Nagios监控生产环境redis集群服务实战

前言： 以前做了cacti上展示redis性能报表图，可以看到redis的性能变化趋势图， 但是还缺了实时报警通知的功能，现在补上这一环节。在redis服务瓶颈或者异常时候即使报警通知，方便dba第一时间处理维护。 1，下载redis监控插件 Redis已经在服务器安装好了，

前言： 以前做了cacti上展示redis性能报表图，可以看到redis的性能变化趋势图，但是还缺了实时报警通知的功能，现在补上这一环节。在redis服务瓶颈或者异常时候即使报警通知，方便dba第一时间处理维护。
1，下载redis监控插件

Redis已经在服务器安装好了，所以直接可以进行监控，redis集群安装请参考：http://blog.itpub.net/26230597/viewspace-1145831/，下载地址为：http://download.csdn.net/detail/mchdba/8023351，有2个版本，一个是perl脚本写成的，一个是php脚本写成的，可以任意选择一个，这里选择的是perl脚本。

2，赋予执行权限

将check_redis.php和check_redis.pl复制到/usr/lib/nagios/plugins/目录，然后赋予执行权限，

[root@wgq_41 plugins]# cd /usr/lib/nagios/plugins/

[root@wgq_41 plugins]# chown -R nagios.nagios check_redis.*

[root@wgq_41 plugins]# chmod 750 check_redis.*

3，定义监控命令

[root@wgq objects] vim /usr/local/nagios/etc/objects/commands.cfg

# add by tim on 20141010,for redis

# check redis

define command {

command_name check_redis

command_line /usr/lib/nagios/plugins/check_redis.pl -H $HOSTADDRESS$ -p $ARG1$ -a $ARG2$ -w $ARG3$ -c $ARG4$ -f

}

4，定义redis监控主机

[root@wgq etc]# vim /usr/local/nagios/etc/hosts.cfg

# No.018，redis master server

define host{

use linux-server

host_name cache-1

alias cache-1

address 10.xxx.3.x0

check_command check-host-alive

max_check_attempts 5

check_period 24x7

contact_groups ops

notification_interval 30

notification_period 24x7

notification_options d,u,r

}

# No.020 cache-3 redis slave server

define host{

use linux-server

host_name cache-3

alias cache-3

address 10.xx.3.x2

check_command check-host-alive

max_check_attempts 5

check_period 24x7

contact_groups ops

notification_interval 30

notification_period 24x7

notification_options d,u,r

}

5，定义redis监控主机组

define hostgroup {

hostgroup_name Redis_Servers

alias Redisservices

members cache-1,cache-2

}

6，定义redis监控服务选项

[root@wgq objects]# vim /usr/local/nagios/etc/objects/services_redis.cfg

# Redis Master 监控选项

define service {

host_name cache-1

servicegroups Redisservices

service_description Redis Master Clients

check_command check_redis!6379!'connected_clients,blocked_clients,client_longest_output_list,client_biggest_input_buf'!200,50,~,~!600,150,~,~

max_check_attempts 5

normal_check_interval 3

retry_check_interval 2

check_period 24x7

notification_interval 10

notification_period 24x7

notification_options w,u,c,r

contact_groups ops

}

define service {

host_name cache-1

servicegroups Redisservices

service_description Redis Master Memory

check_command check_redis!6379!'used_memory_human,used_memory_peak_human'!~,~!~,~

max_check_attempts 5

normal_check_interval 3

retry_check_interval 2

check_period 24x7

notification_interval 10

notification_period 24x7

notification_options w,u,c,r

contact_groups ops

}

define service {

host_name cache-1

servicegroups Redisservices

service_description Redis Master CPU

check_command check_redis!6379!'used_cpu_sys,used_cpu_user,used_cpu_sys_children,used_cpu_user_children'!~,~,~,~!~,~,~,~ ; #未定义监控报警阀值

max_check_attempts 5

normal_check_interval 3

retry_check_interval 2

check_period 24x7

notification_interval 10

notification_period 24x7

notification_options w,u,c,r

contact_groups ops

}

# Redis Slave 监控选项

define service {

host_name cache-3

servicegroups Redisservices

service_description Redis Slave Clients

check_command check_redis!6379!'connected_clients,blocked_clients,client_longest_output_list,client_biggest_input_buf'!200,50,~,~!600,150,~,~

