如何在Docker中的数据库初始化后启动flyway

Question

我有以下docker compose文件（docker-compose-dev.yml）：

version: '3'
services:
  my_sql_db:
    image: percona:latest
    container_name: my_sql_db
    environment:

      MYSQL_ROOT_PASSWORD: password
      MYSQL_DATABASE: abhs
    ports:
    - "3306:3306"

  migration:
    image: boxfuse/flyway:latest
    container_name: flyway_migration
    volumes:
      - ./flyway_scripts/src/main/resources/db/migration:/flyway/sql
    command: -url=jdbc:mysql://my_sql_db:3306/abhs?useUnicode=true&characterEncoding=utf8&useSSL=false -user=root -password=password migrate
    depends_on:
    - my_sql_db

以及下面的docker compose.yml：

version: '3'
services:
  migration:
    image: boxfuse/flyway:latest
    container_name: flyway_migration
    volumes:
     - ./flyway_scripts/src/main/resources/db/migration:/flyway/sql

然后执行以下命令：

docker-compose -f docker-compose.yml -f docker-compose-dev.yml up

它会导致错误：

在日志中，我看到以下内容：

my_sql_db    | Initializing database
flyway_migration  | Flyway Community Edition 5.1.4 by Boxfuse
flyway_migration  |
my_sql_db    | 2018-08-24T08:47:41.616694Z 0 [Warning] 'NO_ZERO_DATE', 'NO_ZERO_IN_DATE' and 'ERROR_FOR_DIVISION_BY_ZERO' sql modes should be used with strict mode. They will be merged with strict mode in a future release.
my_sql_db    | 2018-08-24T08:47:41.616747Z 0 [Warning] 'NO_AUTO_CREATE_USER' sql mode was not set.
flyway_migration  | ERROR:
flyway_migration  | Unable to obtain connection from database (jdbc:mysql://my_sql_db:3306/abhs?useUnicode=true&characterEncoding=utf8&useSSL=false) for user 'root': Could not connect to address=(host=my_sql_db)(port=3306)(type=master) : Connection refused (Connection refused)
flyway_migration  | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
flyway_migration  | SQL State  : 08
flyway_migration  | Error Code : -1
flyway_migration  | Message    : Could not connect to address=(host=my_sql_db)(port=3306)(type=master) : Connection refused (Connection refused)
my_sql_db    | 2018-08-24T08:47:43.024690Z 0 [Warning] InnoDB: New log files created, LSN=45790
flyway_migration  |
my_sql_db    | 2018-08-24T08:47:43.443625Z 0 [Warning] InnoDB: Creating foreign key constraint system tables.
my_sql_db    | 2018-08-24T08:47:43.588008Z 0 [Warning] No existing UUID has been found, so we assume that this is the first time that this server has been started. Generating a new UUID: 5dc59a4f-a77a-11e8-b6cb-0242ac130002.
my_sql_db    | 2018-08-24T08:47:43.760654Z 0 [Warning] Gtid table is not ready to be used. Table 'mysql.gtid_executed' cannot be opened.
my_sql_db    | 2018-08-24T08:47:44.518107Z 0 [Warning] CA certificate ca.pem is self signed.
my_sql_db    | 2018-08-24T08:47:44.925466Z 1 [Warning] root@localhost is created with an empty password ! Please consider switching off the --initialize-insecure option.
my_sql_db    | 2018-08-24T08:47:54.762213Z 1 [Warning] 'user' entry 'root@localhost' ignored in --skip-name-resolve mode.
my_sql_db    | 2018-08-24T08:47:54.762517Z 1 [Warning] 'user' entry 'mysql.session@localhost' ignored in --skip-name-resolve mode.
my_sql_db    | 2018-08-24T08:47:54.762889Z 1 [Warning] 'user' entry 'mysql.sys@localhost' ignored in --skip-name-resolve mode.
my_sql_db    | 2018-08-24T08:47:54.763244Z 1 [Warning] 'db' entry 'performance_schema mysql.session@localhost' ignored in --skip-name-resolve mode.
my_sql_db    | 2018-08-24T08:47:54.763472Z 1 [Warning] 'db' entry 'sys mysql.sys@localhost' ignored in --skip-name-resolve mode.
my_sql_db    | 2018-08-24T08:47:54.763788Z 1 [Warning] 'proxies_priv' entry '@ root@localhost' ignored in --skip-name-resolve mode.
my_sql_db    | 2018-08-24T08:47:54.763928Z 1 [Warning] 'tables_priv' entry 'user mysql.session@localhost' ignored in --skip-name-resolve mode.
