=public –remove-service=redis

허용ip추가 firewall-cmd –permanent –zone=public –add-source=106.xxx.xxx.0/24

특정 port에 대한 ip 제한 sudo firewall-cmd –permanent –add-rich-rule=’rule family=”ipv4” source address=106.xxx.xxx.0/24 port port=”7000” protocol=”tcp” accept’

sudo firewall-cmd –permanent –add-rich-rule=’rule family=”ipv4” source address=222.xxx.xxx.0/24 port port=”7000” protocol=”tcp” accept’

sudo firewall-cmd –permanent –add-rich-rule=’rule family=”ipv4” source address=211.xxx.xxx.xxx/24 port port=”7000” protocol=”tcp” accept’

sudo firewall-cmd –permanent –add-rich-rule=’rule family=”ipv4” source address=106.xxx.xxx.0/24 port port=”7001” protocol=”tcp” accept’

sudo firewall-cmd –permanent –add-rich-rule=’rule family=”ipv4” source address=222.xxx.xxx.0/24 port port=”7001” protocol=”tcp” accept’

sudo firewall-cmd –permanent –add-rich-rule=’rule family=”ipv4” source address=211.xxx.xxx.xxx/24 port port=”7001” protocol=”tcp” accept’

sudo firewall-cmd –permanent –add-rich-rule=’rule family=”ipv4” source address=106.xxx.xxx.0/24 port port=”7002” protocol=”tcp” accept’

sudo firewall-cmd –permanent –add-rich-rule=’rule family=”ipv4” source address=222.xxx.xxx.0/24 port port=”7002” protocol=”tcp” accept’

sudo firewall-cmd –permanent –add-rich-rule=’rule family=”ipv4” source address=211.xxx.xxx.xxx/24 port port=”7002” protocol=”tcp” accept’

방화벽 재시작 firewall-cmd –reload

설정확인 firewall-cmd –list-all

내컴퓨터에서 원격 Sentinel 서버 접속 여부 체크

Sentinel 체크시

C:\Program Files\Redis>redis-cli -h 211.xxx.xxx.xxx -p 7003

211.xxx.xxx.xxx:7000> info sentinel

sentinel_masters:1

sentinel_tilt:0

sentinel_running_scripts:0

sentinel_scripts_queue_length:0

sentinel_simulate_failure_flags:0

master0:name=mymaster,status=ok,address=211.xxx.xxx.xxx:6379,slaves=2,sentinels=3

211.xxx.xxx.xxx:7000>

…..

내컴퓨터에서 원격 Redis 서버 접속 여부 체크

C:\Program Files\Redis>redis-cli -h 211.xxx.xxx.xxx -p 6379

211.xxx.xxx.xxx:6379> AUTH test1234

211.xxx.xxx.xxx:6379> keys * 1) “spring:session:index:org.springframework.session.FindByIndexNameSessionRepository.PRINCIPAL_NAME_INDEX_NAME:testco”

2) “spring:session:sessions:expires:4544302f-778c-446c-8ee8-e56470cf1054”

3) “spring:session:expirations:1622183220000”

4) “spring:session:sessions:4544302f-778c-446c-8ee8-e56470cf1054”

5) “replica_test”

6) “menusCache::POSTAL_MENU”

7) “sentinel_test”

logrotate로 로그 파일 일자별로 생성하도록 설정하기

logrotate는 로그 파일의 회전, 압축, 삭제 및 메일링 등 자동으로 수행할 수 있는 기능입니다. 각 로그 파일을 매일, 매주, 매월 또는 size 별 처리할 수 있도록 logrotate를 구성할 수 있습니다.

/etc/logrotate.conf은 logrotate의 기본 구성 파일이고, /etc/logrotate.d/ 는 특정 서비스 별로 설정할 수 있는 디렉토리입니다.

