forked from prometheus/alertmanager
/
main.go
479 lines (415 loc) · 15.4 KB
/
main.go
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
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
// Copyright 2015 Prometheus Team
// 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.
package main
import (
"context"
"crypto/md5"
"encoding/binary"
"fmt"
"net"
"net/http"
"net/url"
"os"
"os/signal"
"path/filepath"
"runtime"
"strings"
"sync"
"syscall"
"time"
"github.com/go-kit/kit/log"
"github.com/go-kit/kit/log/level"
"github.com/prometheus/alertmanager/api"
"github.com/prometheus/alertmanager/cluster"
"github.com/prometheus/alertmanager/config"
"github.com/prometheus/alertmanager/dispatch"
"github.com/prometheus/alertmanager/inhibit"
"github.com/prometheus/alertmanager/nflog"
"github.com/prometheus/alertmanager/notify"
"github.com/prometheus/alertmanager/provider/mem"
"github.com/prometheus/alertmanager/silence"
"github.com/prometheus/alertmanager/template"
"github.com/prometheus/alertmanager/types"
"github.com/prometheus/alertmanager/ui"
"github.com/prometheus/client_golang/prometheus"
"github.com/prometheus/client_golang/prometheus/promhttp"
"github.com/prometheus/common/promlog"
"github.com/prometheus/common/route"
"github.com/prometheus/common/version"
"github.com/prometheus/prometheus/pkg/labels"
"gopkg.in/alecthomas/kingpin.v2"
)
var (
configHash = prometheus.NewGauge(prometheus.GaugeOpts{
Name: "alertmanager_config_hash",
Help: "Hash of the currently loaded alertmanager configuration.",
})
configSuccess = prometheus.NewGauge(prometheus.GaugeOpts{
Name: "alertmanager_config_last_reload_successful",
Help: "Whether the last configuration reload attempt was successful.",
})
configSuccessTime = prometheus.NewGauge(prometheus.GaugeOpts{
Name: "alertmanager_config_last_reload_success_timestamp_seconds",
Help: "Timestamp of the last successful configuration reload.",
})
alertsActive prometheus.GaugeFunc
alertsSuppressed prometheus.GaugeFunc
requestDuration = prometheus.NewHistogramVec(
prometheus.HistogramOpts{
Name: "alertmanager_http_request_duration_seconds",
Help: "Histogram of latencies for HTTP requests.",
Buckets: []float64{.05, 0.1, .25, .5, .75, 1, 2, 5, 20, 60},
},
[]string{"handler", "method"},
)
responseSize = prometheus.NewHistogramVec(
prometheus.HistogramOpts{
Name: "alertmanager_http_response_size_bytes",
Help: "Histogram of response size for HTTP requests.",
Buckets: prometheus.ExponentialBuckets(100, 10, 7),
},
[]string{"handler", "method"},
)
)
func init() {
prometheus.MustRegister(configSuccess)
prometheus.MustRegister(configSuccessTime)
prometheus.MustRegister(configHash)
prometheus.MustRegister(requestDuration)
prometheus.MustRegister(responseSize)
prometheus.MustRegister(version.NewCollector("alertmanager"))
}
func instrumentHandler(handlerName string, handler http.HandlerFunc) http.HandlerFunc {
return promhttp.InstrumentHandlerDuration(
requestDuration.MustCurryWith(prometheus.Labels{"handler": handlerName}),
promhttp.InstrumentHandlerResponseSize(
responseSize.MustCurryWith(prometheus.Labels{"handler": handlerName}),
handler,
),
)
}
func newAlertMetricByState(marker types.Marker, st types.AlertState) prometheus.GaugeFunc {
return prometheus.NewGaugeFunc(
prometheus.GaugeOpts{
Name: "alertmanager_alerts",
Help: "How many alerts by state.",
ConstLabels: prometheus.Labels{"state": string(st)},
},
func() float64 {
return float64(marker.Count(st))
},
)
}
func newMarkerMetrics(marker types.Marker) {
alertsActive = newAlertMetricByState(marker, types.AlertStateActive)
alertsSuppressed = newAlertMetricByState(marker, types.AlertStateSuppressed)
prometheus.MustRegister(alertsActive)
prometheus.MustRegister(alertsSuppressed)
}
const defaultClusterAddr = "0.0.0.0:9094"
func main() {
if os.Getenv("DEBUG") != "" {
runtime.SetBlockProfileRate(20)
runtime.SetMutexProfileFraction(20)
}
logLevel := &promlog.AllowedLevel{}
if err := logLevel.Set("info"); err != nil {