my_sql_db    | 2018-08-24T08:47:54.764128Z 1 [Warning] 'tables_priv' entry 'sys_config mysql.sys@localhost' ignored in --skip-name-resolve mode.
my_sql_db    | Database initialized
my_sql_db    | MySQL init process in progress...
my_sql_db    | 2018-08-24T08:47:58.970290Z 0 [Warning] 'NO_ZERO_DATE', 'NO_ZERO_IN_DATE' and 'ERROR_FOR_DIVISION_BY_ZERO' sql modes should be used with strict mode. They will be merged with strict mode in a future release.
my_sql_db    | 2018-08-24T08:47:58.970345Z 0 [Warning] 'NO_AUTO_CREATE_USER' sql mode was not set.
my_sql_db    | 2018-08-24T08:47:58.974061Z 0 [Note] mysqld (mysqld 5.7.22-22) starting as process 58 ...
my_sql_db    | 2018-08-24T08:47:58.999651Z 0 [Note] InnoDB: PUNCH HOLE support available
my_sql_db    | 2018-08-24T08:47:58.999685Z 0 [Note] InnoDB: Mutexes and rw_locks use GCC atomic builtins
my_sql_db    | 2018-08-24T08:47:58.999689Z 0 [Note] InnoDB: Uses event mutexes
my_sql_db    | 2018-08-24T08:47:58.999692Z 0 [Note] InnoDB: GCC builtin __atomic_thread_fence() is used for memory barrier
my_sql_db    | 2018-08-24T08:47:58.999695Z 0 [Note] InnoDB: Compressed tables use zlib 1.2.8
my_sql_db    | 2018-08-24T08:47:58.999698Z 0 [Note] InnoDB: Using Linux native AIO
my_sql_db    | 2018-08-24T08:47:59.000153Z 0 [Note] InnoDB: Number of pools: 1
my_sql_db    | 2018-08-24T08:47:59.000426Z 0 [Note] InnoDB: Using CPU crc32 instructions
my_sql_db    | 2018-08-24T08:47:59.002306Z 0 [Note] InnoDB: Initializing buffer pool, total size = 128M, instances = 1, chunk size = 128M
my_sql_db    | 2018-08-24T08:47:59.006893Z 0 [Note] InnoDB: Completed initialization of buffer pool
my_sql_db    | 2018-08-24T08:47:59.013219Z 0 [Note] InnoDB: If the mysqld execution user is authorized, page cleaner thread priority can be changed. See the man page of setpriority().
my_sql_db    | 2018-08-24T08:47:59.024242Z 0 [Note] InnoDB: Crash recovery did not find the parallel doublewrite buffer at /var/lib/mysql/xb_doublewrite
my_sql_db    | 2018-08-24T08:47:59.026263Z 0 [Note] InnoDB: Highest supported file format is Barracuda.
my_sql_db    | 2018-08-24T08:47:59.066469Z 0 [Note] InnoDB: Created parallel doublewrite buffer at /var/lib/mysql/xb_doublewrite, size 3932160 bytes
my_sql_db    | 2018-08-24T08:47:59.071752Z 0 [Note] InnoDB: Creating shared tablespace for temporary tables
my_sql_db    | 2018-08-24T08:47:59.072052Z 0 [Note] InnoDB: Setting file './ibtmp1' size to 12 MB. Physically writing the file full; Please wait ...
my_sql_db    | 2018-08-24T08:47:59.422155Z 0 [Note] InnoDB: File './ibtmp1' size is now 12 MB.
my_sql_db    | 2018-08-24T08:47:59.423325Z 0 [Note] InnoDB: 96 redo rollback segment(s) found. 96 redo rollback segment(s) are active.
my_sql_db    | 2018-08-24T08:47:59.423376Z 0 [Note] InnoDB: 32 non-redo rollback segment(s) are active.
my_sql_db    | 2018-08-24T08:47:59.423900Z 0 [Note] InnoDB: Waiting for purge to start
my_sql_db    | 2018-08-24T08:47:59.474066Z 0 [Note] InnoDB: Percona XtraDB (http://www.percona.com) 5.7.22-22 started; log sequence number 2595255
my_sql_db    | 2018-08-24T08:47:59.474647Z 0 [Note] Plugin 'FEDERATED' is disabled.
my_sql_db    | 2018-08-24T08:47:59.499970Z 0 [Note] Found ca.pem, server-cert.pem and server-key.pem in data directory. Trying to enable SSL support using them.
my_sql_db    | 2018-08-24T08:47:59.500004Z 0 [Note] Skipping generation of SSL certificates as certificate files are present in data directory.
my_sql_db    | 2018-08-24T08:47:59.500382Z 0 [Note] InnoDB: Loading buffer pool(s) from /var/lib/mysql/ib_buffer_pool
my_sql_db    | 2018-08-24T08:47:59.501263Z 0 [Warning] CA certificate ca.pem is self signed.
my_sql_db    | 2018-08-24T08:47:59.522151Z 0 [Note] Skipping generation of RSA key pair as key files are present in data directory.
my_sql_db    | 2018-08-24T08:47:59.531657Z 0 [Note] InnoDB: Buffer pool(s) load completed at 180824  8:47:59

