prometheus/cmd/prometheus/main.go

730 lines
23 KiB
Go
Raw Normal View History

// Copyright 2015 The Prometheus Authors
// 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.
2016-02-17 14:52:44 -08:00
// The main package for the Prometheus server executable.
package main
import (
"context"
2018-01-16 03:10:54 -08:00
"crypto/md5"
"encoding/json"
"fmt"
2017-06-20 09:48:17 -07:00
"net"
"net/http"
_ "net/http/pprof" // Comment this line to disable pprof endpoint.
2017-06-20 09:48:17 -07:00
"net/url"
"os"
"os/signal"
2017-06-20 08:38:01 -07:00
"path/filepath"
"runtime"
2017-06-20 09:48:17 -07:00
"strings"
"sync"
"syscall"
"time"
2017-08-11 11:45:52 -07:00
"github.com/go-kit/kit/log"
"github.com/go-kit/kit/log/level"
"github.com/golang/glog"
"github.com/oklog/oklog/pkg/group"
2017-06-20 09:48:17 -07:00
"github.com/pkg/errors"
"github.com/prometheus/client_golang/prometheus"
2017-06-20 09:48:17 -07:00
"github.com/prometheus/common/model"
"github.com/prometheus/common/version"
prom_runtime "github.com/prometheus/prometheus/pkg/runtime"
2017-06-20 08:38:01 -07:00
"gopkg.in/alecthomas/kingpin.v2"
"github.com/mwitkow/go-conntrack"
"github.com/prometheus/common/promlog"
promlogflag "github.com/prometheus/common/promlog/flag"
"github.com/prometheus/prometheus/config"
2017-11-25 05:13:54 -08:00
"github.com/prometheus/prometheus/discovery"
sd_config "github.com/prometheus/prometheus/discovery/config"
2016-03-01 03:37:22 -08:00
"github.com/prometheus/prometheus/notifier"
"github.com/prometheus/prometheus/promql"
"github.com/prometheus/prometheus/rules"
"github.com/prometheus/prometheus/scrape"
2017-05-10 02:44:13 -07:00
"github.com/prometheus/prometheus/storage"
"github.com/prometheus/prometheus/storage/remote"
"github.com/prometheus/prometheus/storage/tsdb"
"github.com/prometheus/prometheus/util/strutil"
"github.com/prometheus/prometheus/web"
)
2017-06-20 09:48:17 -07:00
var (
configSuccess = prometheus.NewGauge(prometheus.GaugeOpts{
Name: "prometheus_config_last_reload_successful",
Help: "Whether the last configuration reload attempt was successful.",
2017-06-20 09:48:17 -07:00
})
configSuccessTime = prometheus.NewGauge(prometheus.GaugeOpts{
Name: "prometheus_config_last_reload_success_timestamp_seconds",
Help: "Timestamp of the last successful configuration reload.",
2017-06-20 09:48:17 -07:00
})
)
func init() {
prometheus.MustRegister(version.NewCollector("prometheus"))
}
func main() {
if os.Getenv("DEBUG") != "" {
runtime.SetBlockProfileRate(20)
runtime.SetMutexProfileFraction(20)
}
2017-06-20 09:48:17 -07:00
cfg := struct {
configFile string
2017-06-20 09:48:17 -07:00
localStoragePath string
notifier notifier.Options
notifierTimeout model.Duration
forGracePeriod model.Duration
outageTolerance model.Duration
resendDelay model.Duration
web web.Options
tsdb tsdb.Options
lookbackDelta model.Duration
webTimeout model.Duration
queryTimeout model.Duration
queryConcurrency int
queryMaxSamples int
RemoteFlushDeadline model.Duration
2017-06-20 09:48:17 -07:00
prometheusURL string
promlogConfig promlog.Config
2017-06-20 09:48:17 -07:00
}{
notifier: notifier.Options{
Registerer: prometheus.DefaultRegisterer,
},
promlogConfig: promlog.Config{},
2017-06-20 09:48:17 -07:00
}
2017-06-20 08:38:01 -07:00
a := kingpin.New(filepath.Base(os.Args[0]), "The Prometheus monitoring server")
a.Version(version.Print("prometheus"))
a.HelpFlag.Short('h')
a.Flag("config.file", "Prometheus configuration file path.").
Default("prometheus.yml").StringVar(&cfg.configFile)
2017-10-16 15:00:05 -07:00
a.Flag("web.listen-address", "Address to listen on for UI, API, and telemetry.").
2017-06-20 08:38:01 -07:00
Default("0.0.0.0:9090").StringVar(&cfg.web.ListenAddress)
a.Flag("web.read-timeout",
"Maximum duration before timing out read of the request, and closing idle connections.").
Default("5m").SetValue(&cfg.webTimeout)
a.Flag("web.max-connections", "Maximum number of simultaneous connections.").
