k8s部署rocketmq 双主
由于apache 官网的 docker image 是单点，要实现集群方式部署。
rocketmq 分为 nameserver 和 broker ， 对于之间调用频繁的服务，会增加网络压力， 所以 考虑又网络通信 变成 进程间通信， 那么1个pod 中包含两个container。
关于rocketmq 下载编译， https://www.cnblogs.com/fengjian2016/p/10150394.html
拿到 编译后的 apache-rocketmq , 需要修改一些配置， 例如：
. rm -rf 删除 rocketmq/bin/*.cmd
2 .把runserver.sh HEAP_OPTS 变成环境变量， 把runbroker.sh HEAP_OPTS 变成环境变量
3. 指定消息存储路径
[root@rocketmq1 conf]#mkdir -p /data/rocketmq/store
[root@rocketmq1 conf]#mkdir -p /data/rocketmq/store/commitlog
[root@rocketmq1 conf]#mkdir -p /data/rocketmq/logs
4. 指定日志输出路径
/data/rocketmq/conf/ 中的xml文件
sed -i 's#${user.home}#/data/rocketmq/logs#g' *.xml
5. 删除多余配置文件 2m-2s 2master 等等， 启动的时候k8s command 直接指向文件
/data/rocketmq/conf/ 目录下新建 broker-a.conf broker-b.conf 2个文件 6. 配置修改
(1) rocketmqHome 修改
(2) namesrvAddr 修改，强调 通一个pod 必须写localhost 或者 127.0.0.1 要不然调不通
[root@harbor conf]# cat broker-a.conf
rocketmqHome=/data/rocketmq
namesrvAddr=localhost:;rocketmq2:
brokerName=broker-a
brokerClusterName=DefaultCluster
brokerId=
brokerPermission=
#在发送消息时，自动创建服务器不存在的topic，默认创建的队列数
defaultTopicQueueNums=
#是否允许 Broker 自动创建Topic，建议线下开启，线上关闭
autoCreateTopicEnable=true
clusterTopicEnable=true
brokerTopicEnable=true
#是否允许 Broker 自动创建订阅组，建议线下开启，线上关闭
autoCreateSubscriptionGroup=true
sendMessageThreadPoolNums=
pullMessageThreadPoolNums=
queryMessageThreadPoolNums=
adminBrokerThreadPoolNums=
clientManageThreadPoolNums=
consumerManageThreadPoolNums=
heartbeatThreadPoolNums=
endTransactionThreadPoolNums=
flushConsumerOffsetInterval=
flushConsumerOffsetHistoryInterval=
rejectTransactionMessage=false
fetchNamesrvAddrByAddressServer=false
sendThreadPoolQueueCapacity=
pullThreadPoolQueueCapacity=
queryThreadPoolQueueCapacity=
clientManagerThreadPoolQueueCapacity=
consumerManagerThreadPoolQueueCapacity=
heartbeatThreadPoolQueueCapacity=
endTransactionPoolQueueCapacity=
filterServerNums=
longPollingEnable=true
shortPollingTimeMills=
notifyConsumerIdsChangedEnable=true
highSpeedMode=false
commercialEnable=true
commercialTimerCount=
commercialTransCount=
commercialBigCount=
commercialBaseCount=
transferMsgByHeap=true
maxDelayTime=
regionId=DefaultRegion
registerBrokerTimeoutMills=
slaveReadEnable=false
disableConsumeIfConsumerReadSlowly=false
consumerFallbehindThreshold=
brokerFastFailureEnable=true
waitTimeMillsInSendQueue=
waitTimeMillsInPullQueue=
waitTimeMillsInHeartbeatQueue=
waitTimeMillsInTransactionQueue=
startAcceptSendRequestTimeStamp=
traceOn=true
enableCalcFilterBitMap=false
expectConsumerNumUseFilter=
maxErrorRateOfBloomFilter=
filterDataCleanTimeSpan=
filterSupportRetry=false
enablePropertyFilter=false
compressedRegister=false
forceRegister=true
registerNameServerPeriod=
transactionTimeOut=
transactionCheckMax=
transactionCheckInterval=
#Broker 对外服务的监听端口
listenPort=
serverWorkerThreads=
serverCallbackExecutorThreads=
serverSelectorThreads=
serverOnewaySemaphoreValue=
serverAsyncSemaphoreValue=
serverChannelMaxIdleTimeSeconds=
serverSocketSndBufSize=
serverSocketRcvBufSize=
serverPooledByteBufAllocatorEnable=true