看起来flyway在数据库初始化之前启动，因此无法连接到数据库，我看到下面的错误。

我怎样才能解决这个问题？

我谷歌了类似的问题，我找到了以下建议：https://github.com/vishnubob/wait-for-it但我是docker的新手，我不明白如何将其放入我的docker撰写文件

我试图把文件wait-fot-it.sh附近的撰写文件和执行：

command: ["./wait-for-it.sh", "mysql:3306", "--", "-url=jdbc:mysql://my_sql_db:3306/abhs?useUnicode=true&characterEncoding=utf8&useSSL=false -user=root -password=password migrate"]

但是我返回错误：无效的参数：./wait-for-it.sh

我尝试从"重复"主题的方法：

version: '3'
services:
  my_sql_db:
    image: percona:latest
    container_name: my_sql_db
    environment:
      MYSQL_ROOT_PASSWORD: password
      MYSQL_DATABASE: abhs
    ports:
    - "3306:3306"
    healthcheck:
      test: ["CMD", "mysqladmin" ,"ping", "-h", "localhost"]
      timeout: 20s
      retries: 10
  migration:
    image: boxfuse/flyway:latest
    container_name: flyway_migration
    volumes:
      - ./flyway_scripts/src/main/resources/db/migration:/flyway/sql
    command: -url=jdbc:mysql://my_sql_db:3306/abhs?useUnicode=true&characterEncoding=utf8&useSSL=false -user=root -password=password migrate
    depends_on:
      my_sql_db:
        condition: service_healthy

但我看到以下错误：

$ docker-compose -f docker-compose.yml -f docker-compose-dev.yml up
The Compose file '.\docker-compose-dev.yml' is invalid because:
services.migration.depends_on contains an invalid type, it should be an array

对于这种方法，我看到以下错误：

version: '3'
services:
  my_sql_db:
    image: percona:latest
    container_name: my_sql_db
    environment:
      MYSQL_ROOT_PASSWORD: password
      MYSQL_DATABASE: abhs
    ports:
    - "3306:3306"
    healthcheck:
      test: ["CMD", "mysqladmin" ,"ping", "-h", "localhost"]
      timeout: 20s
      retries: 10
  migration:
    image: boxfuse/flyway:latest
    container_name: flyway_migration
    volumes:
      - ./flyway_scripts/src/main/resources/db/migration:/flyway/sql
    command: dockerize wait jdbc:mysql://my_sql_db:3306 -url=jdbc:mysql://my_sql_db:3306/abhs?useUnicode=true&characterEncoding=utf8&useSSL=false -user=root -password=password migrate
    depends_on:
      - my_sql_db

