Databus Relays

概述

Databus Relays主要负责以下两个工作:

  • 从databus源数据库中读取变化行,并序列化为事件流保存至内存中;
  • 接受客户端的请求,并将数据变化事件流返回给客户端。

技术架构

 

  • Event Producer:用来读取数据库的变化事件,转化为AVRO类型并存储至内存中;
  • Circular Buffer:Relay有一个或多个环形的缓冲池用来保存按递增的系统变化号(SCN) 为顺序的变化事件;
  • SCN Writer/Reader:用来读取和写入SCN号至硬盘;
  • RESTFUL interface:它暴露一个restful接口,用来推送数据变化事件至客户端。

数据抓取

目前支持Oracle和Mysql两种数据源的抓取。

Oracle数据抓取

抓取Oracle数据是通过给源表添加一个触发器,在新增和修改的时候记录SCN号作为查询的依据,通过relay定期的查询获取变化的数据。删除和查询不受影响。

Oracle数据库配置

首次部署databus,主要有以下几个步骤(非首次部署,只需要从步骤3开始):

  1. 创建databus表空间、用户、赋权限

    createUser.sh
    -- 创建databus表空间,databus用户,并给databus付权限(参见createUser.sql),注意需要指定datafile,如果表空间名字修改了,则需要修改tablespace文件
    create  tablespace TBS_DATABUS datafile  '${DBDIR}/tbs_databus_01.dbf'  size  50M reuse autoextend  on  next  50M maxsize unlimited extent management  local  uniform  size  2M;
    create  tablespace TBS_DATABUS_IDX datafile  '${DBDIR}/tbs_databus_idx_01.dbf'  size  50M reuse autoextend  on  next  50M maxsize unlimited extent management  local  uniform  size  2M;
    create  user  DATABUS identified  by  DATABUS  default  tablespace TBS_DATABUS  temporary  tablespace temp1;
    grant  create  session,  create  table create  view create  sequence create  procedure create  trigger create  type,  create  job   to  DATABUS;
    grant  query rewrite  to  DATABUS;
    grant  execute  on  dbms_alert  to  DATABUS;
    grant  execute  on  sys.dbms_lock  to  DATABUS;
    grant  select  on  sys.v_$ database  to  DATABUS;
    grant  execute  on  sys.dbms_aq  to  DATABUS;
    grant  execute  on  sys.dbms_aqadm  to  DATABUS;
    grant  execute  on  sys.dbms_aqin  to  DATABUS;
    grant  execute  on  sys.dbms_aq_bqview  to  DATABUS;
    alter  user  DATABUS quota unlimited  on  TBS_DATABUS;
    alter  user  DATABUS quota unlimited  on  TBS_DATABUS_IDX;
  2.    创建databus核心包、表、存储过程等必须的东西,使用databus用户,执行oracle/bin/createSchema.sh databus/databus@posp_boss ${Dir}/BOOK

    createSchema.sh
    #
    #
    # Copyright  2013  LinkedIn Corp. All rights reserved
    #
    # Licensed 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.
    #
    #
    usage() {
       echo  "Usage: $0 <username/password@SID> <SRC_VIEW_DIR>"
    }
    # Database Connection Info
    DB=$ 1
    # Log Directory
    SRC_VIEW_DIR=$ 2
    # Check that DB is set
    if  "x$DB"  "x"  ]
    then
       usage
       exit  1
    fi
    user=`echo $DB | perl -lane  '{my $a = $_; $a =~ s/([^\/]*)\/.*/$1/; print $a;}' `;
    password=`echo $DB | perl -lane  '{my $a = $_; $a =~ s/[^\/]*\/([^@]*)\@.*/$1/; print $a;}' `;
    sid=`echo $DB | perl -lane  '{my $a = $_; $a =~ s/[^\/]*\/[^@]*\@(.*)/$1/; print $a;}' `;
    if  "x$user"  "x"  ] || [  "x$password"  "x"  ] || [  "x$sid"  "x"  ]
    then
       echo  "First argument not in format <username/password@SID>"
       usage
       exit  1
    fi
    if  [ ! -d  "$SRC_VIEW_DIR"  ]
    then
       echo  "Src view directory not found or not a directory"
       usage
       exit  1
    fi
    if  [ ! -f  "$SRC_VIEW_DIR/tablespace"  ]
    then
       echo  "Tablespace file ($SRC_VIEW_DIR/tablespace) not provided"
       usage
       exit  1
    fi
    tbs=`cat $SRC_VIEW_DIR/tablespace`;
    tbs_uc=`echo $tbs | tr  '[A-Z]'  '[a-z]' `
    tbs_lc=`echo $tbs | tr  '[a-z]'  '[A-Z]' `
     
