Flink 配置文件 对于管理员来说,差不多经常调整的就只有 conf 下的flink-conf.yaml : 经过初步的调整,大约有以下模块的参数(未优化)

Licensed to the Apache Software Foundation (ASF) under one

or more contributor license agreements. See the NOTICE file

distributed with this work for additional information

regarding copyright ownership. The ASF licenses this file

to you under the Apache License, Version 2.0 (the

"License"); you may not use this file except in compliance

with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software

distributed under the License is distributed on an "AS IS" BASIS,

WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.

See the License for the specific language governing permissions and

limitations under the License.

==============================================================================

Jobmanager && TaskManager 的基本参数

==============================================================================

jobmanager.rpc.address: flink-master

jobmanager.rpc.port: 6123
jobmanager.heap.size: 32768m
jobstore.cache-size: 52428800
taskmanager.heap.size: 92160m
taskmanager.numberOfTaskSlots: 30
parallelism.default: 3
io.tmp.dirs: /data1/data/flink/local
taskmanager.network.numberOfBuffers: 4096
taskmanager.memory.fraction: 0.7
taskmanager.memory.preallocate: false
taskmanager.debug.memory.startLogThread: 8
taskmanager.debug.memory.logIntervalMs: 3600000
blob.fetch.retries: 50
blob.fetch.num-concurrent: 200
blob.fetch.backlog: 500
task.cancellation-interval: 30000
taskmanager.exit-on-fatal-akka-error: false
taskkmanager.log.path:

==============================================================================

High Availability && and checkpointing

=============================================================================

high-availability: ZOOKEEPER
high-availability.cluster-id: /default_flink_dc
high-availability.zookeeper.quorum: VECS01455:2181,VECS01456:2181,VECS01457:2181,VECS01458:2181,VECS01459:2181
high-availability.storageDir: hdfs://flashHadoop/flink/recovery
high-availability.zookeeper.path.root: /flink
high-availability.zookeeper.path.checkpoint-counter: /checkpoint-counter
high-availability.zookeeper.path.checkpoints: /checkpoints
high-availability.zookeeper.path.jobgraphs: /jobgraphs
high-availability.zookeeper.path.latch: /leaderlatch
high-availability.zookeeper.path.leader: /leader
high-availability.zookeeper.path.mesos-workers: /mesos-workers
high-availability.zookeeper.path.running-registry: /running_job_registry/
high-availability.zookeeper.client.connection-timeout: 18000
high-availability.zookeeper.client.max-retry-attempts: 10
high-availability.zookeeper.client.retry-wait: 5000
high-availability.zookeeper.client.session-timeout: 60000

==============================================================================

HDFS and Fault tolerance and checkpointing

==============================================================================

fs.default-scheme: hdfs://flashHadoop/flink/data
fs.output.always-create-directory: false
fs.overwrite-files: false

state.backend: filesystem
state.backend.async: true
state.backend.fs.memory-threshold: 4096
state.backend.incremental: false
state.backend.local-recovery: false
state.checkpoints.dir: hdfs://flashHadoop/flink/checkpoints_data
state.checkpoints.num-retained: 1
state.savepoints.dir: hdfs://flashHadoop/flink/savepoints_data

==============================================================================

Web Frontend

==============================================================================

web.address: 0.0.0.0
web.access-control-allow-origin: "*"
web.backpressure.cleanup-interval: 600000
web.backpressure.delay-between-samples: 50
web.backpressure.num-samples: 100
web.backpressure.refresh-interval: 60000
web.checkpoints.history: 20
web.history: 20
web.log.path: /data1/data/flink/web/log
web.refresh-interval: 3000
web.submit.enable: true
web.timeout: 10000
web.tmpdir: /data1/data/flink/web/tmp

==============================================================================

Flink Cluster Security Configuration

==============================================================================

test is not set Security

==============================================================================

HistoryServer Config

==============================================================================

historyserver.web.address: 0.0.0.0
historyserver.web.port: 8082
historyserver.archive.fs.dir: hdfs://flashHadoop/flink/completed_jobs/
historyserver.archive.fs.refresh-interval: 2000
historyserver.web.tmpdir: /data1/data/flink/tmp