我看到以下错误：

flyway_migration  | ERROR: Invalid argument: dockerize

wait-for-it.sh内容：

#!/usr/bin/env bash
#   Use this script to test if a given TCP host/port are available

cmdname=$(basename $0)

echoerr() { if [[ $QUIET -ne 1 ]]; then echo "$@" 1>&2; fi }

usage()
{
    cat << USAGE >&2
Usage:
    $cmdname host:port [-s] [-t timeout] [-- command args]
    -h HOST | --host=HOST       Host or IP under test
    -p PORT | --port=PORT       TCP port under test
                                Alternatively, you specify the host and port as host:port
    -s | --strict               Only execute subcommand if the test succeeds
    -q | --quiet                Don't output any status messages
    -t TIMEOUT | --timeout=TIMEOUT
                                Timeout in seconds, zero for no timeout
    -- COMMAND ARGS             Execute command with args after the test finishes
USAGE
    exit 1
}

wait_for()
{
    if [[ $TIMEOUT -gt 0 ]]; then
        echoerr "$cmdname: waiting $TIMEOUT seconds for $HOST:$PORT"
    else
        echoerr "$cmdname: waiting for $HOST:$PORT without a timeout"
    fi
    start_ts=$(date +%s)
    while :
    do
        if [[ $ISBUSY -eq 1 ]]; then
            nc -z $HOST $PORT
            result=$?
        else
            (echo > /dev/tcp/$HOST/$PORT) >/dev/null 2>&1
            result=$?
        fi
        if [[ $result -eq 0 ]]; then
            end_ts=$(date +%s)
            echoerr "$cmdname: $HOST:$PORT is available after $((end_ts - start_ts)) seconds"
            break
        fi
        sleep 1
    done
    return $result
}

wait_for_wrapper()
{
    # In order to support SIGINT during timeout: http://unix.stackexchange.com/a/57692
    if [[ $QUIET -eq 1 ]]; then
        timeout $BUSYTIMEFLAG $TIMEOUT $0 --quiet --child --host=$HOST --port=$PORT --timeout=$TIMEOUT &
    else
        timeout $BUSYTIMEFLAG $TIMEOUT $0 --child --host=$HOST --port=$PORT --timeout=$TIMEOUT &
    fi
    PID=$!
    trap "kill -INT -$PID" INT
    wait $PID
    RESULT=$?
    if [[ $RESULT -ne 0 ]]; then
        echoerr "$cmdname: timeout occurred after waiting $TIMEOUT seconds for $HOST:$PORT"
    fi
    return $RESULT
}

# process arguments
while [[ $# -gt 0 ]]
do
    case "$1" in
        *:* )
        hostport=(${1//:/ })
        HOST=${hostport[0]}
        PORT=${hostport[1]}
        shift 1
        ;;
        --child)
        CHILD=1
        shift 1
        ;;
        -q | --quiet)
        QUIET=1
        shift 1
        ;;
        -s | --strict)
        STRICT=1
        shift 1
        ;;
        -h)
        HOST="$2"
        if [[ $HOST == "" ]]; then break; fi
        shift 2
        ;;
        --host=*)
        HOST="${1#*=}"
        shift 1
        ;;
        -p)
        PORT="$2"
        if [[ $PORT == "" ]]; then break; fi
        shift 2
        ;;
        --port=*)
        PORT="${1#*=}"
        shift 1
        ;;
        -t)
        TIMEOUT="$2"
        if [[ $TIMEOUT == "" ]]; then break; fi
        shift 2
        ;;
        --timeout=*)
        TIMEOUT="${1#*=}"
        shift 1
        ;;
        --)
        shift
        CLI=("$@")
        break
        ;;
        --help)
        usage
        ;;
        *)
        echoerr "Unknown argument: $1"
        usage
        ;;
    esac
done

if [[ "$HOST" == "" || "$PORT" == "" ]]; then
    echoerr "Error: you need to provide a host and port to test."
    usage
fi

