Commit a4895830 authored by 吴晟's avatar 吴晟
Browse files

Merge branch 'fix/documents'

parents a862690e a715220f

docs/README.md

+0 −1
Original line number Diff line number Diff line 
## Documents

[![version](https://img.shields.io/badge/document--version-3.2.5--2017-green.svg)]() 

[![cn doc](https://img.shields.io/badge/document-中文-blue.svg)](README_ZH.md)



  * Getting Started

docs/README_ZH.md

+0 −1
Original line number Diff line number Diff line 
## 中文文档

[![version](https://img.shields.io/badge/document--version-3.2.5--2017-green.svg)]() 

[![EN doc](https://img.shields.io/badge/document-English-blue.svg)](README.md)



  * [项目简介](/README_ZH.md)

docs/cn/Deploy-collector-in-cluster-mode-CN.md

+24 −15
Original line number Diff line number Diff line
@@ -23,49 +23,58 @@ thread_pool.bulk.queue_size: 1000 


### 部署collector

1. 解压安装包`tar -xvf skywalking-collector.tar.gz`，windows用户可以选择zip包

1. 运行`bin/startup.sh`启动。windows用户为.bat文件。

2. 设置Collector集群模式



集群模式主要依赖Zookeeper的注册和应用发现能力。所以，你只需要调整 `config/application.yml`中的host和port配置，使用实际IP和端口，代替默认配置。

其次，将storage的注释取消，并修改为Elasticsearch集群的节点地址信息。





- `config/application.yml`

```

cluster:

# Zookeeper地址配置

# 配置zookeeper集群信息

  zookeeper:

    hostPort: localhost:2181

    sessionTimeout: 100000

# agent_server, agent_stream, ui, collector_inside中配置的IP都是Collector所使用的IP地址

agent_server:

naming:

# 配置探针使用的host和port

  jetty:

    host: localhost

    # The port used

    port: 10800

    context_path: /

agent_stream:

  grpc:

remote:

  gRPC:

    host: localhost

    port: 11800

agent_gRPC:

  gRPC:

    host: localhost

    port: 11800

agent_jetty:

  jetty:

    host: localhost

    port: 12800

    context_path: /

agent_stream:

  default:

    buffer_file_path: ../buffer/

    buffer_offset_max_file_size: 10M

    buffer_segment_max_file_size: 500M

ui:

  jetty:

    host: localhost

    port: 12800

    context_path: /

collector_inside:

  grpc:

    host: localhost

    port: 11800

# 配置 Elasticsearch 集群连接信息

storage:

  elasticsearch:

    cluster_name: CollectorDBCluster

    cluster_transport_sniffer: true

    # Elastic Search地址信息

    cluster_nodes: localhost:9300

    index_shards_number: 2

    index_replicas_number: 0

    ttl: 7

```



## Collector集群模式启动

集群模式主要依赖Zookeeper的注册和应用发现能力。所以，你只需要调整 `config/application.yml`中，agent_server, agent_stream, ui, collector_inside这些配置项的ip信息，使用真实的IP地址或者hostname，Collector就会使用集群模式运行。