useEpollNativeSelector=false
clientWorkerThreads=
clientCallbackExecutorThreads=
clientOnewaySemaphoreValue=
clientAsyncSemaphoreValue=
connectTimeoutMillis=
channelNotActiveInterval=
clientChannelMaxIdleTimeSeconds=
clientSocketSndBufSize=
clientSocketRcvBufSize=
clientPooledByteBufAllocatorEnable=false
clientCloseSocketIfTimeout=false
useTLS=false
storePathRootDir=/data/rocketmq/store
storePathCommitLog=/root/rocketmq/store/commitlog
#commitLog每个文件的大小默认1G
mapedFileSizeCommitLog=
#ConsumeQueue每个文件默认存60W条，根据业务情况调整,
mapedFileSizeConsumeQueue=
enableConsumeQueueExt=false
mappedFileSizeConsumeQueueExt=
bitMapLengthConsumeQueueExt=
flushIntervalCommitLog=
commitIntervalCommitLog=
useReentrantLockWhenPutMessage=false
flushCommitLogTimed=false
flushIntervalConsumeQueue=
cleanResourceInterval=
deleteCommitLogFilesInterval=
deleteConsumeQueueFilesInterval=
destroyMapedFileIntervalForcibly=
redeleteHangedFileInterval=
#删除文件时间点，默认凌晨 2点
deleteWhen=
#检测物理文件磁盘空间
diskMaxUsedSpaceRatio=
#文件保留时间，默认 小时
fileReservedTime=
putMsgIndexHightWater=
#限制的消息大小
maxMessageSize=
checkCRCOnRecover=true
flushCommitLogLeastPages=
commitCommitLogLeastPages=
flushLeastPagesWhenWarmMapedFile=
flushConsumeQueueLeastPages=
flushCommitLogThoroughInterval=
commitCommitLogThoroughInterval=
flushConsumeQueueThoroughInterval=
maxTransferBytesOnMessageInMemory=
maxTransferCountOnMessageInMemory=
maxTransferBytesOnMessageInDisk=
maxTransferCountOnMessageInDisk=
accessMessageInMemoryMaxRatio=
messageIndexEnable=true
maxHashSlotNum=
maxIndexNum=
maxMsgsNumBatch=
messageIndexSafe=false
haListenPort=
haSendHeartbeatInterval=
haHousekeepingInterval=
haTransferBatchSize=
haMasterAddress=
haSlaveFallbehindMax=
brokerRole=ASYNC_MASTER
flushDiskType=ASYNC_FLUSH
syncFlushTimeout=
messageDelayLevel=1s 5s 10s 30s 1m 2m 3m 4m 5m 6m 7m 8m 9m 10m 20m 30m 1h 2h
flushDelayOffsetInterval=
cleanFileForciblyEnable=true
warmMapedFileEnable=false
offsetCheckInSlave=false
debugLockEnable=false
duplicationEnable=false
diskFallRecorded=true
osPageCacheBusyTimeOutMills=
defaultQueryMaxNum=
transientStorePoolEnable=false
transientStorePoolSize=
fastFailIfNoBufferInStorePool=false
[root@harbor conf]# cat broker-b.conf
rocketmqHome=/data/rocketmq
namesrvAddr=rocketmq1:;localhost:
brokerName=broker-b
brokerClusterName=DefaultCluster
brokerId=
brokerPermission=
#在发送消息时，自动创建服务器不存在的topic，默认创建的队列数
defaultTopicQueueNums=
#是否允许 Broker 自动创建Topic，建议线下开启，线上关闭
autoCreateTopicEnable=true
clusterTopicEnable=true
brokerTopicEnable=true
#是否允许 Broker 自动创建订阅组，建议线下开启，线上关闭
autoCreateSubscriptionGroup=true
sendMessageThreadPoolNums=
pullMessageThreadPoolNums=
queryMessageThreadPoolNums=
adminBrokerThreadPoolNums=
clientManageThreadPoolNums=
consumerManageThreadPoolNums=
heartbeatThreadPoolNums=
endTransactionThreadPoolNums=
flushConsumerOffsetInterval=
flushConsumerOffsetHistoryInterval=
rejectTransactionMessage=false
fetchNamesrvAddrByAddressServer=false
sendThreadPoolQueueCapacity=
pullThreadPoolQueueCapacity=
