docker-gen

docker-gen

基于Docker容器元数据的动态配置文件生成工具

docker-gen是一个基于Docker容器元数据的文件生成工具。它能够生成多种配置文件,包括日志、反向代理和服务发现等。支持多种安装方式,如主机安装、容器捆绑或独立运行。通过其灵活的模板系统和功能丰富的API,docker-gen简化了Docker环境的管理和配置过程。

docker-gen容器元数据模板渲染配置生成DockerGithub开源项目

docker-gen

Tests GitHub release Docker Image Size Docker stars Docker pulls

docker-gen is a file generator that renders templates using docker container meta-data.

It can be used to generate various kinds of files for:

  • Centralized logging - fluentd, logstash or other centralized logging tools that tail the containers JSON log file or files within the container.
  • Log Rotation - logrotate files to rotate container JSON log files
  • Reverse Proxy Configs - nginx, haproxy, etc. reverse proxy configs to route requests from the host to containers
  • Service Discovery - Scripts (python, bash, etc..) to register containers within etcd, hipache, etc..

Installation

There are three common ways to run docker-gen:

  • on the host
  • bundled in a container with another application
  • separate standalone containers

Host Install

Download the version you need, untar, and install to your PATH.

wget https://github.com/nginx-proxy/docker-gen/releases/download/0.12.0/docker-gen-linux-amd64-0.12.0.tar.gz tar xvzf docker-gen-linux-amd64-0.12.0.tar.gz ./docker-gen

Bundled Container Install

Docker-gen can be bundled inside of a container along-side applications.

nginx-proxy/nginx-proxy trusted build is an example of running docker-gen within a container along-side nginx. jwilder/docker-register is an example of running docker-gen within a container to do service registration with etcd.

Separate Container Install

It can also be run as two separate containers using the nginx-proxy/docker-gen image, together with virtually any other image.

This is how you could run the official nginx image and have docker-gen generate a reverse proxy config in the same way that nginx-proxy works. You may want to do this to prevent having the docker socket bound to a publicly exposed container service.

Start nginx with a shared volume:

docker run -d -p 80:80 --name nginx -v /tmp/nginx:/etc/nginx/conf.d -t nginx

Fetch the template and start the docker-gen container with the shared volume:

mkdir -p /tmp/templates && cd /tmp/templates curl -o nginx.tmpl https://raw.githubusercontent.com/nginx-proxy/docker-gen/main/templates/nginx.tmpl docker run -d --name nginx-gen --volumes-from nginx \ -v /var/run/docker.sock:/tmp/docker.sock:rw \ -v /tmp/templates:/etc/docker-gen/templates \ -t nginxproxy/docker-gen -notify-sighup nginx -watch -only-exposed /etc/docker-gen/templates/nginx.tmpl /etc/nginx/conf.d/default.conf

Start a container, taking note of any Environment variables a container expects. See the top of a template for details.

docker run --env VIRTUAL_HOST='example.com' --env VIRTUAL_PORT=80 ...

Usage

$ docker-gen
Usage: docker-gen [options] template [dest]

Generate files from docker container meta-data

Options:
  -config value
      config files with template directives. Config files will be merged if this option is specified multiple times. (default [])
  -endpoint string
      docker api endpoint (tcp|unix://..). Default unix:///var/run/docker.sock
  -interval int
      notify command interval (secs)
  -keep-blank-lines
      keep blank lines in the output file
  -notify restart xyz
      run command after template is regenerated (e.g restart xyz)
  -notify-output
      log the output(stdout/stderr) of notify command
  -notify-sighup container-ID
      send HUP signal to container.
      Equivalent to 'docker kill -s HUP container-ID', or `-notify-container container-ID -notify-signal 1`.
      You can pass this option multiple times to send HUP to multiple containers.
  -notify-container container-ID
      send -notify-signal signal (defaults to 1 / HUP) to container.
      You can pass this option multiple times to notify multiple containers.
  -notify-filter key=value
      container filter for notification (e.g -notify-filter name=foo).
      You can pass this option multiple times to combine filters with AND.
      https://docs.docker.com/engine/reference/commandline/ps/#filter
  -notify-signal signal
      signal to send to the -notify-container and -notify-filter. -1 to call docker restart. Defaults to 1 aka. HUP.
      All available signals available on the dockerclient
      https://github.com/fsouza/go-dockerclient/blob/main/signal.go
  -only-exposed
      only include containers with exposed ports
  -only-published
      only include containers with published ports (implies -only-exposed)
  -include-stopped
      include stopped containers
  -tlscacert string
      path to TLS CA certificate file (default "~/.docker/machine/machines/default/ca.pem")
  -tlscert string
      path to TLS client certificate file (default "~/.docker/machine/machines/default/cert.pem")
  -tlskey string
      path to TLS client key file (default "~/.docker/machine/machines/default/key.pem")
  -tlsverify
      verify docker daemon's TLS certicate (default true)
  -version
      show version
  -watch
      watch for container changes
  -wait
      minimum (and/or maximum) duration to wait after each container change before triggering