    echo  "User : $user"
    echo  "Password : $password"
    echo  "SID : $sid"
    echo  "DB : $DB"
    echo  "TBS : $tbs_uc"
    #
    # create sequenes,tables
    #
    ../fwk/schema/ddl_sqs.sh  "$DB"
    ../fwk/schema/ddl_tab.sh  "$DB"  "$tbs_uc"
    # add column txn
    # for  t in ${SRC_VIEW_DIR}/*\.tab;
    # do
    #  echo  "Processing Table definition File : $t"
    #  table=`echo $t | perl -lane  '{ $a = $_; $a =~ s/.*\/(.*)\.tab/$1/; print $a; }' `
    #  echo  "Creating Table $table. File is $t"
    #  echo  "sqlplus ${DB} " @$t " ${sid} ${user} ${password} ${tbs_lc}"
    #  sqlplus ${DB} << __EOF__
    #    @$t  ${sid} ${user} ${password} ${tbs_lc}
    #__EOF__
    #done
    #
    # create contains/indexes
    #
    ../fwk/schema/ddl_con.sh  "$DB"  "$tbs_uc"
    ../fwk/schema/ddl_ind.sh  "$DB"  "$tbs_uc"
    #
    # crete procedure/packages/functions
    #
    for  t in ${SRC_VIEW_DIR}/*\.tab;
    do
       table=`echo $t | perl -lane  '{ $a = $_; $a =~ s/.*\/(.*)\.tab/$1/; print $a; }' `
       echo  "Setting up Alerts for  $table. File is $t"
       ../fwk/schema/ddl_prc.sh  "$DB"  "$table"
    done
    #
    # create databus views
    #
    ../fwk/schema/ddl_vw.sh  "$DB"
    #
    # create business table's view
    #
    # for  t in ${SRC_VIEW_DIR}/*\.view;
    # do
    #  echo  "Processing View definition File : $t"
    #  view=`echo $t | perl -lane  '{ $a = $_; $a =~ s/.*\/(.*)\.view/$1/; print $a; }' `
    #  echo  "Creating view sy\$$view"
    #  pattern= "sy\\$" ;
    #  pattern+= "${view}" ;
    #  wc=`grep -ic  "${pattern}"  $t`
    #   if  [ $wc -lt  1  ];
    #  then
    #     echo  "View names should start with sy\$. Offending view file ($t)" ;
    #     exit  1
    #  fi
    #  sqlplus ${DB}  << __EOF__
    #     @$t  ${sid} ${user} ${password} ${tbs_lc};
    #__EOF__
    #done
    #
    # create sync_alert
    #
    ## execute rest
    for  t in ${SRC_VIEW_DIR}/*\.tab;
    do
       table=`echo $t | perl -lane  '{ $a = $_; $a =~ s/.*\/(.*)\.tab/$1/; print $a; }' `
       echo  "Creating trigger for source $table"
       ../fwk/schema/ddl_prc1.sh   "$DB"  "$table"
    done
     