queryThreadPoolQueueCapacity=
clientManagerThreadPoolQueueCapacity=
consumerManagerThreadPoolQueueCapacity=
heartbeatThreadPoolQueueCapacity=
endTransactionPoolQueueCapacity=
filterServerNums=
longPollingEnable=true
shortPollingTimeMills=
notifyConsumerIdsChangedEnable=true
highSpeedMode=false
commercialEnable=true
commercialTimerCount=
commercialTransCount=
commercialBigCount=
commercialBaseCount=
transferMsgByHeap=true
maxDelayTime=
regionId=DefaultRegion
registerBrokerTimeoutMills=
slaveReadEnable=false
disableConsumeIfConsumerReadSlowly=false
consumerFallbehindThreshold=
brokerFastFailureEnable=true
waitTimeMillsInSendQueue=
waitTimeMillsInPullQueue=
waitTimeMillsInHeartbeatQueue=
waitTimeMillsInTransactionQueue=
startAcceptSendRequestTimeStamp=
traceOn=true
enableCalcFilterBitMap=false
expectConsumerNumUseFilter=
maxErrorRateOfBloomFilter=
filterDataCleanTimeSpan=
filterSupportRetry=false
enablePropertyFilter=false
compressedRegister=false
forceRegister=true
registerNameServerPeriod=
transactionTimeOut=
transactionCheckMax=
transactionCheckInterval=
#Broker 对外服务的监听端口
listenPort=
serverWorkerThreads=
serverCallbackExecutorThreads=
serverSelectorThreads=
serverOnewaySemaphoreValue=
serverAsyncSemaphoreValue=
serverChannelMaxIdleTimeSeconds=
serverSocketSndBufSize=
serverSocketRcvBufSize=
serverPooledByteBufAllocatorEnable=true
useEpollNativeSelector=false
clientWorkerThreads=
clientCallbackExecutorThreads=
clientOnewaySemaphoreValue=
clientAsyncSemaphoreValue=
connectTimeoutMillis=
channelNotActiveInterval=
clientChannelMaxIdleTimeSeconds=
clientSocketSndBufSize=
clientSocketRcvBufSize=
clientPooledByteBufAllocatorEnable=false
clientCloseSocketIfTimeout=false
useTLS=false
storePathRootDir=/data/rocketmq/store
storePathCommitLog=/root/rocketmq/store/commitlog
#commitLog每个文件的大小默认1G
mapedFileSizeCommitLog=
#ConsumeQueue每个文件默认存60W条，根据业务情况调整,
mapedFileSizeConsumeQueue=
enableConsumeQueueExt=false
mappedFileSizeConsumeQueueExt=
bitMapLengthConsumeQueueExt=
flushIntervalCommitLog=
commitIntervalCommitLog=
useReentrantLockWhenPutMessage=false
flushCommitLogTimed=false
flushIntervalConsumeQueue=
cleanResourceInterval=
deleteCommitLogFilesInterval=
deleteConsumeQueueFilesInterval=
destroyMapedFileIntervalForcibly=
redeleteHangedFileInterval=
#删除文件时间点，默认凌晨 2点
deleteWhen=
#检测物理文件磁盘空间
diskMaxUsedSpaceRatio=
#文件保留时间，默认 小时
fileReservedTime=
putMsgIndexHightWater=
#限制的消息大小
maxMessageSize=
checkCRCOnRecover=true
flushCommitLogLeastPages=
commitCommitLogLeastPages=
flushLeastPagesWhenWarmMapedFile=
flushConsumeQueueLeastPages=
flushCommitLogThoroughInterval=
commitCommitLogThoroughInterval=
flushConsumeQueueThoroughInterval=
maxTransferBytesOnMessageInMemory=
maxTransferCountOnMessageInMemory=
maxTransferBytesOnMessageInDisk=
maxTransferCountOnMessageInDisk=
accessMessageInMemoryMaxRatio=
messageIndexEnable=true
maxHashSlotNum=
maxIndexNum=
maxMsgsNumBatch=
messageIndexSafe=false
haListenPort=
haSendHeartbeatInterval=
haHousekeepingInterval=
haTransferBatchSize=