Arguments:
  template - path to a template to generate
  dest - path to write the template. If not specfied, STDOUT is used

Environment Variables:
  DOCKER_HOST - default value for -endpoint
  DOCKER_CERT_PATH - directory path containing key.pem, cert.pm and ca.pem
  DOCKER_TLS_VERIFY - enable client TLS verification]

If no <dest> file is specified, the output is sent to stdout. Mainly useful for debugging.

Configuration file

Using the -config flag from above you can tell docker-gen to use the specified config file instead of command-line options. Multiple templates can be defined and they will be executed in the order that they appear in the config file.

An example configuration file, docker-gen.cfg can be found in the examples folder.

Configuration File Syntax

[[config]] # Starts a configuration section dest = "path/to/a/file" # path to write the template. If not specfied, STDOUT is used notifycmd = "/etc/init.d/foo reload" # run command after template is regenerated (e.g restart xyz) onlyexposed = true # only include containers with exposed ports template = "/path/to/a/template/file.tmpl" # path to a template to generate watch = true # watch for container changes wait = "500ms:2s" # debounce changes with a min:max duration. Only applicable if watch = true [config.NotifyContainers] # Starts a notify container section containername = 1 # container name followed by the signal to send container_id = 1 # or the container id can be used followed by the signal to send

Putting it all together here is an example configuration file.

[[config]] template = "/etc/nginx/nginx.conf.tmpl" dest = "/etc/nginx/sites-available/default" onlyexposed = true notifycmd = "/etc/init.d/nginx reload" [[config]] template = "/etc/logrotate.conf.tmpl" dest = "/etc/logrotate.d/docker" watch = true [[config]] template = "/etc/docker-gen/templates/nginx.tmpl" dest = "/etc/nginx/conf.d/default.conf" watch = true wait = "500ms:2s" [config.NotifyContainers] nginx = 1 # 1 is a signal number to be sent; here SIGHUP e75a60548dc9 = 1 # a key can be either container name (nginx) or ID

Templating

The templates used by docker-gen are written using the Go text/template language. In addition to the built-in functions supplied by Go, docker-gen uses sprig and some additional functions to make it simpler (or possible) to generate your desired output. Some templates rely on environment variables within the container to make decisions on what to generate from the template.

Several templates may be parsed at once by using a semicolon (;) to delimit the template value. This can be used as a proxy for Golang's nested template functionality. In all cases, the main rendered template should go first.

[[config]]
template = "/etc/docker-gen/templates/nginx.tmpl;/etc/docker-gen/templates/header.tmpl"
dest = "/etc/nginx/conf.d/default.conf"
watch = true
wait = "500ms:2s"

Emit Structure

Within the templates, the object emitted by docker-gen will be a structure consisting of following Go structs:

type RuntimeContainer struct { ID string Created time.Time Addresses []Address Networks []Network Gateway string Name string Hostname string Image DockerImage Env map[string]string Volumes map[string]Volume Node SwarmNode Labels map[string]string IP string IP6LinkLocal string IP6Global string Mounts []Mount State State } type Address struct { IP string IP6LinkLocal string IP6Global string Port string HostPort string Proto string HostIP string } type Network struct { IP string Name string Gateway string EndpointID string IPv6Gateway string GlobalIPv6Address string MacAddress string GlobalIPv6PrefixLen int IPPrefixLen int Internal bool } type DockerImage struct { Registry string Repository string Tag string } type Mount struct { Name string Source string Destination string Driver string Mode string RW bool } type Volume struct { Path string HostPath string ReadWrite bool } type SwarmNode struct { ID string Name string Address Address } type State struct { Running bool Health Health } type Health struct { Status string } // Accessible from the root in templates as .Docker type Docker struct { Name string NumContainers int NumImages int Version string ApiVersion string GoVersion string OperatingSystem string Architecture string CurrentContainerID string } // Host environment variables accessible from root in templates as .Env