panic(err)
}
var (
configFile = kingpin.Flag("config.file", "Alertmanager configuration file name.").Default("alertmanager.yml").String()
dataDir = kingpin.Flag("storage.path", "Base path for data storage.").Default("data/").String()
retention = kingpin.Flag("data.retention", "How long to keep data for.").Default("120h").Duration()
alertGCInterval = kingpin.Flag("alerts.gc-interval", "Interval between alert GC.").Default("30m").Duration()
logLevelString = kingpin.Flag("log.level", "Only log messages with the given severity or above.").Default("info").Enum("debug", "info", "warn", "error")
externalURL = kingpin.Flag("web.external-url", "The URL under which Alertmanager is externally reachable (for example, if Alertmanager is served via a reverse proxy). Used for generating relative and absolute links back to Alertmanager itself. If the URL has a path portion, it will be used to prefix all HTTP endpoints served by Alertmanager. If omitted, relevant URL components will be derived automatically.").String()
routePrefix = kingpin.Flag("web.route-prefix", "Prefix for the internal routes of web endpoints. Defaults to path of --web.external-url.").String()
listenAddress = kingpin.Flag("web.listen-address", "Address to listen on for the web interface and API.").Default(":9093").String()
clusterBindAddr = kingpin.Flag("cluster.listen-address", "Listen address for cluster.").
Default(defaultClusterAddr).String()
clusterAdvertiseAddr = kingpin.Flag("cluster.advertise-address", "Explicit address to advertise in cluster.").String()
peers = kingpin.Flag("cluster.peer", "Initial peers (may be repeated).").Strings()
peerTimeout = kingpin.Flag("cluster.peer-timeout", "Time to wait between peers to send notifications.").Default("15s").Duration()
gossipInterval = kingpin.Flag("cluster.gossip-interval", "Interval between sending gossip messages. By lowering this value (more frequent) gossip messages are propagated across the cluster more quickly at the expense of increased bandwidth.").Default(cluster.DefaultGossipInterval.String()).Duration()
pushPullInterval = kingpin.Flag("cluster.pushpull-interval", "Interval for gossip state syncs. Setting this interval lower (more frequent) will increase convergence speeds across larger clusters at the expense of increased bandwidth usage.").Default(cluster.DefaultPushPullInterval.String()).Duration()
tcpTimeout = kingpin.Flag("cluster.tcp-timeout", "Timeout for establishing a stream connection with a remote node for a full state sync, and for stream read and write operations.").Default(cluster.DefaultTcpTimeout.String()).Duration()
probeTimeout = kingpin.Flag("cluster.probe-timeout", "Timeout to wait for an ack from a probed node before assuming it is unhealthy. This should be set to 99-percentile of RTT (round-trip time) on your network.").Default(cluster.DefaultProbeTimeout.String()).Duration()
probeInterval = kingpin.Flag("cluster.probe-interval", "Interval between random node probes. Setting this lower (more frequent) will cause the cluster to detect failed nodes more quickly at the expense of increased bandwidth usage.").Default(cluster.DefaultProbeInterval.String()).Duration()
settleTimeout = kingpin.Flag("cluster.settle-timeout", "Maximum time to wait for cluster connections to settle before evaluating notifications.").Default(cluster.DefaultPushPullInterval.String()).Duration()
reconnectInterval = kingpin.Flag("cluster.reconnect-interval", "Interval between attempting to reconnect to lost peers.").Default(cluster.DefaultReconnectInterval.String()).Duration()
peerReconnectTimeout = kingpin.Flag("cluster.reconnect-timeout", "Length of time to attempt to reconnect to a lost peer.").Default(cluster.DefaultReconnectTimeout.String()).Duration()
)
kingpin.Version(version.Print("alertmanager"))
kingpin.CommandLine.GetFlag("help").Short('h')
kingpin.Parse()
logLevel.Set(*logLevelString)
logger := promlog.New(*logLevel)
