在Docker上永久保存数据的方法是通过Prometheus

12 月 ago

新, 韵

4 minutes

背景 – Background.

我想在Docker中使用Prometheus，但每次重新构建都会导致监视数据被初始化，所以我想要持久化数据。

目标环境

> docker-compose version
docker-compose version 1.23.2, build 1110ad01
docker-py version: 3.6.0
CPython version: 3.6.6
OpenSSL version: OpenSSL 1.1.0h  27 Mar 2018
> docker image ls
REPOSITORY                   TAG                 IMAGE ID            CREATED             SIZE
prom/prometheus              v2.6.1              5517f7057e72        9 days ago          97.8MB

顺便提一下，本文中用到的代码可以在这里找到。

解决办法

/etc/prometheus/data 被设为本地存储。

> docker-compose exec prometheus ls -la /etc/prometheus
total 12
drwxr-xr-x    1 nobody   nogroup       4096 Jan 25 03:00 .
drwxr-xr-x    1 root     root          4096 Jan 25 06:04 ..
lrwxrwxrwx    1 nobody   nogroup         39 Jan 15 20:13 console_libraries -> /usr/share/prometheus/console_libraries
lrwxrwxrwx    1 nobody   nogroup         31 Jan 15 20:13 consoles -> /usr/share/prometheus/consoles/
lrwxrwxrwx    1 root     root            11 Jan 15 20:13 data -> /prometheus
-rw-r--r--    1 root     root           682 Jan 24 06:48 prometheus.yml
> docker-compose exec prometheus ls -la /etc/prometheus/data/
total 4
drwxr-xr-x    5 nobody   nogroup        170 Jan 25 06:01 .
drwxr-xr-x    1 root     root          4096 Jan 25 06:04 ..
drwxr-xr-x    4 nobody   nogroup        136 Jan 25 03:19 data
-rw-r--r--    1 nobody   nogroup          0 Jan 25 06:01 lock
drwxr-xr-x    3 nobody   nogroup        102 Jan 25 06:01 wal

所以，如果将此挂载到本地适当的数据目录，就可以了。

那时的docker-compose.yml、Dockerfile和prometheus.yml将如下所示。

version: '2'
services:
  prometheus:
    build: .
    ports:
     - "9090:9090"
    volumes:
     - $PWD/prometheus-data:/etc/prometheus/data

FROM prom/prometheus
ADD prometheus.yml /etc/prometheus/

global:
  scrape_interval: 15s
  external_labels:
    monitor: 'codelab-monitor'
scrape_configs:
  - job_name: 'prometheus'
    scrape_interval: 5s
    static_configs:
      - targets: ['localhost:9090']

使用以下命令来启动：
运行以下命令即可启动。

> docker-compose up -d
Creating network "prometheus-test_default" with the default driver
Creating prometheus-test_prometheus_1 ... done

通过查看图表（HTTP请求总数），我们可以看到系统已成功启动，并且图表显示请求数量在增加。

我将从这里将容器卸下，并尝试重新启动。

> docker-compose down
Stopping prometheus-test_prometheus_1 ... done
Removing prometheus-test_prometheus_1 ... done
Removing network prometheus-test_default
> docker-compose up -d
Creating network "prometheus-test_default" with the default driver
Creating prometheus-test_prometheus_1 ... done

然后，您就可以看到一次删除容器时的图表，如下所示。

失败谈论

只需要一个选项：同时进行以下两个项目。

/etc/prometheus/をカレントディレクトリとマウント

由于挂载的yml文件在构建过程中被删除了，导致启动时找不到prometheus.yml文件，出现错误，让我苦恼了一个小时。

> docker-compose up
Creating network "prometheus-test_default" with the default driver
Creating prometheus-test_prometheus_1 ... done
Attaching to prometheus-test_prometheus_1
prometheus_1  | level=info ts=2019-01-25T03:19:16.5079408Z caller=main.go:243 msg="Starting Prometheus" version="(version=2.6.1, branch=HEAD, revision=b639fe140c1f71b2cbad3fc322b17efe60839e7e)"
prometheus_1  | level=info ts=2019-01-25T03:19:16.5080482Z caller=main.go:244 build_context="(go=go1.11.4, user=root@4c0e286fe2b3, date=20190115-19:12:04)"
prometheus_1  | level=info ts=2019-01-25T03:19:16.5080827Z caller=main.go:245 host_details="(Linux 4.9.125-linuxkit #1 SMP Fri Sep 7 08:20:28 UTC 2018 x86_64 433cf28249ba (none))"
prometheus_1  | level=info ts=2019-01-25T03:19:16.5082197Z caller=main.go:246 fd_limits="(soft=1048576, hard=1048576)"
prometheus_1  | level=info ts=2019-01-25T03:19:16.5088957Z caller=main.go:247 vm_limits="(soft=unlimited, hard=unlimited)"
prometheus_1  | level=info ts=2019-01-25T03:19:16.5120108Z caller=main.go:561 msg="Starting TSDB ..."
prometheus_1  | level=info ts=2019-01-25T03:19:16.5125678Z caller=web.go:429 component=web msg="Start listening for connections" address=0.0.0.0:9090
prometheus_1  | level=info ts=2019-01-25T03:19:16.5382688Z caller=main.go:571 msg="TSDB started"
prometheus_1  | level=info ts=2019-01-25T03:19:16.5388817Z caller=main.go:631 msg="Loading configuration file" filename=prometheus.yml
prometheus_1  | level=info ts=2019-01-25T03:19:16.539723Z caller=main.go:430 msg="Stopping scrape discovery manager..."
prometheus_1  | level=info ts=2019-01-25T03:19:16.5398383Z caller=main.go:444 msg="Stopping notify discovery manager..."
prometheus_1  | level=info ts=2019-01-25T03:19:16.540191Z caller=main.go:466 msg="Stopping scrape manager..."
prometheus_1  | level=info ts=2019-01-25T03:19:16.5409244Z caller=main.go:440 msg="Notify discovery manager stopped"
prometheus_1  | level=info ts=2019-01-25T03:19:16.5410353Z caller=main.go:426 msg="Scrape discovery manager stopped"
prometheus_1  | level=info ts=2019-01-25T03:19:16.5411454Z caller=main.go:460 msg="Scrape manager stopped"
prometheus_1  | level=info ts=2019-01-25T03:19:16.5413125Z caller=manager.go:664 component="rule manager" msg="Stopping rule manager..."
prometheus_1  | level=info ts=2019-01-25T03:19:16.5416848Z caller=manager.go:670 component="rule manager" msg="Rule manager stopped"
prometheus_1  | level=info ts=2019-01-25T03:19:16.5489596Z caller=notifier.go:521 component=notifier msg="Stopping notification manager..."
prometheus_1  | level=info ts=2019-01-25T03:19:16.5493186Z caller=main.go:615 msg="Notifier manager stopped"
prometheus_1  | level=error ts=2019-01-25T03:19:16.5501995Z caller=main.go:624 err="error loading config from \"prometheus.yml\": couldn't load configuration (--config.file=\"prometheus.yml\"): open prometheus.yml: no such file or directory"
prometheus-test_prometheus_1 exited with code 1

请参考

Yusuke-Shimizu/prometheus-test