From ce210cdee307024d8c1f227637ab79fcdf56e631 Mon Sep 17 00:00:00 2001 From: fortrue Date: Wed, 28 Mar 2018 20:27:04 +0800 Subject: [PATCH] 1.modify default ServerTimeout and KeepAlive to 0 2.improve doc --- conf/cluster.conf | 4 ++-- conf/sentinel.conf | 4 ++-- doc/config_CN.md | 28 +++++++++++++++++++--------- src/Conf.h | 4 ++-- 4 files changed, 25 insertions(+), 15 deletions(-) diff --git a/conf/cluster.conf b/conf/cluster.conf index 0e33722..6f77385 100644 --- a/conf/cluster.conf +++ b/conf/cluster.conf @@ -6,10 +6,10 @@ ## [StaticSlaveReadPriority [0-100]] #default 0 ## [DynamicSlaveReadPriority [0-100]] #default 0 ## [RefreshInterval number[s|ms|us]] #default 1, means 1 second -## [ServerTimeout number[s|ms|us]] #default 1, server connection socket read/write timeout +## [ServerTimeout number[s|ms|us]] #default 0, server connection socket read/write timeout ## [ServerFailureLimit number] #default 10 ## [ServerRetryTimeout number[s|ms|us]] #default 1 -## [KeepAlive seconds] #default 120, server connection tcp keepalive +## [KeepAlive seconds] #default 0, server connection tcp keepalive ## Servers { ## + addr diff --git a/conf/sentinel.conf b/conf/sentinel.conf index 207f9b2..2975cb5 100644 --- a/conf/sentinel.conf +++ b/conf/sentinel.conf @@ -10,10 +10,10 @@ ## [StaticSlaveReadPriority [0-100]] #default 0 ## [DynamicSlaveReadPriority [0-100]] #default 0 ## [RefreshInterval number[s|ms|us]] #default 1, means 1 second -## [ServerTimeout number[s|ms|us]] #default 1, server connection socket read/write timeout +## [ServerTimeout number[s|ms|us]] #default 0, server connection socket read/write timeout ## [ServerFailureLimit number] #default 10 ## [ServerRetryTimeout number[s|ms|us]] #default 1 -## [KeepAlive seconds] #default 120, server connection tcp keepalive +## [KeepAlive seconds] #default 0, server connection tcp keepalive ## Sentinels { ## + addr ## ... diff --git a/doc/config_CN.md b/doc/config_CN.md index e793eba..71a6933 100644 --- a/doc/config_CN.md +++ b/doc/config_CN.md @@ -222,9 +222,11 @@ predixy支持Redis Sentinel和Redis Cluster来使用redis，一个配置里这 [MasterReadPriority [0-100]] [StaticSlaveReadPriority [0-100]] [DynamicSlaveReadPriority [0-100]] - [RefreshInterval seconds] + [RefreshInterval number[s|ms|us]] + [ServerTimeout number[s|ms|us]] [ServerFailureLimit number] - [ServerRetryTimeout seconds] + [ServerRetryTimeout number[s|ms|us]] + [KeepAlive seconds] Sentinels { + addr ... @@ -242,12 +244,14 @@ predixy支持Redis Sentinel和Redis Cluster来使用redis，一个配置里这 + Hash: 指定对key算哈希的方法，当前只支持atol和crc16 + HashTag: 指定哈希标签，不指定的话为{} + Distribution: 指定分布key的方法，当前只支持modula和random -+ MasterReadPriority: 读写分离功能，从redis master节点执行读请求的优先级，为0则禁止读redis master -+ StaticSlaveReadPriority: 读写分离功能，从静态redis slave节点执行读请求的优先级，所谓静态节点，是指在本配置文件中显示列出的redis节点 -+ DynamicSlaveReadPolicy: 功能见上，所谓动态节点是指在本配置文件中没有列出，但是通过redis sentinel动态发现的节点 -+ RefreshInterval: predixy会周期性的请求redis sentinel以获取最新的集群信息，该参数以秒为单位指定刷新周期 -+ ServerFailureLimit: 一个redis实例出现多少次才错误以后将其标记为失效 -+ ServerRetryTimeout: 一个redis实例失效后多久后去检查其是否恢复正常 ++ MasterReadPriority: 读写分离功能，从redis master节点执行读请求的优先级，为0则禁止读redis master，不指定的话为50 ++ StaticSlaveReadPriority: 读写分离功能，从静态redis slave节点执行读请求的优先级，所谓静态节点，是指在本配置文件中显示列出的redis节点，不指定的话为0 ++ DynamicSlaveReadPolicy: 功能见上，所谓动态节点是指在本配置文件中没有列出，但是通过redis sentinel动态发现的节点，不指定的话为0 ++ RefreshInterval: predixy会周期性的请求redis sentinel以获取最新的集群信息，该参数以秒为单位指定刷新周期，不指定的话为1秒 ++ ServerTimeout: 请求在predixy中最长的处理/等待时间，如果超过该时间redis还没有响应的话，那么predixy会关闭同redis的连接，并给客户端一个错误响应，对于blpop这种阻塞式命令，该选项不起作用，为0则禁止此功能，即如果redis不返回就一直等待，不指定的话为0 ++ ServerFailureLimit: 一个redis实例出现多少次才错误以后将其标记为失效，不指定的话为10 ++ ServerRetryTimeout: 一个redis实例失效后多久后去检查其是否恢复正常，不指定的话为1秒 ++ KeepAlive: predixy与redis的连接tcp keepalive时间，为0则禁止此功能，不指定的话为0 + Sentinels: 里面定义redis sentinel实例的地址 + Group: 定义一个redis组，Group的名字应该和redis sentinel里面的名字一致，Group里可以显示列出redis的地址，列出的话就是上面提到的静态节点 @@ -262,8 +266,10 @@ predixy支持Redis Sentinel和Redis Cluster来使用redis，一个配置里这 StaticSlaveReadPriority 50 DynamicSlaveReadPriority 50 RefreshInterval 1 + ServerTimeout 1 ServerFailureLimit 10 ServerRetryTimeout 1 + KeepAlive 120 Sentinels { + 10.2.2.2:7500 + 10.2.2.3:7500 @@ -287,8 +293,10 @@ predixy支持Redis Sentinel和Redis Cluster来使用redis，一个配置里这 [StaticSlaveReadPriority [0-100]] [DynamicSlaveReadPriority [0-100]] [RefreshInterval seconds] + [ServerTimeout number[s|ms|us]] [ServerFailureLimit number] - [ServerRetryTimeout seconds] + [ServerRetryTimeout number[s|ms|us]] + [KeepAlive seconds] Servers { + addr ... @@ -308,8 +316,10 @@ predixy支持Redis Sentinel和Redis Cluster来使用redis，一个配置里这 StaticSlaveReadPriority 50 DynamicSlaveReadPriority 50 RefreshInterval 1 + ServerTimeout 1 ServerFailureLimit 10 ServerRetryTimeout 1 + KeepAlive 120 Servers { + 192.168.2.107:2211 + 192.168.2.107:2212 diff --git a/src/Conf.h b/src/Conf.h index 98cf2c4..04e9a17 100644 --- a/src/Conf.h +++ b/src/Conf.h @@ -50,10 +50,10 @@ struct ServerPoolConf int staticSlaveReadPriority = 0; int dynamicSlaveReadPriority = 0; long refreshInterval = 1000000; //us - long serverTimeout = 1000000; //us + long serverTimeout = 0; //us int serverFailureLimit = 10; long serverRetryTimeout = 1000000; //us - int keepalive = 120; //seconds + int keepalive = 0; //seconds int databases = 1; };