AllTraceSink.java
4.16 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
package com.yoho.trace.online.sink;
import com.alibaba.fastjson.JSONObject;
import com.yoho.trace.anaylzer.model.ApiTraceResult;
import com.yoho.trace.store.HBasePool;
import org.apache.commons.lang3.StringUtils;
import org.apache.flink.api.java.tuple.Tuple2;
import org.apache.flink.configuration.Configuration;
import org.apache.flink.streaming.api.functions.sink.RichSinkFunction;
import org.apache.flink.streaming.api.functions.sink.SinkFunction;
import org.apache.hadoop.hbase.TableName;
import org.apache.hadoop.hbase.client.Connection;
import org.apache.hadoop.hbase.client.Durability;
import org.apache.hadoop.hbase.client.HTable;
import org.apache.hadoop.hbase.client.Put;
import org.apache.hadoop.hbase.util.Bytes;
import java.io.IOException;
/**
* Created by mingdan.ge on 2019/11/11.
*/
public class AllTraceSink extends RichSinkFunction<Tuple2<String, ApiTraceResult>> {
private Connection conn;
private HTable table;
@Override
public void open(Configuration parameters) throws Exception {
conn = HBasePool.getConnection();
table = (HTable) conn.getTable(TableName.valueOf("all_trace"));
table.setWriteBufferSize(1024 * 1024 * 20);
table.setAutoFlush(false, true);//不单个提交
}
@Override
public void invoke(Tuple2<String, ApiTraceResult> value, SinkFunction.Context context) throws Exception {
ApiTraceResult apiTraceResult = value.f1;
String[] md5Tags = StringUtils.split(apiTraceResult.getTraceMd5(), '.');
Put put1 = new Put(Bytes.toBytes( md5Tags[1] + ":" + apiTraceResult.getTraceStartTime()/1000 + ":" + apiTraceResult.getTraceId()));
//其实不推荐关闭WAL，不过关了的确可以提升性能...因为HBase在写数据前会先写WAL，以保证在异常情况下，HBase可以按照WAL的记录来恢复还未持久化的数据。
put1.setDurability(Durability.SKIP_WAL);
put1.addColumn(Bytes.toBytes("data"), Bytes.toBytes("spans"), Bytes.toBytes(JSONObject.toJSONString(apiTraceResult.getSpans())));
put1.addColumn(Bytes.toBytes("data"), Bytes.toBytes("traceid"), Bytes.toBytes(apiTraceResult.getTraceId()));
put1.addColumn(Bytes.toBytes("data"), Bytes.toBytes("starttime"), Bytes.toBytes(apiTraceResult.getTraceStartTime()/1000));
put1.addColumn(Bytes.toBytes("data"), Bytes.toBytes("traceMd5"), Bytes.toBytes(apiTraceResult.getTraceMd5()));
put1.addColumn(Bytes.toBytes("data"), Bytes.toBytes("duration"), Bytes.toBytes(apiTraceResult.getDuration()));
if(StringUtils.isNotEmpty(apiTraceResult.getRegion())){
put1.addColumn(Bytes.toBytes("data"), Bytes.toBytes("region"), Bytes.toBytes(apiTraceResult.getRegion()));
}
Put put2 = new Put(Bytes.toBytes( apiTraceResult.getApiName() + ":" + apiTraceResult.getTraceStartTime()/1000 + ":" + apiTraceResult.getTraceId()));
//其实不推荐关闭WAL，不过关了的确可以提升性能...因为HBase在写数据前会先写WAL，以保证在异常情况下，HBase可以按照WAL的记录来恢复还未持久化的数据。
put2.setDurability(Durability.SKIP_WAL);
put2.addColumn(Bytes.toBytes("data"), Bytes.toBytes("spans"), Bytes.toBytes(JSONObject.toJSONString(apiTraceResult.getSpans())));
put2.addColumn(Bytes.toBytes("data"), Bytes.toBytes("traceid"), Bytes.toBytes(apiTraceResult.getTraceId()));
put2.addColumn(Bytes.toBytes("data"), Bytes.toBytes("starttime"), Bytes.toBytes(apiTraceResult.getTraceStartTime()/1000));
put2.addColumn(Bytes.toBytes("data"), Bytes.toBytes("traceMd5"), Bytes.toBytes(apiTraceResult.getTraceMd5()));
put2.addColumn(Bytes.toBytes("data"), Bytes.toBytes("duration"), Bytes.toBytes(apiTraceResult.getDuration()));
if(StringUtils.isNotEmpty(apiTraceResult.getRegion())){
put2.addColumn(Bytes.toBytes("data"), Bytes.toBytes("region"), Bytes.toBytes(apiTraceResult.getRegion()));
}
table.put(put1);
table.put(put2);
}
@Override
public void close() throws Exception {
try {
table.close();
} catch (IOException e) {
e.printStackTrace();
}
super.close();
conn.close();
}
}