For example, this is a JSON version of an emitted RuntimeContainer struct:

{ "ID": "71e9768075836eb38557adcfc71a207386a0c597dbeda240cf905df79b18cebf", "Addresses": [ { "IP": "172.17.0.4", "Port": "22", "Proto": "tcp", "HostIP": "192.168.10.24", "HostPort": "2222" } ], "Gateway": "172.17.42.1", "Node": { "ID": "I2VY:P7PF:TZD5:PGWB:QTI7:QDSP:C5UD:DYKR:XKKK:TRG2:M2BL:DFUN", "Name": "docker-test", "Address": { "IP": "192.168.10.24" } }, "Labels": { "operatingsystem": "Ubuntu 14.04.2 LTS", "storagedriver": "devicemapper", "anything_foo": "something_bar" }, "IP": "172.17.0.4", "Name": "docker_register", "Hostname": "71e976807583", "Image": { "Registry": "jwilder", "Repository": "docker-register" }, "Env": { "ETCD_HOST": "172.17.42.1:4001", "PATH": "/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin", "DOCKER_HOST": "unix:///var/run/docker.sock", "HOST_IP": "172.17.42.1" }, "Volumes": { "/mnt": { "Path": "/mnt", "HostPath": "/Users/joebob/tmp", "ReadWrite": true } } }

Functions

  • Functions from Go
  • Functions from Sprig v3, except for those that have the same name as one of the following functions.
  • closest $array $value: Returns the longest matching substring in $array that matches $value
  • coalesce ...: Returns the first non-nil argument.
  • comment $delimiter $string: Returns $string with each line prefixed by $delimiter (helpful for debugging combined with Sprig toPrettyJson: {{ toPrettyJson $ | comment "#" }}).
  • contains $map $key: Returns true if $map contains $key. Takes maps from string to any type.
  • dir $path: Returns an array of filenames in the specified $path.
  • exists $path: Returns true if $path refers to an existing file or directory. Takes a string.
  • eval $templateName [$data]: Evaluates the named template like Go's built-in template action, but instead of writing out the result it returns the result as a string so that it can be post-processed. The $data argument may be omitted, which is equivalent to passing nil.
  • groupBy $containers $fieldPath: Groups an array of RuntimeContainer instances based on the values of a field path expression $fieldPath. A field path expression is a dot-delimited list of map keys or struct member names specifying the path from container to a nested value, which must be a string. Returns a map from the value of the field path expression to an array of containers having that value. Containers that do not have a value for the field path in question are omitted.
  • groupByWithDefault $containers $fieldPath $defaultValue: Returns the same as groupBy, but containers that do not have a value for the field path are instead included in the map under the $defaultValue key.
  • groupByKeys $containers $fieldPath: Returns the same as groupBy but only returns the keys of the map.
  • groupByMulti $containers $fieldPath $sep: Like groupBy, but the string value specified by $fieldPath is first split by $sep into a list of strings. A container whose $fieldPath value contains a list of strings will show up in the map output under each of those strings.
  • groupByLabel $containers $label: Returns the same as groupBy but grouping by the given label's value. Containers that do not have the $label set are omitted.
  • groupByLabelWithDefault $containers $label $defaultValue: Returns the same as groupBy but grouping by the given label's value. Containers that do not have the $label set are included in the map under the $defaultValue key.
  • include $file: Returns content of $file, and empty string if file reading error.
  • intersect $slice1 $slice2: Returns the strings that exist in both string slices.
  • fromYaml $string / mustFromYaml $string: Similar to Sprig's fromJson / mustFromJson, but for YAML.
  • toYaml $dict / mustToYaml $dict: Similar to Sprig's toJson / mustToJson, but for YAML.
  • keys $map: Returns the keys from $map. If $map is nil, a nil is returned. If $map is not a map, an error will be thrown.
  • sortStringsAsc $strings: Returns a slice of strings $strings sorted in ascending order.
  • sortStringsDesc $strings: Returns a slice of strings $strings sorted in descending (reverse) order.
  • sortObjectsByKeysAsc $objects $fieldPath: Returns the array $objects, sorted in ascending order based on the values of a field path

