Influxdb 数据写入流程

数据写入流程分析

1. 本篇不涉及存储层的写入，只分析写入请求的处理流程

Influxdb名词介绍

1. 如果想搞清楚Influxdb数据写入流程，Influxdb本身的用法和其一些主要的专用词还是要明白是什么意思，比如measurement, field key,field value, tag key, tag value, tag set, line protocol, point, series, query, retention policy等;
2. 相关的专用名词解释可参考： InfluxDB glossary of terms

分析入口

1. 我们还是以http写请求为入口来分析，在httpd/handler.go中创建Handler时有如下代码:

    Route{
            "write", // Data-ingest route.
            "POST", "/write", true, writeLogEnabled, h.serveWrite,
        }

因此对写入请求的处理就在函数 func (h *Handler) serveWrite(w http.ResponseWriter, r *http.Request, user meta.User)中。

2. Handler.serveWrite流程梳理:
   2.1 获取写入的db并判断db是否存在

database := r.URL.Query().Get("db")
    if database == "" {
        h.httpError(w, "database is required", http.StatusBadRequest)
        return
    }

    if di := h.MetaClient.Database(database); di == nil {
        h.httpError(w, fmt.Sprintf("database not found: %q", database), http.StatusNotFound)
        return
    }

2.2 权限验证

if h.Config.AuthEnabled {
        if user == nil {
            h.httpError(w, fmt.Sprintf("user is required to write to database %q", database), http.StatusForbidden)
            return
        }

        if err := h.WriteAuthorizer.AuthorizeWrite(user.ID(), database); err != nil {
            h.httpError(w, fmt.Sprintf("%q user is not authorized to write to database %q", user.ID(), database), http.StatusForbidden)
            return
        }
    }

2.3 获取http请求的body部分，如需gzip解压缩则解压，并且作body size的校验，因为有body size大小限制

    body := r.Body
    if h.Config.MaxBodySize > 0 {
        body = truncateReader(body, int64(h.Config.MaxBodySize))
    }
    ...
    _, err := buf.ReadFrom(body)

2.4 从http body中解析出 points

points, parseError := models.ParsePointsWithPrecision(buf.Bytes(), time.Now().UTC(),
                       r.URL.Query().Get("precision"))

2.5 将解析出的points写入db

h.PointsWriter.WritePoints(database, r.URL.Query().Get("rp"), consistency, user, points); 

Points的解析

1. 将http body解析成Points是写入前的最主要的一步, 相关内容定义在 models/points.go中;
2. 我们先来看一下一条写入语句是什么样子的： insert test_mea_1,tag1=v1,tag2=v2 cpu=1,memory=10
   其中test_mea_1是measurement, tag key是tag1和tag2, 对应的tag value是v1和v2, field key是cpu和memory, field value是1和10;
3. 先来看下point的定义，它实现了Point interface

type point struct {
    time time.Time

    //这个 key包括了measurement和tag set, 且tag set是排序好的   
    key []byte

    // text encoding of field data
    fields []byte

    // text encoding of timestamp
    ts []byte

    // cached version of parsed fields from data
    cachedFields map[string]interface{}

    // cached version of parsed name from key
    cachedName string

    // cached version of parsed tags
    cachedTags Tags

    //用来遍历所有的field
    it fieldIterator
}

4. 解析出Points

func ParsePointsWithPrecision(buf []byte, defaultTime time.Time, precision string) ([]Point, error) {
    points := make([]Point, 0, bytes.Count(buf, []byte{'\n'})+1)
    var (
        pos    int
        block  []byte
        failed []string
    )
    for pos < len(buf) {
        pos, block = scanLine(buf, pos)
        pos++
  
        ...

        pt, err := parsePoint(block[start:], defaultTime, precision)
        if err != nil {
            failed = append(failed, fmt.Sprintf("unable to parse '%s': %v", string(block[start:]), err))
        } else {
            points = append(points, pt)
        }

    }

    return points, nil
}

这里的解析并没有用正则之类的方案，纯的字符串逐次扫描,这里不详细展开说了.

PointsWriter分析

1. 定义在coordinator/points_writer.go中
2. 主要负责将数据写入到本地的存储,我们重点分析下WritePointsPrivileged

func (w *PointsWriter) WritePointsPrivileged(database, retentionPolicy string, consistencyLevel models.ConsistencyLevel, points []models.Point) error {
    ....
    
    //将point按time对应到相应的Shar上, 这个对应关系存储在shardMappings里, 这个MapShareds我们后面会分析
    shardMappings, err := w.MapShards(&WritePointsRequest{Database: database, RetentionPolicy: retentionPolicy, Points: points})
    if err != nil {
        return err
    }

    // Write each shard in it's own goroutine and return as soon as one fails.
    ch := make(chan error, len(shardMappings.Points))
    for shardID, points := range shardMappings.Points {
    
        // 每个 Shard启动一个goroutine作写入操作, 真正的写入操作w.writeToShard
        go func(shard *meta.ShardInfo, database, retentionPolicy string, points []models.Point) {
            err := w.writeToShard(shard, database, retentionPolicy, points)
            if err == tsdb.ErrShardDeletion {
                err = tsdb.PartialWriteError{Reason: fmt.Sprintf("shard %d is pending deletion", shard.ID), Dropped: len(points)}
            }
            ch <- err
        }(shardMappings.Shards[shardID], database, retentionPolicy, points)
    }
    ...
    
    // 写入超时会return ErrTimeout
    timeout := time.NewTimer(w.WriteTimeout)
    defer timeout.Stop()
    for range shardMappings.Points {
        select {
        case <-w.closing:
            return ErrWriteFailed
        case <-timeout.C:
            atomic.AddInt64(&w.stats.WriteTimeout, 1)
            // return timeout error to caller
            return ErrTimeout
        case err := <-ch:
            if err != nil {
                return err
            }
        }
    }
    return err
}

3. Point到Shard的映谢
   3.1 先根据point的time找到对应的ShardGroup, 没有就创建新的ShardGroup;
   3.2 按Point的key(measurement + tag set取hash)来散

sgi.Shards[hash%uint64(len(sgi.Shards))]