TIMEOUT=${TIMEOUT:-15}
STRICT=${STRICT:-0}
CHILD=${CHILD:-0}
QUIET=${QUIET:-0}

# check to see if timeout is from busybox?
# check to see if timeout is from busybox?
TIMEOUT_PATH=$(realpath $(which timeout))
if [[ $TIMEOUT_PATH =~ "busybox" ]]; then
        ISBUSY=1
        BUSYTIMEFLAG="-t"
else
        ISBUSY=0
        BUSYTIMEFLAG=""
fi

if [[ $CHILD -gt 0 ]]; then
    wait_for
    RESULT=$?
    exit $RESULT
else
    if [[ $TIMEOUT -gt 0 ]]; then
        wait_for_wrapper
        RESULT=$?
    else
        wait_for
        RESULT=$?
    fi
fi

if [[ $CLI != "" ]]; then
    if [[ $RESULT -ne 0 && $STRICT -eq 1 ]]; then
        echoerr "$cmdname: strict mode, refusing to execute subprocess"
        exit $RESULT
    fi
    exec "${CLI[@]}"
else
    exit $RESULT
fi

我也尝试了这个：

version: '3'
services:
  my_sql_db:
    image: percona:latest
    container_name: my_sql_db
    environment:
      MYSQL_ROOT_PASSWORD: password
      MYSQL_DATABASE: abhs
    ports:
    - "3306:3306"
    healthcheck:
      test: ["CMD", "mysqladmin" ,"ping", "-h", "localhost"]
      timeout: 20s
      retries: 10
  migration:
    image: boxfuse/flyway:latest
    container_name: flyway_migration
    volumes:
     - ./flyway_scripts/src/main/resources/db/migration:/flyway/sql
    entrypoint: ["wait-for-it.sh", "mysql:3306", "--", "docker-entrypoint.sh"]      
    command: -url=jdbc:mysql://my_sql_db:3306/abhs?useUnicode=true&characterEncoding=utf8&useSSL=false -user=root -password=password migrate
    depends_on:
      - my_sql_db

这会导致错误：

正在创建flyway\u迁移。。。错误

ERROR: for flyway_migration  Cannot start service migration: OCI runtime create failed: container_linux.go:348: starting container process caused "exec: \"wait-for-it.sh\": executable file not found in $PATH": unknown

ERROR: for migration  Cannot start service migration: OCI runtime create failed: container_linux.go:348: starting container process caused "exec: \"wait-for-it.sh\": executable file not found in $PATH": unknown
Encountered errors while bringing up the project.

Answer 1

或者，您可以创建一个migrate.dev.sh并使用docker run。。boxfuse/flyway：最新版本等。。

这是一个示例文件

$FLYWAY_PASSWORD=$PGPASSWORD
$FLYWAY_URL=jdbc:postgresql://$DB_HOST:$PORT/mydb
$FLYWAY_USER=postgres

docker run \
  --rm \
  -e FLYWAY_PASSWORD=$FLYWAY_PASSWORD \
  -e FLYWAY_URL=$FLYWAY_URL \
  -e FLYWAY_USER=$FLYWAY_USER \
  -e FLYWAY_SCHEMAS=$FLYWAY_SCHEMAS \
  -v $(pwd)/sql:/flyway/sql \
   boxfuse/flyway:latest $1

然后你可以这样称呼它

migrate.dev.sh info或migrate.dev.sh migrate

假设您的迁移在中。/sql/

Answer 2

使用flyway 5.2.0，可以添加参数connectRetries，该参数指定flyway在1秒间隔内尝试重新连接的最大次数。

command: -connectRetries=20 -url=jdbc:mysql://my_sql_db:3306/abhs?useUnicode=true&characterEncoding=utf8&useSSL=false -user=root -password=password migrate

Flyway文档

Answer 3

对于错误：

Invalid argument: ./wait-for-it.sh

和

Invalid argument: dockerize

这是因为Flyway容器的入口点是Flyway可执行文件，并且您指定的命令内容作为参数附加到入口点。因此，实际上，容器正在运行以下内容：

flyway dockerize ...

或者

flyway wait-for-it.sh ...