编辑推荐精选

TRAE编程

TRAE编程

AI辅助编程,代码自动修复

Trae是一种自适应的集成开发环境(IDE),通过自动化和多元协作改变开发流程。利用Trae,团队能够更快速、精确地编写和部署代码,从而提高编程效率和项目交付速度。Trae具备上下文感知和代码自动完成功能,是提升开发效率的理想工具。

AI工具TraeAI IDE协作生产力转型热门
博思AIPPT

博思AIPPT

AI一键生成PPT,就用博思AIPPT!

博思AIPPT,新一代的AI生成PPT平台,支持智能生成PPT、AI美化PPT、文本&链接生成PPT、导入Word/PDF/Markdown文档生成PPT等,内置海量精美PPT模板,涵盖商务、教育、科技等不同风格,同时针对每个页面提供多种版式,一键自适应切换,完美适配各种办公场景。

AI办公办公工具AI工具博思AIPPTAI生成PPT智能排版海量精品模板AI创作热门
潮际好麦

潮际好麦

AI赋能电商视觉革命,一站式智能商拍平台

潮际好麦深耕服装行业,是国内AI试衣效果最好的软件。使用先进AIGC能力为电商卖家批量提供优质的、低成本的商拍图。合作品牌有Shein、Lazada、安踏、百丽等65个国内外头部品牌,以及国内10万+淘宝、天猫、京东等主流平台的品牌商家,为卖家节省将近85%的出图成本,提升约3倍出图效率,让品牌能够快速上架。

iTerms

iTerms

企业专属的AI法律顾问

iTerms是法大大集团旗下法律子品牌,基于最先进的大语言模型(LLM)、专业的法律知识库和强大的智能体架构,帮助企业扫清合规障碍,筑牢风控防线,成为您企业专属的AI法律顾问。

SimilarWeb流量提升

SimilarWeb流量提升

稳定高效的流量提升解决方案,助力品牌曝光

稳定高效的流量提升解决方案,助力品牌曝光

Sora2视频免费生成

Sora2视频免费生成

最新版Sora2模型免费使用,一键生成无水印视频

最新版Sora2模型免费使用,一键生成无水印视频

Transly

Transly

实时语音翻译/同声传译工具

Transly是一个多场景的AI大语言模型驱动的同声传译、专业翻译助手,它拥有超精准的音频识别翻译能力,几乎零延迟的使用体验和支持多国语言可以让你带它走遍全球,无论你是留学生、商务人士、韩剧美剧爱好者,还是出国游玩、多国会议、跨国追星等等,都可以满足你所有需要同传的场景需求,线上线下通用,扫除语言障碍,让全世界的语言交流不再有国界。

讯飞绘文

讯飞绘文

选题、配图、成文,一站式创作,让内容运营更高效

讯飞绘文,一个AI集成平台,支持写作、选题、配图、排版和发布。高效生成适用于各类媒体的定制内容,加速品牌传播,提升内容营销效果。

热门AI辅助写作AI工具讯飞绘文内容运营AI创作个性化文章多平台分发AI助手
商汤小浣熊

商汤小浣熊

最强AI数据分析助手

小浣熊家族Raccoon,您的AI智能助手,致力于通过先进的人工智能技术,为用户提供高效、便捷的智能服务。无论是日常咨询还是专业问题解答,小浣熊都能以快速、准确的响应满足您的需求,让您的生活更加智能便捷。

imini AI

imini AI

像人一样思考的AI智能体

imini 是一款超级AI智能体,能根据人类指令,自主思考、自主完成、并且交付结果的AI智能体。

下拉加载更多