haMasterAddress=
haSlaveFallbehindMax=
brokerRole=ASYNC_MASTER
flushDiskType=ASYNC_FLUSH
syncFlushTimeout=
messageDelayLevel=1s 5s 10s 30s 1m 2m 3m 4m 5m 6m 7m 8m 9m 10m 20m 30m 1h 2h
flushDelayOffsetInterval=
cleanFileForciblyEnable=true
warmMapedFileEnable=false
offsetCheckInSlave=false
debugLockEnable=false
duplicationEnable=false
diskFallRecorded=true
osPageCacheBusyTimeOutMills=
defaultQueryMaxNum=
transientStorePoolEnable=false
transientStorePoolSize=
fastFailIfNoBufferInStorePool=false
broker Dockerfile
[root@harbor rocketmq-broker]# cat Dockerfile
FROM centos MAINTAINER fengjian <fengjian@senyint.com> ENV TZ "Asia/Shanghai"
ENV TERM xterm
ENV MALLOC_ARENA_MAX=
ENV JAVA_HOME /data/jdk
ENV CLASSPATH $JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar
ENV CATALINA_HOME /data/tomcat
ENV CATALINA_BASE /data/tomcat
ENV PATH $PATH:$JAVA_HOME/bin:$CATALINA_HOME/lib:$CATALINA_HOME/bin
ENV ROCKETMQ_HOME /data/rocketmq RUN rm /etc/yum.repos.d/* -rf
ADD Centos-7.repo /etc/yum.repos.d/
ADD epel.repo /etc/yum.repos.d/
RUN yum clean all
RUN yum -y install net-tools bind-utils sysstat ADD localtime /etc/
ADD jdk.tar.gz /data
ADD profile /etc
ADD sysctl.conf /etc
ADD 20-nproc.conf /etc/security/limits.d/
ADD limits.conf /etc/security/ ADD rocketmq.tar.gz /data/
WORKDIR ${ROCKETMQ_HOME}/bin EXPOSE 10909 10911
nameserver Dockerfile
FROM centos MAINTAINER fengjian <fengjian@senyint.com> ENV TZ "Asia/Shanghai"
ENV TERM xterm
ENV MALLOC_ARENA_MAX=
ENV JAVA_HOME /data/jdk
ENV CLASSPATH $JAVA_HOME/lib/dt.jar:$JAVA_HOME/lib/tools.jar
ENV CATALINA_HOME /data/tomcat
ENV CATALINA_BASE /data/tomcat
ENV PATH $PATH:$JAVA_HOME/bin:$CATALINA_HOME/lib:$CATALINA_HOME/bin
ENV ROCKETMQ_HOME /data/rocketmq RUN rm /etc/yum.repos.d/* -rf
ADD Centos-7.repo /etc/yum.repos.d/
ADD epel.repo /etc/yum.repos.d/
RUN yum clean all
RUN yum -y install net-tools bind-utils sysstat ADD localtime /etc/
ADD jdk.tar.gz /data
ADD profile /etc
ADD sysctl.conf /etc
ADD 20-nproc.conf /etc/security/limits.d/
ADD limits.conf /etc/security/ ADD rocketmq.tar.gz /data/
WORKDIR ${ROCKETMQ_HOME}/bin EXPOSE 9876 ENTRYPOINT ["sh","mqnamesrv"]
cat ceph-rocketmq1-pvc.yaml
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
name: rocketmq1-brokerstore-pvc
annotations:
volume.beta.kubernetes.io/storage-class: ceph-rbd-rocketmq
spec:
accessModes:
- ReadWriteOnce #必须写ReadWriteOnce,否则报错。
resources:
requests:
storage: 20Gi
---
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
name: rocketmq1-brokerlog-pvc
annotations:
volume.beta.kubernetes.io/storage-class: ceph-rbd-rocketmq
spec:
accessModes:
- ReadWriteOnce #必须写ReadWriteOnce,否则报错。
resources:
requests:
storage: 20Gi
---
kind: PersistentVolumeClaim
apiVersion: v1
metadata:
name: rocketmq1-namesrvlog-pvc
annotations:
volume.beta.kubernetes.io/storage-class: ceph-rbd-rocketmq
spec:
accessModes:
- ReadWriteOnce #必须写ReadWriteOnce,否则报错。
resources:
requests:
storage: 20Gi
rocketmq1 yaml文件
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
name: rocketmq1