max_check_attempts 5

normal_check_interval 3

retry_check_interval 2

check_period 24x7

notification_interval 10

notification_period 24x7

notification_options w,u,c,r

contact_groups ops

}

define service {

host_name cache-3

servicegroups Redisservices

service_description Redis Slave Memory

check_command check_redis!6379!'used_memory_human,used_memory_peak_human'!~,~!~,~

max_check_attempts 5

normal_check_interval 3

retry_check_interval 2

check_period 24x7

notification_interval 10

notification_period 24x7

notification_options w,u,c,r

contact_groups ops

}

define service {

host_name cache-3

servicegroups Redisservices

service_description Redis Slave CPU

check_command check_redis!6379!'used_cpu_sys,used_cpu_user,used_cpu_sys_children,used_cpu_user_children'!~,~,~,~!~,~,~,~ ; #未定义监控报警阀值

max_check_attempts 5

normal_check_interval 3

retry_check_interval 2

check_period 24x7

notification_interval 10

notification_period 24x7

notification_options w,u,c,r

contact_groups ops

}

赋予nagios用户执行权限

[root@wgq objects]# chown -R nagios.nagios services_redis.cfg

[root@wgq objects]# chmod 777 services_redis.cfg

添加监控服务项到nagios.cfg

[root@wgq etc]# vim /usr/local/nagios/etc/nagios.cfg

cfg_file=/usr/local/nagios/etc/objects/services_redis.cfg

7，测试redis监控服务

执行命令/usr/lib/nagios/plugins/check_redis.pl -H cache-1 -a 'connected_clients,blocked_clients' -w ~,~ -c ~,~ -m -M 4G -A -R -T 来测试下redis监控是否正常运行

[root@wgq plugins]# /usr/lib/nagios/plugins/check_redis.pl -H 10.2xx.3.x0 -a 'connected_clients,blocked_clients' -w ~,~ -c ~,~ -m -M 4G -A -R -T

OK: REDIS 2.8.8 on 10.2xx.3.x0:6379 has 1 databases (db0) with 28497 keys, up 76 days 2 hours - response in 0.004s, hitrate is 12.83%, memory use is 194.14M (peak 205.14M, 6.49% of max, fragmentation 1.37%), connected_clients is 35, blocked_clients is 11 | redis_build_id=d322d411218ade61 total_connections_received=341191c used_memory_lua=33792 aof_rewrite_buffer_length=0 used_memory_rss=278749184B redis_git_dirty=0 loading=0 redis_mode=standalone latest_fork_usec=5588 repl_backlog_first_byte_offset=0 sync_partial_ok=0 master_repl_offset=0 uptime_in_days=76c aof_rewrite_scheduled=0 lru_clock=3649276 rdb_bgsave_in_progress=0 rejected_connections=0 repl_backlog_active=0 aof_delayed_fsync=1 sync_full=0 process_id=7776 used_memory_human=194.14M aof_current_rewrite_time_sec=-1 used_memory=203570960 aof_enabled=1 blocked_clients=11 aof_last_bgrewrite_status=ok aof_rewrite_in_progress=0 sync_partial_err=0 used_cpu_sys_children=2222.75 connected_slaves=0 repl_backlog_histlen=0 uptime_in_seconds=6576292c repl_backlog_size=1048576 os=Linux 2.6.32-358.el6.x86_64 x86_64 used_cpu_sys=32640.80 aof_pending_bio_fsync=0 connected_clients=35 rdb_last_bgsave_time_sec=1 used_memory_peak_human=205.14M run_id=d1fc098d26fa4bbcef3eabeec6d19a858f03dd00 rdb_last_bgsave_status=ok pubsub_patterns=8 client_biggest_input_buf=0 keyspace_hits=42175896c rdb_last_save_time=1412935342 rdb_changes_since_last_save=318 db0_keys=28497 db0_expires=7 db0_avg_ttl=34003 aof_pending_rewrite=0 aof_buffer_length=0 config_file=/usr/local/redis-2.8.8/etc/redis.conf pubsub_channels=0 used_cpu_user_children=21375.34 hz=10 aof_last_rewrite_time_sec=2 aof_last_write_status=ok aof_base_size=82883253 used_cpu_user=18460.42 keyspace_misses=286602797c tcp_port=6379 total_commands_processed=797581196c mem_fragmentation_ratio=1.37 aof_current_size=146485850 rdb_current_bgsave_time_sec=-1 client_longest_output_list=0 instantaneous_ops_per_sec=114 evicted_keys=0c used_memory_peak=215106272B expired_keys=58977c total_keys=28497 total_expires=7 response_time=0.003802s hitrate=12.8281% memory_utilization=6.49013519287109%

