VictoriaMetrics/app/vlinsert/journald/journald.go

257 lines
8.2 KiB
Go
Raw Permalink Normal View History

package journald
import (
"bytes"
"encoding/binary"
"flag"
"fmt"
"io"
"net/http"
"regexp"
"slices"
"strconv"
"strings"
"time"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlinsert/insertutils"
"github.com/VictoriaMetrics/VictoriaMetrics/app/vlstorage"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/bytesutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/encoding/zstd"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/flagutil"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logstorage"
"github.com/VictoriaMetrics/VictoriaMetrics/lib/writeconcurrencylimiter"
"github.com/VictoriaMetrics/metrics"
)
const (
journaldEntryMaxNameLen = 64
)
var (
bodyBufferPool bytesutil.ByteBufferPool
allowedJournaldEntryNameChars = regexp.MustCompile(`^[A-Z_][A-Z0-9_]+`)
)
var (
journaldStreamFields = flagutil.NewArrayString("journald.streamFields", "Journal fields to be used as stream fields. "+
"See the list of allowed fields at https://www.freedesktop.org/software/systemd/man/latest/systemd.journal-fields.html.")
journaldIgnoreFields = flagutil.NewArrayString("journald.ignoreFields", "Journal fields to ignore. "+
"See the list of allowed fields at https://www.freedesktop.org/software/systemd/man/latest/systemd.journal-fields.html.")
journaldTimeField = flag.String("journald.timeField", "__REALTIME_TIMESTAMP", "Journal field to be used as time field. "+
"See the list of allowed fields at https://www.freedesktop.org/software/systemd/man/latest/systemd.journal-fields.html.")
journaldTenantID = flag.String("journald.tenantID", "0:0", "TenantID for logs ingested via the Journald endpoint.")
journaldIncludeEntryMetadata = flag.Bool("journald.includeEntryMetadata", false, "Include journal entry fields, which with double underscores.")
)
func getCommonParams(r *http.Request) (*insertutils.CommonParams, error) {
cp, err := insertutils.GetCommonParams(r)
if err != nil {
return nil, err
}
if cp.TenantID.AccountID == 0 && cp.TenantID.ProjectID == 0 {
tenantID, err := logstorage.ParseTenantID(*journaldTenantID)
if err != nil {
return nil, fmt.Errorf("cannot parse -journald.tenantID=%q for journald: %w", *journaldTenantID, err)
}
cp.TenantID = tenantID
}
if cp.TimeField != "" {
cp.TimeField = *journaldTimeField
}
if len(cp.StreamFields) == 0 {
cp.StreamFields = *journaldStreamFields
}
if len(cp.IgnoreFields) == 0 {
cp.IgnoreFields = *journaldIgnoreFields
}
cp.MsgFields = []string{"MESSAGE"}
return cp, nil
}
// RequestHandler processes Journald Export insert requests
func RequestHandler(path string, w http.ResponseWriter, r *http.Request) bool {
switch path {
case "/upload":
if r.Header.Get("Content-Type") != "application/vnd.fdo.journal" {
httpserver.Errorf(w, r, "only application/vnd.fdo.journal encoding is supported for Journald")
return true
}
handleJournald(r, w)
return true
default:
return false
}
}
// handleJournald parses Journal binary entries
func handleJournald(r *http.Request, w http.ResponseWriter) {
startTime := time.Now()
requestsJournaldTotal.Inc()
if err := vlstorage.CanWriteData(); err != nil {
httpserver.Errorf(w, r, "%s", err)
return
}
reader := r.Body
var err error
wcr := writeconcurrencylimiter.GetReader(reader)
data, err := io.ReadAll(wcr)
if err != nil {
httpserver.Errorf(w, r, "cannot read request body: %s", err)
return
}
writeconcurrencylimiter.PutReader(wcr)
bb := bodyBufferPool.Get()
defer bodyBufferPool.Put(bb)
if r.Header.Get("Content-Encoding") == "zstd" {
bb.B, err = zstd.Decompress(bb.B[:0], data)
if err != nil {
httpserver.Errorf(w, r, "cannot decompress zstd-encoded request with length %d: %s", len(data), err)
return
}
data = bb.B
}
cp, err := getCommonParams(r)
if err != nil {
httpserver.Errorf(w, r, "cannot parse common params from request: %s", err)
return
}
lmp := cp.NewLogMessageProcessor()
n, err := parseJournaldRequest(data, lmp, cp)
lmp.MustClose()
if err != nil {
errorsTotal.Inc()
httpserver.Errorf(w, r, "cannot parse Journald protobuf request: %s", err)
return
}
rowsIngestedJournaldTotal.Add(n)
// update requestJournaldDuration only for successfully parsed requests
// There is no need in updating requestJournaldDuration for request errors,
// since their timings are usually much smaller than the timing for successful request parsing.