==============================================================================

==============================================================================

Environment 角色环境 申明

==============================================================================

env.java.opts: -XX:+UseG1GC -XX:+UnlockDiagnosticVMOptions -XX:+UnlockExperimentalVMOptions -XX:+G1SummarizeConcMark -XX:MaxGCPauseMillis=100 -XX:-ResizePLAB -XX:+ParallelRefProcEnabled -XX:ParallelGCThreads=24 -XX:ConcGCThreads=16 -XX:G1HeapWastePercent=3 -XX:InitiatingHeapOccupancyPercent=35 -XX:G1MixedGCLiveThresholdPercent=85

env.java.opts.historyserver: -Xmx16g -Xms16g -Xmn4g -XX:+UseG1GC -XX:+UnlockDiagnosticVMOptions -XX:+UnlockExperimentalVMOptions -XX:+G1SummarizeConcMark -XX:MaxGCPauseMillis=100 -XX:-ResizePLAB -XX:+ParallelRefProcEnabled -XX:+AlwaysPreTouch -XX:ParallelGCThreads=24 -XX:ConcGCThreads=16 -XX:G1HeapWastePercent=3 -XX:InitiatingHeapOccupancyPercent=35 -XX:G1MixedGCLiveThresholdPercent=85 -verbose:gc -XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:+PrintGCDateStamps -Xloggc:/data1/data/flink/gc/gc-historyserver-hostname.log

env.java.opts.jobmanager: -XX:+UseG1GC -XX:+UnlockDiagnosticVMOptions -XX:+UnlockExperimentalVMOptions -XX:+G1SummarizeConcMark -XX:MaxGCPauseMillis=100 -XX:-ResizePLAB -XX:+ParallelRefProcEnabled -XX:ParallelGCThreads=24 -XX:ConcGCThreads=16 -XX:G1HeapWastePercent=3 -XX:InitiatingHeapOccupancyPercent=35 -XX:G1MixedGCLiveThresholdPercent=85 -verbose:gc -XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:+PrintGCDateStamps -Xloggc:/data1/data/flink/gc/gc-jobmanager-hostname.log

env.java.opts.taskmanager: -XX:+UseG1GC -XX:+UnlockDiagnosticVMOptions -XX:+UnlockExperimentalVMOptions -XX:+G1SummarizeConcMark -XX:MaxGCPauseMillis=100 -XX:-ResizePLAB -XX:+ParallelRefProcEnabled -XX:ParallelGCThreads=24 -XX:ConcGCThreads=16 -XX:G1HeapWastePercent=3 -XX:InitiatingHeapOccupancyPercent=35 -XX:G1MixedGCLiveThresholdPercent=85 -verbose:gc -XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:+PrintGCDateStamps -Xloggc:/data1/data/flink/gc/gc-taskmanager-hostname.log

env.log.dir: /data1/data/flink/log

env.log.max: 1

flink-conf.yaml的更多相关文章

  1. flink部署操作-flink standalone集群安装部署

    flink集群安装部署 standalone集群模式 必须依赖 必须的软件 JAVA_HOME配置 flink安装 配置flink 启动flink 添加Jobmanager/taskmanager 实 ...

  2. flink Standalone Cluster

    Requirements Software Requirements Flink runs on all UNIX-like environments, e.g. Linux, Mac OS X, a ...

  3. Flink集群部署

    部署方式 一般来讲有三种方式: Local Standalone Flink On Yarn/Mesos/K8s… 单机模式 参考上一篇Flink从入门到放弃(入门篇2)-本地环境搭建&构建第 ...

  4. Flink(二)CentOS7.5搭建Flink1.6.1分布式集群

    一. Flink的下载 安装包下载地址:http://flink.apache.org/downloads.html  ,选择对应Hadoop的Flink版本下载 [admin@node21 soft ...

  5. Flink的安装配置

    一. Flink的下载 安装包下载地址:http://flink.apache.org/downloads.html  ,选择对应Hadoop的Flink版本下载 [admin@node21 soft ...