주요 설정 옵션

yearly // 매년

monthly // 매월

weekly // 매주

daily // 매일

rotate 5 // log 파일이 5개가 되면 첫번째 생성된 log 파일 삭제 후 생성.

dateext // 로그 파일에 날짜 부여

compress // 로그파일 압축여부

daily,weekly,monthly // 로그파일분할 시간지정.

delaycompress // 최근파일외 전부압축

errors “emailid” // 에러발생시 메일발솔여부 지정.

missingok // 로그파일분실시 에러로 인정안함.

notifempty // 로그파일이 제로인경우 분할하지않음.

olddir “dir” // 분할된 로그파일을 지정한 디렉토리에 저장함.

postrotate // 로그파일분할후 실행할 script지정.

prerotate // 로그파일분할전 실행할 script지정.

rotate ‘n’ // 저장할 로그파일개수 지정

sharedscripts // 한세트의 로그파일에 한번만 로그분할작업 실행.

size=’logsize’ // 로그파일이 지정된 logsize보다 클떄만 분할

Redis 서버 별 로그 설정

# cd /etc/logrotate.d

# vim redis-6379

/usr/local/src/redis/logs/6379.log {

​ daily

​ rotate 7

​ copytruncate

​ delaycompress

​ compress

​ dateext

​ notifempty

​ missingok }

# cp redis-6379 ./redis-6380

# vim redis-6380

/usr/local/src/redis/logs/6380.log {

​ daily

​ rotate 7

​ copytruncate

​ delaycompress

​ compress

​ dateext

​ notifempty

​ missingok }

# cp redis-6380 ./redis-6381

# vim redis-6381

/usr/local/src/redis/logs/6381.log {

​ daily

​ rotate 7

​ copytruncate

​ delaycompress

​ compress

​ dateext

​ notifempty

​ missingok }

위 형식대로 sentinel 로그도 생성해줍니다.

#cp redis-6380 ./redis-sentinel_7000

#cp redis-sentinel_7000 ./redis-sentinel_7001

#cp redis-sentinel_7000 ./redis-sentinel_7002

logrotate 실행 : 기본 값은 새벽 4시에 자동으로 실행 되며, 시간대를 변경하려면 /etc/cron.daily/에 있는 logrotate파일을 다른 이동시키고, 름은 logrotate.sh로 변경, etc/cron.d/ 아래 하나의 파일을 하나 생성하여, 그안에 스크립트를 작성해 주면 된다.

0 0 * * * root /usr/logrotate/logrotate.sh

지금 강제 실행은 아래 명령으로 하면 됩니다.

# /usr/sbin/logrotate -vf /etc/logrotate.conf

reading config file /etc/logrotate.conf

including /etc/logrotate.d

reading config file bootlog

reading config file chrony

reading config file firewalld

reading config file redis-6379

reading config file redis-6380

reading config file redis-6381

reading config file redis-sentinel_7000

reading config file redis-sentinel_7001

reading config file redis-sentinel_7002

reading config file syslog

reading config file wpa_supplicant

reading config file yum

Allocating hash table for state file, size 15360 B

Handling 14 logs

…

rotating pattern: /usr/local/src/redis/logs/6379.log forced from command line (7 rotations)

empty log files are not rotated, old logs are removed

considering log /usr/local/src/redis/logs/6379.log

log needs rotating

rotating log /usr/local/src/redis/logs/6379.log, log->rotateCount is 7

dateext suffix ‘-20210531’

glob pattern ‘-[0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]’

glob finding logs to compress failed

glob finding old rotated logs failed

copying /usr/local/src/redis/logs/6379.log to /usr/local/src/redis/logs/6379.log-20210531

set default create context to unconfined_u:object_r:usr_t:s0

truncating /usr/local/src/redis/logs/6379.log

….

# ls -lah /usr/local/src/redis/logs

합계 4.9M

drwxr-xr-x. 2 root root 4.0K 5월 31 13:55 .

drwxrwxr-x. 12 root root 4.0K 5월 28 11:34 ..