    #
    # create business table's trigger
    #
    # for  t in ${SRC_VIEW_DIR}/*\.tab;
    # do
    #  table=`echo $t | perl -lane  '{ $a = $_; $a =~ s/.*\/(.*)\.tab/$1/; print $a; }' `
    #  echo  "Creating trigger for source $table"
    #  ../fwk/schema/ddl_trg.sh  "$DB"  "$table"
    #done
    echo  "INFO: compiling types"
    sqlplus $DB << __EOF__
    exec compile_allobjects;
    __EOF__
     
    echo  "INFO: Sync_core seeding "
    sqlplus $DB << __EOF__
       insert into sync_core_settings values ( 'N' );
       commit;
       show errors;
    __EOF__
     
    #
    # sources setting
    #
    #echo  "INFO: sy\$SOURCES setup"
    #i= 0 ;
    # for  t in ${SRC_VIEW_DIR}/*\.tab;
    # do
    #  echo  "Processing Table definition File : $t"
    #  table=`echo $t | perl -lane  '{ $a = $_; $a =~ s/.*\/(.*)\.tab/$1/; print $a; }' `
    #  echo  "Creating Table $table"
    #  sqlplus $DB << __EOF__
    #    insert into sy\$sources values( '$table' ,$i);
    #__EOF__
    #  ((i++))
    #done
  3. 修改表结构,增加一列 TXN NUMBER(posp_boss)

    person.sql
    -- 创建表,添加TXN列
    create  table  posp_boss.person
    (
             id number  primary  key ,
             first_name  varchar (120)  not  null ,
             last_name  varchar (120)  not  null ,
             birth_date  date ,
             deleted  varchar (5)  default  'false'  not  null ,
             txn number
    );
     
  4. 将源表权限赋给databus(posp_boss)

    grant.sql
    grant  insert , update , select  on  posp_boss.PERSON  to  databus;
  5. 给posp_boss赋databus.sync_core包的执行权限(databus)

    grant.sql
    grant  execute  on  databus.sync_core  to  posp_boss;
  6. 创建索引(posp_boss)

    index.sql
    -- 创建索引(posp_boss)
    create  index  posp_boss.PERSON_txn  on  POSP_BOSS.PERSON(txn) tablespace index_data;
  7. 创建表视图,注意一定要把TXN列包括进去,并且要把ID映射为KEY

    book_vw.sql
    -- 主键ID 映射为 KEY
    CREATE  OR  REPLACE  FORCE  VIEW  sy$person
    AS
    SELECT
       txn,
       id  key ,
       first_name,
       last_name,
       birth_date,
       deleted
    FROM
       posp_boss.person;
  8. 新增sy$sources表配置,注意value的值必须小于等于125

    insert.sql
    -- 注意sourceName区分大小写,对应到上面触发器里面的sync_core.getTxn('POSP_BOSS.PERSON')的POSP_BOSS.PERSON
    insert  into  databus.sy$sources  values ( 'POSP_BOSS.PERSON' ,1);
  9.  创建触发器

    trigger.sql
    -- 注意和sy$sources插入的值一致
    CREATE TRIGGER PERSON_TRG
       before insert or update on POSP_BOSS.PERSON
       referencing old as old  new  as  new
       for  each row
    begin
       if  (updating and : new .txn <  0 ) then
         : new .txn := -: new .txn;
       else
         : new .txn := databus.sync_core.getTxn( 'POSP_BOSS.PERSON' );
       end  if ;
    end;

    至此,针对于Oracle的数据抓取数据端的配置就全部配置完毕了。

     