其次，将elasticsearch的注释取消，并修改集群的节点地址信息。
 


3. 运行`bin/startup.sh`启动。windows用户为.bat文件。

docs/cn/Deploy-collector-in-standalone-mode-CN.md

+5 −48
Original line number Diff line number Diff line 
## 用途说明

单机模式使用本地H2数据库，不支持集群部署。主要用于：预览、功能测试、演示和低压力系统。

# 用途说明

单机模式使用本地H2数据库，不支持集群部署。主要用于：预览、功能测试、演示和低压力系统。你可选择使用Elasticsearch作为



## 所需的第三方软件

- JDK8+
@@ -10,55 +10,12 @@ 
## Quick Start

Collector单机模拟启动简单，提供和集群模式相同的功能，单机模式下除端口被占用的情况下，直接启动即可。



### 部署collector

## 部署collector

1. 解压安装包`tar -xvf skywalking-collector.tar.gz`，windows用户可以选择zip包

1. 运行`bin/startup.sh`启动。windows用户为.bat文件。



- `config/application.yml`

```

# 单机模式下无需配置集群相关信息

#cluster:

#  zookeeper:

#    hostPort: localhost:2181

#    sessionTimeout: 100000

# agent_server, agent_stream, ui, collector_inside中配置的IP都是Collector所使用的IP地址

agent_server:

  jetty:

    host: localhost

    # The port used

    port: 10800

    context_path: /

agent_stream:

  grpc:

    host: localhost

    port: 11800

  jetty:

    host: localhost

    port: 12800

    context_path: /

ui:

  jetty:

    host: localhost

    port: 12800

    context_path: /

collector_inside:

  grpc:

    host: localhost

    port: 11800



#storage:

#  elasticsearch:

#    cluster_name: CollectorDBCluster

#    cluster_transport_sniffer: true

#    Elastic Search地址信息

#    cluster_nodes: localhost:9300

#    index_shards_number: 2

#    index_replicas_number: 0

```



## 使用Elastic Search代替H2存储

由于H2数据库性能的局限性，我们在单机模式下，也支持使用ElasticSearch 5.3作为存储。你需要安装对应的ElasticSearch 5.3，并取消

- 在单机模式下除了支持内置的H2数据库运行，也支持其他的存储（当前已支持的ElasticSearch 5.3），安装storage注释，修改配置信息即可。取消Storage相关配置节的注释。

- 在单机模式下除了支持内置的H2数据库运行，也支持其他的存储（当前已支持的ElasticSearch 5.3），取消Storage相关配置节的注释，并修改配置。

```yaml

#storage:

#  elasticsearch:
@@ -69,7 +26,7 @@ collector_inside: 
#    index_replicas_number: 0

```



### 部署Elasticsearch

## 部署Elasticsearch

- 修改`elasticsearch.yml`文件

  - 设置 `cluster.name: CollectorDBCluster`。此名称需要和collector配置文件一致。

  - 设置 `node.name: anyname`, 可以设置为任意名字，如Elasticsearch为集群模式，则每个节点名称需要不同。

docs/en/Deploy-collector-in-cluster-mode.md

+23 −18
Original line number Diff line number Diff line
@@ -21,53 +21,58 @@ thread_pool.bulk.queue_size: 1000 


- Start Elasticsearch



## Single Node Mode Collector

Single Node collector is easy to deploy, and provides same features as cluster mode. You can use almost all default config to run in this mode. And attention, all the default configs of single node mode, depend on running the collector, traced application, ElasticSearch and Zookeeper in the same machine. 



### Deploy collector servers

1. Run `tar -xvf skywalking-collector.tar.gz`

1. Run `bin/startup.sh`

2. Config collector in cluster mode.



Cluster mode depends on Zookeeper register and application discovery capabilities. So, you just need to adjust the IP config items in `config/application.yml`. Change IP and port configs of agent_server, agent_stream, ui, collector_inside, replace them to the real ip or hostname which you want to use for cluster.



- `config/application.yml`

```

cluster:

  # The address of Zookeeper

# The Zookeeper cluster for collector cluster management.

  zookeeper:

    hostPort: localhost:2181

    sessionTimeout: 100000

# IPs in agent_server, agent_stream, ui, collector_inside are addresses of Collector

agent_server:

naming:

# Host and port used for agent config

  jetty:

    host: localhost

    # The port used

    port: 10800

    context_path: /

agent_stream:

  grpc:

remote:

  gRPC:

    host: localhost

    port: 11800

agent_gRPC:

  gRPC:

    host: localhost

    port: 11800

agent_jetty:

  jetty:

    host: localhost

    port: 12800

    context_path: /

agent_stream:

  default:

    buffer_file_path: ../buffer/

    buffer_offset_max_file_size: 10M

    buffer_segment_max_file_size: 500M

ui:

  jetty:

    host: localhost

    port: 12800

    context_path: /

collector_inside:

  grpc:

    host: localhost

    port: 11800

# Config Elasticsearch cluster connection info.

storage:

  elasticsearch:

    cluster_name: CollectorDBCluster

    cluster_transport_sniffer: true

    # The address of Elastic Search

    cluster_nodes: localhost:9300

    index_shards_number: 2

    index_replicas_number: 0

    ttl: 7

```



## Cluster Mode Collector

Cluster mode depends on Zookeeper register and application discovery capabilities. So, you just need to adjust the IP config items in `config/application.yml`. Change IP and port configs of agent_server, agent_stream, ui, collector_inside, replace them to the real ip or hostname which you want to use for cluster.
 


3. Run `bin/startup.sh`