requestJournaldDuration.UpdateDuration(startTime)
}
var (
rowsIngestedJournaldTotal = metrics.NewCounter(`vl_rows_ingested_total{type="journald", format="journald"}`)
requestsJournaldTotal = metrics.NewCounter(`vl_http_requests_total{path="/insert/journald/upload",format="journald"}`)
errorsTotal = metrics.NewCounter(`vl_http_errors_total{path="/insert/journald/upload",format="journald"}`)
requestJournaldDuration = metrics.NewHistogram(`vl_http_request_duration_seconds{path="/insert/journald/upload",format="journald"}`)
)
// See https://systemd.io/JOURNAL_EXPORT_FORMATS/#journal-export-format
func parseJournaldRequest(data []byte, lmp insertutils.LogMessageProcessor, cp *insertutils.CommonParams) (rowsIngested int, err error) {
var fields []logstorage.Field
var ts int64
var size uint64
var name, value string
var line []byte
currentTimestamp := time.Now().UnixNano()
for len(data) > 0 {
idx := bytes.IndexByte(data, '\n')
switch {
case idx > 0:
// process fields
line = data[:idx]
data = data[idx+1:]
case idx == 0:
// next message or end of file
// double new line is a separator for the next message
if len(fields) > 0 {
if ts == 0 {
ts = currentTimestamp
}
lmp.AddRow(ts, fields)
rowsIngested++
fields = fields[:0]
}
// skip newline separator
data = data[1:]
continue
case idx < 0:
return rowsIngested, fmt.Errorf("missing new line separator, unread data left=%d", len(data))
}
idx = bytes.IndexByte(line, '=')
// could b either e key=value\n pair
// or just key\n
// with binary data at the buffer
if idx > 0 {
name = bytesutil.ToUnsafeString(line[:idx])
value = bytesutil.ToUnsafeString(line[idx+1:])
} else {
name = bytesutil.ToUnsafeString(line)
if len(data) == 0 {
return rowsIngested, fmt.Errorf("unexpected zero data for binary field value of key=%s", name)
}
// size of binary data encoded as le i64 at the begging
idx, err := binary.Decode(data, binary.LittleEndian, &size)
if err != nil {
return rowsIngested, fmt.Errorf("failed to extract binary field %q value size: %w", name, err)
}
// skip binary data sise
data = data[idx:]
if size == 0 {
return rowsIngested, fmt.Errorf("unexpected zero binary data size decoded %d", size)
}
if int(size) > len(data) {
return rowsIngested, fmt.Errorf("binary data size=%d cannot exceed size of the data at buffer=%d", size, len(data))
}
value = bytesutil.ToUnsafeString(data[:size])
data = data[int(size):]
// binary data must has new line separator for the new line or next field
if len(data) == 0 {
return rowsIngested, fmt.Errorf("unexpected empty buffer after binary field=%s read", name)
}
lastB := data[0]
if lastB != '\n' {
return rowsIngested, fmt.Errorf("expected new line separator after binary field=%s, got=%s", name, string(lastB))
}
data = data[1:]
}
// https://github.com/systemd/systemd/blob/main/src/libsystemd/sd-journal/journal-file.c#L1703
if len(name) > journaldEntryMaxNameLen {
return rowsIngested, fmt.Errorf("journald entry name should not exceed %d symbols, got: %q", journaldEntryMaxNameLen, name)
}
if !allowedJournaldEntryNameChars.MatchString(name) {
return rowsIngested, fmt.Errorf("journald entry name should consist of `A-Z0-9_` characters and must start from non-digit symbol")
}
if name == cp.TimeField {
ts, err = strconv.ParseInt(value, 10, 64)
if err != nil {
return 0, fmt.Errorf("failed to parse Journald timestamp, %w", err)
}
ts *= 1e3
continue
}
if slices.Contains(cp.MsgFields, name) {
name = "_msg"
}
if *journaldIncludeEntryMetadata || !strings.HasPrefix(name, "__") {
fields = append(fields, logstorage.Field{
Name: name,
Value: value,
})
}
}
if len(fields) > 0 {
if ts == 0 {
ts = currentTimestamp
}
lmp.AddRow(ts, fields)
rowsIngested++
}
return rowsIngested, nil
}