level.Info(logger).Log("msg", "Starting Alertmanager", "version", version.Info())
level.Info(logger).Log("build_context", version.BuildContext())
err := os.MkdirAll(*dataDir, 0777)
if err != nil {
level.Error(logger).Log("msg", "Unable to create data directory", "err", err)
os.Exit(1)
}
var peer *cluster.Peer
if *clusterBindAddr != "" {
peer, err = cluster.Create(
log.With(logger, "component", "cluster"),
prometheus.DefaultRegisterer,
*clusterBindAddr,
*clusterAdvertiseAddr,
*peers,
true,
*pushPullInterval,
*gossipInterval,
*tcpTimeout,
*probeTimeout,
*probeInterval,
)
if err != nil {
level.Error(logger).Log("msg", "unable to initialize gossip mesh", "err", err)
os.Exit(1)
}
}
stopc := make(chan struct{})
var wg sync.WaitGroup
wg.Add(1)
notificationLogOpts := []nflog.Option{
nflog.WithRetention(*retention),
nflog.WithSnapshot(filepath.Join(*dataDir, "nflog")),
nflog.WithMaintenance(15*time.Minute, stopc, wg.Done),
nflog.WithMetrics(prometheus.DefaultRegisterer),
nflog.WithLogger(log.With(logger, "component", "nflog")),
}
notificationLog, err := nflog.New(notificationLogOpts...)
if err != nil {
level.Error(logger).Log("err", err)
os.Exit(1)
}
if peer != nil {
c := peer.AddState("nfl", notificationLog, prometheus.DefaultRegisterer)
notificationLog.SetBroadcast(c.Broadcast)
}
marker := types.NewMarker()
newMarkerMetrics(marker)
silenceOpts := silence.Options{
SnapshotFile: filepath.Join(*dataDir, "silences"),
Retention: *retention,
Logger: log.With(logger, "component", "silences"),
Metrics: prometheus.DefaultRegisterer,
}
silences, err := silence.New(silenceOpts)
if err != nil {
level.Error(logger).Log("err", err)
os.Exit(1)
}
if peer != nil {
c := peer.AddState("sil", silences, prometheus.DefaultRegisterer)
silences.SetBroadcast(c.Broadcast)
}
// Start providers before router potentially sends updates.
wg.Add(1)
go func() {
silences.Maintenance(15*time.Minute, filepath.Join(*dataDir, "silences"), stopc)
wg.Done()
}()
defer func() {
close(stopc)
wg.Wait()
}()
// Peer state listeners have been registered, now we can join and get the initial state.
if peer != nil {
err = peer.Join(
*reconnectInterval,
*peerReconnectTimeout,
)
if err != nil {
level.Warn(logger).Log("msg", "unable to join gossip mesh", "err", err)
}
ctx, cancel := context.WithTimeout(context.Background(), *settleTimeout)
defer func() {
cancel()
if err := peer.Leave(10 * time.Second); err != nil {
level.Warn(logger).Log("msg", "unable to leave gossip mesh", "err", err)
}
}()
go peer.Settle(ctx, *gossipInterval*10)
}
alerts, err := mem.NewAlerts(marker, *alertGCInterval)
if err != nil {
level.Error(logger).Log("err", err)
os.Exit(1)
}
defer alerts.Close()
var (
inhibitor *inhibit.Inhibitor
tmpl *template.Template
pipeline notify.Stage
disp *dispatch.Dispatcher
)
defer disp.Stop()
apiv := api.New(
alerts,
silences,
func(matchers []*labels.Matcher) dispatch.AlertOverview {
return disp.Groups(matchers)
},
marker.Status,
peer,
logger,
)
amURL, err := extURL(*listenAddress, *externalURL)
if err != nil {
level.Error(logger).Log("err", err)
os.Exit(1)
}
waitFunc := func() time.Duration { return 0 }
if peer != nil {
waitFunc = clusterWait(peer, *peerTimeout)
}
timeoutFunc := func(d time.Duration) time.Duration {
if d < notify.MinTimeout {
d = notify.MinTimeout
}
return d + waitFunc()
}
var hash float64
reload := func() (err error) {
level.Info(logger).Log("msg", "Loading configuration file", "file", *configFile)
defer func() {
if err != nil {
level.Error(logger).Log("msg", "Loading configuration file failed", "file", *configFile, "err", err)
configSuccess.Set(0)
} else {
configSuccess.Set(1)
configSuccessTime.Set(float64(time.Now().Unix()))
configHash.Set(hash)
}
}()
conf, plainCfg, err := config.LoadFile(*configFile)
if err != nil {
return err
}
hash = md5HashAsMetricValue(plainCfg)
err = apiv.Update(conf, time.Duration(conf.Global.ResolveTimeout))
if err != nil {
return err
}
tmpl, err = template.FromGlobs(conf.Templates...)