[root@wgq plugins]#

8，查看redis监控服务状态

先重新加载nagios，使刚添加的redis监控配置生效

[root@wgq objects]# service nagios reload

Running configuration check...

Reloading nagios configuration...

done

[root@wgq objects]#

redis监控服务界面，如下图所示：

9，操作过程中的报错处理过程

报错：

[root@wgq_line_cache_3_41 plugins]# ./check_redis.pl --help

Can't locate Redis.pm in @INC (@INC contains: /usr/local/lib64/perl5 /usr/local/share/perl5 /usr/lib64/perl5/vendor_perl /usr/share/perl5/vendor_perl /usr/lib64/perl5 /usr/share/perl5 .) at ./check_redis.pl line 421.

BEGIN failed--compilation aborted at ./check_redis.pl line 421.

[root@wgq_line_cache_3_41 plugins]#

[root@wgq_line_cache_3_41 plugins]# perl -MCPAN -e shell

Terminal does not support AddHistory.

cpan shell -- CPAN exploration and modules installation (v1.9402)

Enter 'h' for help.

cpan[1]> install Redis

…

Can't locate Module/Build/Tiny.pm in @INC (@INC contains: /usr/local/lib64/perl5 /usr/local/share/perl5 /usr/lib64/perl5/vendor_perl /usr/share/perl5/vendor_perl /usr/lib64/perl5 /usr/share/perl5 .) at Build.PL line 2.

BEGIN failed--compilation aborted at Build.PL line 2.

Warning: No success on command[/usr/bin/perl Build.PL --installdirs site]

Warning (usually harmless): 'YAML' not installed, will not store persistent state

DAMS/Redis-1.976.tar.gz

/usr/bin/perl Build.PL --installdirs site -- NOT OK

Running Build test

Make had some problems, won't test

Running Build install

Make had some problems, won't install

Could not read '/root/.cpan/build/Redis-1.976-Zhz6xI/META.yml'. Falling back to other methods to determine prerequisites……

YAML是以数据为*的标记语言，其使用ASCII码（如连字符、问号、冒号、逗号等）构造数据块（标量值或哈希码）。和XML相同，YAML也是一种机器可识别语言，并能和多种脚本语言相结合，其中一种便是Perl，需要安装YAML，如下执行：

cpan[2]>install YAML

……

Appending installation info to /usr/lib64/perl5/perllocal.pod

INGY/YAML-1.12.tar.gz

/usr/bin/make install -- OK

CPAN: YAML loaded ok (v1.12)

PS：这里可能会安装失败，失败原因是网络连接，可以多执行几次install YAML就会成功。

再继续执行install Redis，有如下提示信息

cpan[4]> install Redis

Running install for module 'Redis'

Running Build for D/DA/DAMS/Redis-1.976.tar.gz

Has already been unwrapped into directory /root/.cpan/build/Redis-1.976-cUL4rt

'/usr/bin/perl Build.PL --installdirs site' returned status 512, won't make

Running Build test

Make had some problems, won't test

Running Build install

Make had some problems, won't install

cpan[5]>

Build失败，Build.PL故障了，需要重新安装下执行命令install Build

cpan[5]> install Build

成功后，再执行install Redis

cpan[6]> install Redis

Redis安装执行成功。

参考文档：http://exchange.nagios.org/directory/Plugins/Databases/check_redis-2Epl/details