spec:
replicas:
template:
metadata:
labels:
app: rocketmq1
spec:
containers:
- name: namesrv
image: 192.168.200.10/fengjian/rocketmq-nameserver:4.3
imagePullPolicy: Always
ports:
- containerPort:
env:
- name: ROCKETMQ_HEAP_OPTS
value : "-Xms1g -Xmx1g -Xmn512m"
volumeMounts:
- name: rocketmq1namesrvlog
mountPath: /data/rocketmq/logs
- name: broker
image: 192.168.200.10/fengjian/rocketmq-broker:4.3
imagePullPolicy: Always
ports:
- name: brockerserver
containerPort:
- name: brockerslave
containerPort:
command: ["/data/rocketmq/bin/mqbroker"]
args: ["-c", "/data/rocketmq/conf/broker-a.conf"]
env:
- name: ROCKETMQ_HEAP_OPTS
value : "-Xms2g -Xmx2g -Xmn1g"
volumeMounts:
- name: brokerlogs
mountPath: /data/rocketmq/logs
- name: brokerstore
mountPath: /data/rocketmq/store
volumes:
- name: rocketmq1namesrvlog
persistentVolumeClaim:
claimName: rocketmq1-namesrvlog-pvc
- name: brokerlogs
persistentVolumeClaim:
claimName: rocketmq1-brokerlog-pvc
- name: brokerstore
persistentVolumeClaim:
claimName: rocketmq1-brokerstore-pvc ---
apiVersion: v1
kind: Service
metadata:
name: rocketmq1
spec:
ports:
- name: brokerslave
port:
protocol: TCP
targetPort:
- name: brokerserver
port:
protocol: TCP
targetPort:
- name: namesrvport
port:
protocol: TCP
targetPort:
selector:
app: rocketmq1
还要提前定义pvc 卷
vim rocketmq2.yaml
apiVersion: extensions/v1beta1
kind: Deployment
metadata:
name: rocketmq1
spec:
replicas:
template:
metadata:
labels:
app: rocketmq1
spec:
containers:
- name: namesrv
image: 192.168.200.10/fengjian/rocketmq-nameserver:4.3
imagePullPolicy: Always
ports:
- containerPort:
env:
- name: ROCKETMQ_HEAP_OPTS
value : "-Xms1g -Xmx1g -Xmn512m"
volumeMounts:
- name: rocketmq1namesrvlog
mountPath: /data/rocketmq/logs
- name: broker
image: 192.168.200.10/fengjian/rocketmq-broker:4.3
imagePullPolicy: Always
ports:
- name: brockerserver
containerPort:
- name: brockerslave
containerPort:
command: ["/data/rocketmq/bin/mqbroker"]
args: ["-c", "/data/rocketmq/conf/broker-b.conf"]
env:
- name: ROCKETMQ_HEAP_OPTS
value : "-Xms2g -Xmx2g -Xmn1g"
volumeMounts:
- name: brokerlogs
mountPath: /data/rocketmq/logs
- name: brokerstore
mountPath: /data/rocketmq/store
volumes:
- name: rocketmq1namesrvlog
persistentVolumeClaim:
claimName: rocketmq1-namesrvlog-pvc
- name: brokerlogs
persistentVolumeClaim:
claimName: rocketmq1-brokerlog-pvc
- name: brokerstore
persistentVolumeClaim:
claimName: rocketmq1-brokerstore-pvc ---
apiVersion: v1
kind: Service
metadata:
name: rocketmq1
spec:
ports:
- name: brokerslave
port:
protocol: TCP
targetPort:
- name: brokerserver
port:
protocol: TCP
targetPort:
- name: namesrvport
port:
protocol: TCP
targetPort:
selector:
app: rocketmq1
k8s部署rocketmq 双主的更多相关文章
- Docker 部署 RocketMQ 双主双从模式（ 版本v4.7.0）
文章转载自:http://www.mydlq.club/article/96/ 系统环境: 系统版本:CentOS 7.8 RocketMQ 版本:4.7.0 Docker 版本:19.03.13 一 ...
- Centos6.9下RocketMQ3.4.6高可用集群部署记录(双主双从+Nameserver+Console)
之前的文章已对RocketMQ做了详细介绍,这里就不再赘述了,下面是本人在测试和生产环境下RocketMQ3.4.6高可用集群的部署手册,在此分享下: 1) 基础环境 ip地址 主机名 角色 192. ...