Mysql数据抓取

Mysql的数据抓取比较简单

  • 创建一个slave的帐号,因为binlog日志分析是基于主从复制的模式来实现的
  • 开启Mysql的binlog日志,设置日志名称,这个名称是后面需要用到的,默认mysql-bin,注意,binlog日志默认是不开启的,开启后需要重启mysql服务
  • 设置binlog日志格式为ROW,默认是STATEMENT。binlog_format  = ROW ,只有ROW模式才会记录受影响的行数,Databus默认只获取影响行数的事件

    my.cnf
    server-id               =  1
    log_bin                 = mysql-bin
    expire_logs_days        =  10
    max_binlog_size         = 100M
    binlog_format           = ROW
  • 配置数据源,注意sources的id必须与sy$sources中的value一致

    sources.json
    {
         "name"  "boss" ,
         "id"   1 ,
         "uri"  "mysql://repl%2F123456@localhost:3306/1/mysql-bin" ,
         "slowSourceQueryThreshold"  2000 ,
         "sources"  :
         [
             { "id"  1 ,
              "name"  "com.linkedin.events.example.person.Person" ,
              "uri" "lijiang.person" ,
              "partitionFunction"  "constant:1"
             }
         ]
    }

     

    uri的格式为:mysql://用户%2F密码@host:port/serverID/binlog文件名称
    另外需要注意sources里对应数据源的uri,必需带上数据库名称,格式为 db.table 

  • 对于Mysql的数据抓取,很多数据类型在Avro序列化时会被转换为string

部署normal_replay

  1. 配置relay sources,sources的id必须与sy$sources的value一致。注意oracle和mysql的配置是不一样的。

    source.json
    # oracle
    {
         "name"  "person" ,
         "id"   1 ,
         "uri"  "jdbc:oracle:thin:lijiang/lijiang@192.168.16.239:51521:afc1" ,
         "slowSourceQueryThreshold"  2000 ,
         "sources"  :
         [
             { "id"  1 ,
              "name"  "com.linkedin.events.example.person.Person" ,
              "uri" "lijiang.person" ,
              "partitionFunction"  "constant:1"
             }
         ]
    }
     
    # mysql
    {
         "name"  "boss" ,
         "id"   1 ,
         "uri"  "mysql://repl%2F123456@localhost:3306/1/mysql-bin" ,
         "slowSourceQueryThreshold"  2000 ,
         "sources"  :
         [
             { "id"  1 ,
              "name"  "com.linkedin.events.example.person.Person" ,
              "uri" "lijiang.person" ,
              "partitionFunction"  "constant:1"
             }
         ]
    }
  2. 添加 avro 配置文件至schemas_registry文件夹中,关于avro的详细结束参见Apache Avro

    book.avsc
    {
       "name"  "Person_V1" ,
       "doc"  "Auto-generated Avro schema for sy$person. Generated at Dec 04, 2012 05:07:05 PM PST" ,
       "type"  "record" ,
       "meta"  "dbFieldName=sy$person;pk=key;" ,
       "namespace"  "com.linkedin.events.example.person" ,
       "fields"  : [ {
         "name"  "txn" ,
         "type"  : [  "long" "null"  ],
         "meta"  "dbFieldName=TXN;dbFieldPosition=0;"
       }, {
         "name"  "key" ,
         "type"  : [  "long" "null"  ],
         "meta"  "dbFieldName=KEY;dbFieldPosition=1;"
       }, {
         "name"  "firstName" ,
         "type"  : [  "string" "null"  ],
         "meta"  "dbFieldName=FIRST_NAME;dbFieldPosition=2;"
       }, {
         "name"  "lastName" ,
         "type"  : [  "string" "null"  ],
         "meta"  "dbFieldName=LAST_NAME;dbFieldPosition=3;"
       }, {
         "name"  "birthDate" ,
         "type"  : [  "long" "null"  ],
         "meta"  "dbFieldName=BIRTH_DATE;dbFieldPosition=4;"
       }, {
         "name"  "deleted" ,
         "type"  : [  "string" "null"  ],
         "meta"  "dbFieldName=DELETED;dbFieldPosition=5;"
       } ]
    }
  3. 启动relay

    starup.sh
    ./bin/startup.sh relay

至此,Relay和数据库都已经配置和部署完成!

评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值