-rw-r–r–. 1 root root 0 5월 31 13:55 6379.log

-rw-r–r–. 1 root root 352 5월 31 13:55 6379.log-20210531

-rw-r–r–. 1 root root 0 5월 31 13:55 6380.log

-rw-r–r–. 1 root root 602K 5월 31 13:55 6380.log-20210531

-rw-r–r–. 1 root root 0 5월 31 13:55 6381.log

-rw-r–r–. 1 root root 598K 5월 31 13:55 6381.log-20210531

-rw-r–r–. 1 root root 0 5월 27 10:12 AUTH

-rw-r–r–. 1 root root 0 5월 27 10:12 info

-rw-r–r–. 1 root root 0 5월 31 13:55 redis-sentinel_7000.log

-rw-r–r–. 1 root root 1.3M 5월 31 13:55 redis-sentinel_7000.log-20210531

-rw-r–r–. 1 root root 0 5월 31 13:55 redis-sentinel_7001.log

-rw-r–r–. 1 root root 1.3M 5월 31 13:55 redis-sentinel_7001.log-20210531

-rw-r–r–. 1 root root 0 5월 31 13:55 redis-sentinel_7002.log

-rw-r–r–. 1 root root 1.3M 5월 31 13:55 redis-sentinel_7002.log-20210531

모니터링 - redis-stat

redis-stat ( https://github.com/junegunn/redis-stat )

#yum install git

#git clone https://github.com/junegunn/redis-stat.git

#cd redis-stat

#gem install redis-stat

#gem -v

#gem list grep redis-stat

Spring jedis보다 lettuce를 써야되는 이유

여러 쓰레드에서 단일 jedis 인스턴스를 공유하려 할때 jedis는 쓰레드에 안전하지 않다. 멀티쓰레드환경에서 고려해야할 상황이 따른다.

안전한 방법은 pooling(Thread-pool)과 같은 jedis-pool을 사용하는 것이지만 물리적인 비용의(connection할 인스턴스를 미리 만들어놓고 대기하는 연결비용의 증가) 증가가 따른다.

반면 lettuce는 netty 라이브러리 위에서 구축되었고, connection 인스턴스((StatefulRedisConnection))를 여러 쓰레드에서 공유가 가능하다.(Thread-safe)

=> 혼동할 수 있는데, connection 인스턴스의 공유라는 점에서 Thread-safe를 논점을 둔 점이고 Single-Thread의 레디스에 데이터에 접근할 때는 또다르게 고려해야 할 점이다.

redis:
  host: 211.xxx.xxx.xxx
  port: 7000
  connect-timeout: 10s
  password: test1234
  sentinel:
    master: mymaster
    nodes:
      - 211.xxx.xxx.xxx:7000
      - 211.xxx.xxx.xxx:7001
      - 211.xxx.xxx.xxx:7002
  lettuce:
    shutdown-timeout: 500ms

@Configuration
@EnableRedisHttpSession(maxInactiveIntervalInSeconds = 1800)
public class SessionConfig extends AbstractHttpSessionApplicationInitializer {
    private final RedisProperties redisProperties;

    public SessionConfig(RedisProperties redisProperties) {
        super(SessionConfig.class);
        this.redisProperties = redisProperties;
    }

    @Bean
    public RedisConnectionFactory connectionFactory() {

        RedisSentinelConfiguration sentinelConfiguration = new RedisSentinelConfiguration()
                .master(redisProperties.getSentinel().getMaster());
        redisProperties.getSentinel().getNodes().forEach(s -> {
            String[] ipSet = s.split(":");
            sentinelConfiguration.sentinel(ipSet[0], Integer.parseInt(ipSet[1]));
        });
        sentinelConfiguration.setPassword(redisProperties.getPassword().isEmpty() ? RedisPassword.none() : RedisPassword.of(redisProperties.getPassword()));

        return new LettuceConnectionFactory(sentinelConfiguration, LettuceClientConfiguration.defaultConfiguration());

    }
}

참고:

https://dltlabs.medium.com/how-to-setup-redis-replication-c9cc89ba6c03

https://medium.com/trendyol-tech/high-availability-with-redis-sentinel-and-spring-lettuce-client-9da40525fc82

https://crystalcube.co.kr/176

https://jojoldu.tistory.com/418

댓글남기기