- 49.RocketMQ 双主搭建（本文非EamonSec原创）
声明:本文非EamonSec原创,copy自网上下载的某个个文件 1.RocketMQ介绍 1.1. 简介 RocketMQ 是一款分布式.队列模型的消息中间件,具有以下特点: 能够保证严格的消息顺序 ...
- rocketmq双主模式
1.官网 https://rocketmq.apache.org/ 官方安装文档 https://rocketmq.apache.org/docs/quick-start/ 2.rocketmq多主配 ...
- rocketmq双主发送消息 SLAVE_NOT_AVAILABLE 状态
RocketMQ最佳实践之Producer 投递状态 发送消息时,将得到包含SendStatus的SendResult.首先,我们假设消息的isWaitStoreMsgOK = true(默认是tru ...
- 双主MySQL+keepalived高可用配置
部署双节点双主数据库服务器mysql 分别在二台服务器上安装mysql,要求同版本或主服务器比从服务器版本高. 主机mysql配置: Vi /etc/my.cnf [mysqld] Log-bin=m ...
- RocketMQ学习笔记（16）----RocketMQ搭建双主双从（异步复制）集群
1. 修改RocketMQ默认启动端口 由于只有两台机器,部署双主双从需要四个节点,所以只能修改rocketmq的默认启动端口,从官网下载rocketmq的source文件,解压后使用idea打开,全 ...
- MySQL双主+keeplived安装部署说明
MySQL双主+keeplived安装部署说明 一.环境介绍 1.1.规划 序号 类别 版本 主机名 IP 端口 备注 1 OS CentOS release 6.9 (Final) (minimal ...
- 003.MMM双主-双从读写分离部署
一 前期规划 1.1 主机规划 1.2 虚拟IP规划 1.3 用户列表 提示:以上角色需要在所有节点添加. 1.4 整体架构 1.4 hosts修改 1 [root@localhost ~]# vi ...
随机推荐
- 设计模式-----Builder模式
前言 近日,看到Myabtis中组件中SqlSessionFactory由SqlSessionFactoryBuilder().build()生成时,且采用Builder模式,遂记录学习之. SqlS ...
- 排版-标题及table
排版-标题 在标题中还可以包含small标签,可以用来标记副标题 副标题灰色,比主标题小 <h1>我是标题 <small>我是小标题</small></ ...
- [PHP] 数据结构-单链表头插法PHP实现
1.创建头结点 2.创建新结点 3.新结点next指向头结点next 4.头结点next指向新结点 <?php class Node{ public $data; public $next; } ...
- 微信公共号：CTO技术总监
业务价值胜过技术策略: 战略目标胜过具体项目的效益: 内置的互操作胜过定制的集成: 共享服务胜过特定目标的实现: 灵活性胜过优化: 不断演进地提炼胜过在最开始追求完美!
- 51Testing专访史亮：测试人员在国外
不久前,我接受了51Testing的访问,讨论了软件测试的一些问题.以下是全文. 1.史亮老师,作为我们51Testing的老朋友,能和我们说说您最近在忙些什么吗? 自2011年起,我加入Micros ...
- 编码最佳实践——Liskov替换原则
Liskov替换原则(Liskov Substitution Principle)是一组用于创建继承层次结构的指导原则.按照Liskov替换原则创建的继承层次结构中,客户端代码能够放心的使用它的任意类 ...
- Python中For循环
1. for i in range(10): print(i) 输出结果 F:\py\pyProject\venv\Scripts\python.exe F:/py/pyProject/venv/wh ...
- Appium学习——安装appium Server
安装appium Server 下载地址:百度网盘的下载链接:https://pan.baidu.com/s/1pKMwdfX 下载后, AppiumForWindows.zip 进行解压,点击 ap ...
- [20171223]grid用户的环境变量问题.txt
[20171223]grid用户的环境变量问题.txt --//oracle 11G 安装RAC,一般需要建立grid用户,使用这个用户管理asm,群集信息.--//在安装过程中,同事的疑问实际上也是 ...
- python第三十六天-----类中的特殊成员方法
__doc__ 查看尖的描述信息 __module__表示当前操作的对象所在的模块 __class__表示当前操作的对象所属的类 __init__构造方法 通过类创建对象自动执行 __del__析构方 ...