mysql binlog解析概要

1,dump协议：

根据数据库的ip+port创建socket,如果创建成功，说明链接建立成功，接下来是使用dump协议订阅binlog

链接建立成功之后,服务端会主动向客户端发送如下问候信息greeting(可以理解为经java转换后，是一个java对象)，

在下面的代码中可以看到greeting中的信息：

this.context.setServerStatus(greeting.getServerStatus());//
this.context.setServerVersion(greeting.getServerVersion().toString());
this.context.setServerCollation(greeting.getServerCollation());
this.context.setServerCapabilities(greeting.getServerCapabilities());
this.context.setThreadId(greeting.getThreadId());
this.context.setProtocolVersion(greeting.getProtocolVersion());
this.context.setScramble(greeting.getScramble1().toString() + greeting.getScramble2().toString());

然后如下：
将ctx=this.context;

final XSerializer s = new XSerializer(64);
s.writeInt(buildClientCapabilities(), 4);
s.writeInt(this.maximumPacketLength, 4);
s.writeInt(this.clientCollation > 0 ? this.clientCollation : ctx.getServerCollation(), 1);
s.writeBytes((byte)0, 23); // Fixed, all 0
s.writeNullTerminatedString(StringColumn.valueOf(this.user.getBytes(this.encoding)));
s.writeInt(20, 1); // the length of the SHA1 encrypted password
s.writeBytes(MySQLUtils.password41OrLater(this.password.getBytes(this.encoding), ctx.getScramble().getBytes(this.encoding)));
if(this.initialSchema != null) s.writeNullTerminatedString(StringColumn.valueOf(this.initialSchema.getBytes(this.encoding)));

//transport可以理解为是一个socket的包装后的东西
final RawPacket request = new RawPacket();
request.setSequence(1);
request.setPacketBody(s.toByteArray());
request.setLength(request.getPacketBody().length);
transport.getOutputStream().writePacket(request);
transport.getOutputStream().flush();

然后可以在下面的代码中得到mysql的相应：

final Packet response = transport.getInputStream().readPacket();
if(response.getPacketBody()[0] == ErrorPacket.PACKET_MARKER) {
   final ErrorPacket error = ErrorPacket.valueOf(response);
   LOGGER.info("login failed, user: {}, error: {}", this.user, error);
   throw new TransportException(error);
} else if(response.getPacketBody()[0] == OKPacket.PACKET_MARKER) {
   final OKPacket ok = OKPacket.valueOf(response);
   LOGGER.info("login successfully, user: {}, detail: {}", this.user, ok);
} else {
   LOGGER.warn("login failed, unknown packet: ", response);
   throw new RuntimeException("assertion failed, invalid packet: " + response);
}

2,dump报文格式：

下方表格来自mysql官网链接：http://dev.mysql.com/doc/internals/en/mysql-packet.html

Type	Name	Description
int<3>	payload_length	Length of the payload. The number of bytes in the packet beyond the initial 4 bytes that make up the packet header.
int<1>	sequence_id	Sequence ID
string<var>	payload	[len=payload_length] payload of the packet

报文对超大packet的支持方式如下：

当一个packet过大 (超过1<<24-1byte ~= 16 MB) 时, 传输需要对packet进行切割, 参看这里

注意, 在A上生成binlog时, 是可以容纳大于16MB的packet的, 也就是原binlog里存在超大的event, 需要在传输时加以限制

切割packet没什么特别之处, 仅需要注意包格式, 一个20MB的event的传输packet格式举例为 (此处用16MB便于描述, 应为1<<24-1byte):

    packet 1
        4字节 packet header
        1字节 值为[00], 是binlog event的特征标志
        16MB-1字节 为第一段数据

    packet 2
        4字节 packet header
        20MB-16MB+1字节 为第二段数据

需要注意的是之后的packet时不带有[00]特征位的. 而包的大小计算范围为除去前4字节的全部字节

上面的讲解对应的代码为：

// Parse packet
final int packetLength = is.readInt(3);
final int packetSequence = is.readInt(1);//超大数据包切割后的顺序
is.setReadLimit(packetLength); // Ensure the packet boundary

//
final int packetMarker = is.readInt(1);//特征位
if(packetMarker != OKPacket.PACKET_MARKER) { // 0x00
   if((byte)packetMarker == ErrorPacket.PACKET_MARKER) {
      final ErrorPacket packet = ErrorPacket.valueOf(packetLength, packetSequence, packetMarker, is);
      throw new RuntimeException(packet.toString());
   } else if((byte)packetMarker == EOFPacket.PACKET_MARKER) {
      final EOFPacket packet = EOFPacket.valueOf(packetLength, packetSequence, packetMarker, is);
      throw new RuntimeException(packet.toString());
   } else {
      throw new RuntimeException("assertion failed, invalid packet marker: " + packetMarker);
   }
}

接下来直接解析mysql binlog即可，header获取如下：

// Parse the event header
final BinlogEventV4HeaderImpl header = new BinlogEventV4HeaderImpl();
header.setTimestamp(is.readLong(4) * 1000L);//timestamp
header.setEventType(is.readInt(1));//type_code
header.setServerId(is.readLong(4));//server_id
header.setEventLength(is.readInt(4));//event_length
header.setNextPosition(is.readLong(4));//next_position
header.setFlags(is.readInt(2));//flags
header.setBinlogFileName(this.binlogFileName);
header.setTimestampOfReceipt(System.currentTimeMillis());

body的解析不同事件格式不一样，类似header那样取值

注：代码来自开源项目open replicator