军哥，反馈一个关于redis的一个奇怪的问题

ffspace810

2016年9月在centos 6.5 x64中部署了lnmp1.3版，开启了redis（默认安装，没有修改相关配置），线上业务是discuz。从今年1月6号开始，磁盘IO写突然变得不稳定，可以看一下附件里的监控截图。在iotop里发现是redis的IO写出现问题，只要把discuz配置文件中redis的地址删除，IO写就稳定了。现在不知道为什么会这样，我把redis日志功能打开了，正在观察中。这个问题军哥知道该怎样排查吗？求指点




[ 本帖最后由 ffspace810 于 2017-1-12 12:04 编辑 ]

ffspace810

开启redis日志后，重启redis时日志如下：

2207:M 12 Jan 11:41:33.892 # WARNING: The TCP backlog setting of 511 cannot be enforced because /proc/sys/net/core/somaxconn is set to the lower value of 128.
2207:M 12 Jan 11:41:33.892 # Server started, Redis version 3.2.0
2207:M 12 Jan 11:41:33.892 # 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.
2207:M 12 Jan 11:41:33.892 # 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.
2207:M 12 Jan 11:41:33.894 * DB loaded from disk: 0.001 seconds
2207:M 12 Jan 11:41:33.894 * The server is now ready to accept connections on port 6379

licess

redis不大了解，可以按日志里的提示修改里面的参数试试

ffspace810

WARNING的部分改好以后还是不行，已经停止redis服务了。。。改用memcache