这两个参数都不是Flyway命令行的有效参数。

需要更新入口点，正如您在P.S.5中所做的那样。但是，您随后遇到了错误：

"wait-for-it.sh": executable file not found in $PATH"

这是因为wait-for-it.sh（和dockerize）在Flyway容器中不可用。

您可以创建扩展Flyway容器的Dockerfile，然后添加或复制脚本，例如：

FROM boxfuse/flyway:latest

RUN mkdir /flyway/bin

ADD wait-for-it.sh /flyway/bin/wait-for-it.sh

RUN chmod 755 /flyway/bin/wait-for-it.sh

或挂载包含脚本/可执行文件的卷：

version: '3'
services:
  ...
  migration:
    image: boxfuse/flyway:latest
    container_name: flyway_migration
    volumes:
     - ../sql:/flyway/sql
     - ../bin:/flyway/bin
    entrypoint: ["/flyway/bin/dockerize", "-wait", "tcp://my_sql_db:3306", "-timeout", "15s", "--", "flyway"]      
    ...

其中本地目录。/bin包含dockerize（或等待它.sh）。

这应该足以让dockerize/wait-for-it.sh工作。但是，这两个工具都只检查端口是否可用，而不检查数据库本身是否已经准备好为请求提供服务。

也就是说，使用docker-composev2.1depends_on条件语法可能是一个合理的方法。正如你在评论中提到的，这种语法在v3中被删除了，很多人对此不满意。

然而，正如一位Docker开发人员在评论中所说：

如果您不打算使用swarm服务，就没有理由使用v3格式。

另一种方法是扩展Flyway容器以添加自定义MySQL healthcheck脚本，类似于docker compose文档中显示的Postgres脚本：

#!/bin/bash
# wait-for-mysql.sh

set -e

host="$1"
shift
cmd="$@"

until MYSQL_PWD=$MYSQL_ROOT_PASSWORD /usr/bin/mysql --host="$host" --user="root" --execute "SHOW DATABASES;"; do
  >&2 echo "MySQL is unavailable - sleeping"
  sleep 1
done

>&2 echo "MySQL is up - executing command"
exec $cmd

然后创建Dockerfile以扩展Flyway，安装MySQL客户端并添加以下脚本：

FROM boxfuse/flyway:latest

RUN apt-get update && \
    apt-get install -y mysql-client && \
    mkdir /flyway/bin

ADD wait-for-mysql.sh /flyway/bin/wait-for-mysql.sh

RUN chmod 755 /flyway/bin/wait-for-mysql.sh

然后，您可以在撰写文件中使用自定义飞道图像：

version: '3'
services:
  my_sql_percona:
    ...
  migration:
    build: ./flyway_mysql_client
    container_name: flyway_migration
    environment:
      MYSQL_ROOT_PASSWORD: password
    volumes:
     - ../sql:/flyway/sql
    entrypoint: ["bash", "/flyway/bin/wait-for-mysql.sh", "my_sql_percona", "--", "flyway"]      
    command: -url=jdbc:mysql://my_sql_db:3306/abhs?useUnicode=true&characterEncoding=utf8&useSSL=false -user=root -password=password migrate
    depends_on:
      - my_sql_percona

这种方法的缺点是，您需要为每个容器的依赖项扩展每个容器，并使用自定义健康检查脚本。

v2.1依赖于：condition语法似乎已被删除，以支持v3中的重启策略。但是，它们嵌套在deploy部分下，该部分：

仅在部署到具有docker stack deploy的群集时生效，docker compose up和docker compose run会忽略此项。

因此，进一步的选择是放弃docker compose并在docker swarm上运行，如下所示：

向Flyway容器添加故障时的重新启动策略：

version: '3'
services:
  my_sql_percona:
    ...
  migration:
    image: boxfuse/flyway:latest
    ...
    depends_on:
      - my_sql_percona
    deploy:
      restart_policy:
        condition: on-failure

创建群集（本例中为单节点）：

docker swarm init --advertise-addr <your-ip-address>

部署服务：

docker stack deploy --compose-file docker-compose.yml flyway_mysql

然后，swarm每次退出Flyway容器时都会重新启动它，直到它最终成功退出。

虽然这似乎是可行的，但我不确定这是最好的方法。例如，如果Flyway容器由于迁移脚本中的错误而退出，swarm将继续重新启动容器，即使它永远不会成功。

我用这五种不同的方法创建了一个存储库。

就个人而言，我认为我应该使用v2.1方法，因为healthcheck保存在数据库容器中，而不是在依赖它的每个容器中复制。不过，我不需要使用swarm服务，所以选择适合您的服务。：-）