if err != nil {
return err
}
tmpl.ExternalURL = amURL
inhibitor.Stop()
disp.Stop()
inhibitor = inhibit.NewInhibitor(alerts, conf.InhibitRules, marker, logger)
pipeline = notify.BuildPipeline(
conf.Receivers,
tmpl,
waitFunc,
inhibitor,
silences,
notificationLog,
marker,
peer,
logger,
)
disp = dispatch.NewDispatcher(alerts, dispatch.NewRoute(conf.Route, nil), pipeline, marker, timeoutFunc, logger)
go disp.Run()
go inhibitor.Run()
return nil
}
if err := reload(); err != nil {
os.Exit(1)
}
// Make routePrefix default to externalURL path if empty string.
if routePrefix == nil || *routePrefix == "" {
*routePrefix = amURL.Path
}
*routePrefix = "/" + strings.Trim(*routePrefix, "/")
router := route.New().WithInstrumentation(instrumentHandler)
if *routePrefix != "/" {
router = router.WithPrefix(*routePrefix)
}
webReload := make(chan chan error)
ui.Register(router, webReload, logger)
apiv.Register(router.WithPrefix("/api/v1"))
level.Info(logger).Log("msg", "Listening", "address", *listenAddress)
go listen(*listenAddress, router, logger)
var (
hup = make(chan os.Signal)
hupReady = make(chan bool)
term = make(chan os.Signal, 1)
)
signal.Notify(hup, syscall.SIGHUP)
signal.Notify(term, os.Interrupt, syscall.SIGTERM)
go func() {
<-hupReady
for {
select {
case <-hup:
reload()
case errc := <-webReload:
errc <- reload()
}
}
}()
// Wait for reload or termination signals.
close(hupReady) // Unblock SIGHUP handler.
<-term
level.Info(logger).Log("msg", "Received SIGTERM, exiting gracefully...")
}
// clusterWait returns a function that inspects the current peer state and returns
// a duration of one base timeout for each peer with a higher ID than ourselves.
func clusterWait(p *cluster.Peer, timeout time.Duration) func() time.Duration {
return func() time.Duration {
return time.Duration(p.Position()) * timeout
}
}
func extURL(listen, external string) (*url.URL, error) {
if external == "" {
hostname, err := os.Hostname()
if err != nil {
return nil, err
}
_, port, err := net.SplitHostPort(listen)
if err != nil {
return nil, err
}
external = fmt.Sprintf("http://%s:%s/", hostname, port)
}
u, err := url.Parse(external)
if err != nil {
return nil, err
}
ppref := strings.TrimRight(u.Path, "/")
if ppref != "" && !strings.HasPrefix(ppref, "/") {
ppref = "/" + ppref
}
u.Path = ppref
return u, nil
}
func listen(listen string, router *route.Router, logger log.Logger) {
if err := http.ListenAndServe(listen, router); err != nil {
level.Error(logger).Log("msg", "Listen error", "err", err)
os.Exit(1)
}
}
func md5HashAsMetricValue(data []byte) float64 {
sum := md5.Sum(data)
// We only want 48 bits as a float64 only has a 53 bit mantissa.
smallSum := sum[0:6]
var bytes = make([]byte, 8)
copy(bytes, smallSum)
return float64(binary.LittleEndian.Uint64(bytes))
}