  6. 一张图轻松掌握 Flink on YARN 应用启动全流程(上)

    Flink 支持 Standalone 独立部署和 YARN.Kubernetes.Mesos 等集群部署模式,其中 YARN 集群部署模式在国内的应用越来越广泛.Flink 社区将推出 Flink ...

  7. kerberos系列之flink认证配置

    大数据安全系列的其它文章 https://www.cnblogs.com/bainianminguo/p/12548076.html-----------安装kerberos https://www. ...

  8. 分享一个Flink checkpoint失败的问题和解决办法

    本文来自: PerfMa技术社区 PerfMa(笨马网络)官网 接触Flink一段时间了,遇到了一些问题,其中有一个checkpoint失败导致作业重启的问题,遇到了很多次,重启之后一般也能恢复正常, ...

  9. flink集群模式安装配置

    一.手动下载安装包 wget http://mirrors.tuna.tsinghua.edu.cn/apache/flink/flink-1.6.1/flink-1.6.1-bin-hadoop27 ...

  10. Linux使用docker安装flink

    配置文件不是原始的 修改过,可以根据自己的配置文件来 jobmanager配置 flink-conf.yaml ############################################ ...

随机推荐

  1. Git 密钥对处理

    生成密钥对: ssh-keygen -t rsa cd .ssh ls      id_rsa          私钥      id_rsa.pub   公钥

  2. ubuntu18.04修改网卡名称为eth0

    1.修改grub文件 vim /etc/default/grub 查找 GRUB_CMDLINE_LINUX="" 修改为 GRUB_CMDLINE_LINUX="net ...

  3. PowerShell或命令行运行javac xx.java提示“编码GBK的不可映射字符”

    由于JDK是国际版的,我们在用javac编译时,编译程序首先会获得我们操作系统默认采用的编码格式(GBK),然后JDK就把Java源文件从GBK编码格式转换为Java内部默认的Unicode格式放入内 ...

  4. LeetCode算法题-Find Smallest Letter Greater Than Target(Java实现)

    这是悦乐书的第306次更新,第326篇原创 01 看题和准备 今天介绍的是LeetCode算法题中Easy级别的第175题(顺位题号是744).给定一个仅包含小写字母的有序字符数组,并给定目标字母目标 ...

  5. Docker:pipeline编写基本技巧- jenkins配置通过免交互方式拉取git源码管理仓库的代码

    工作中,从git仓库拉取代码有2种方式:交互式和非交互式 什么是交互式?就是拉取需要权限才能访问的代码时,需要输入密码 免交互式呢? 是通过密钥,私钥的方式,让服务端信任客户端,产生信任后,任何一次客 ...

  6. TensorRT&Sample&Python[yolov3_onnx]

    本文是基于TensorRT 5.0.2基础上,关于其内部的yolov3_onnx例子的分析和介绍. 本例子展示一个完整的ONNX的pipline,在tensorrt 5.0的ONNX-TensorRT ...

  7. neutron二

    第四篇neutron— 网络实践   一.虚拟机获取 ip:   用 namspace 隔离 DHCP 服务   Neutron 通过 dnsmasq 提供 DHCP 服务,而 dnsmasq 通过 ...

  8. dedecms织梦的不同栏目调用不同banner图的方法

    在做织梦站的时候我们会有不同的栏目,比如联系我们,产品中心等等,banner也不一样,方法如下: 我们可以使用织梦的顶级栏目ID标签,把图片命名成顶级栏目typeid ,代码如下: <img s ...

  9. 简单实现Python调用有道API接口(最新的)

    # ''' # Created on 2018-5-26 # # @author: yaoshuangqi # ''' import urllib.request import urllib.pars ...

  10. Python开发【第十六篇】:AJAX全套(转)

    作者:武沛齐 出处:http://www.cnblogs.com/wupeiqi/ 本文版权归作者和博客园共有,欢迎转载,但未经作者同意必须保留此段声明,且在文章页面明显位置给出原文连接. 概述 对于 ...