Default("512").IntVar(&cfg.web.MaxConnections)
a.Flag("web.external-url",
"The URL under which Prometheus is externally reachable (for example, if Prometheus is served via a reverse proxy). Used for generating relative and absolute links back to Prometheus itself. If the URL has a path portion, it will be used to prefix all HTTP endpoints served by Prometheus. If omitted, relevant URL components will be derived automatically.").
PlaceHolder("<URL>").StringVar(&cfg.prometheusURL)
a.Flag("web.route-prefix",
"Prefix for the internal routes of web endpoints. Defaults to path of --web.external-url.").
PlaceHolder("<path>").StringVar(&cfg.web.RoutePrefix)
a.Flag("web.user-assets", "Path to static asset directory, available at /user.").
PlaceHolder("<path>").StringVar(&cfg.web.UserAssetsPath)
a.Flag("web.enable-lifecycle", "Enable shutdown and reload via HTTP request.").
Default("false").BoolVar(&cfg.web.EnableLifecycle)
2017-06-20 08:38:01 -07:00
2018-03-20 08:58:19 -07:00
a.Flag("web.enable-admin-api", "Enable API endpoints for admin control actions.").
2017-07-10 00:29:41 -07:00
Default("false").BoolVar(&cfg.web.EnableAdminAPI)
2017-06-20 08:38:01 -07:00
a.Flag("web.console.templates", "Path to the console template directory, available at /consoles.").
Default("consoles").StringVar(&cfg.web.ConsoleTemplatesPath)
a.Flag("web.console.libraries", "Path to the console library directory.").
Default("console_libraries").StringVar(&cfg.web.ConsoleLibrariesPath)
a.Flag("web.page-title", "Document title of Prometheus instance.").
Default("Prometheus Time Series Collection and Processing Server").StringVar(&cfg.web.PageTitle)
2017-06-20 08:38:01 -07:00
a.Flag("storage.tsdb.path", "Base path for metrics storage.").
Default("data/").StringVar(&cfg.localStoragePath)
a.Flag("storage.tsdb.min-block-duration", "Minimum duration of a data block before being persisted. For use in testing.").
Hidden().Default("2h").SetValue(&cfg.tsdb.MinBlockDuration)
2017-06-20 08:38:01 -07:00
a.Flag("storage.tsdb.max-block-duration",
"Maximum duration compacted blocks may span. For use in testing. (Defaults to 10% of the retention period).").
Hidden().PlaceHolder("<duration>").SetValue(&cfg.tsdb.MaxBlockDuration)
2017-06-20 08:38:01 -07:00
2018-03-20 09:54:36 -07:00
a.Flag("storage.tsdb.retention", "How long to retain samples in storage.").
2017-06-20 08:38:01 -07:00
Default("15d").SetValue(&cfg.tsdb.Retention)
a.Flag("storage.tsdb.no-lockfile", "Do not create lockfile in data directory.").
Default("false").BoolVar(&cfg.tsdb.NoLockfile)
a.Flag("storage.remote.flush-deadline", "How long to wait flushing sample on shutdown or config reload.").
Default("1m").PlaceHolder("<duration>").SetValue(&cfg.RemoteFlushDeadline)
a.Flag("storage.remote.read-sample-limit", "Maximum overall number of samples to return via the remote read interface, in a single query. 0 means no limit.").
Default("5e7").IntVar(&cfg.web.RemoteReadSampleLimit)
a.Flag("storage.remote.read-concurrent-limit", "Maximum number of concurrent remote read calls. 0 means no limit.").
Default("10").IntVar(&cfg.web.RemoteReadConcurrencyLimit)
a.Flag("rules.alert.for-outage-tolerance", "Max time to tolerate prometheus outage for restoring 'for' state of alert.").
Default("1h").SetValue(&cfg.outageTolerance)
a.Flag("rules.alert.for-grace-period", "Minimum duration between alert and restored 'for' state. This is maintained only for alerts with configured 'for' time greater than grace period.").
Default("10m").SetValue(&cfg.forGracePeriod)
a.Flag("rules.alert.resend-delay", "Minimum amount of time to wait before resending an alert to Alertmanager.").
Default("1m").SetValue(&cfg.resendDelay)
2018-03-20 09:54:36 -07:00
a.Flag("alertmanager.notification-queue-capacity", "The capacity of the queue for pending Alertmanager notifications.").
2017-06-20 08:38:01 -07:00
Default("10000").IntVar(&cfg.notifier.QueueCapacity)
a.Flag("alertmanager.timeout", "Timeout for sending alerts to Alertmanager.").
2017-06-20 09:48:17 -07:00
Default("10s").SetValue(&cfg.notifierTimeout)
2017-06-20 08:38:01 -07:00
a.Flag("query.lookback-delta", "The delta difference allowed for retrieving metrics during expression evaluations.").
Default("5m").SetValue(&cfg.lookbackDelta)
a.Flag("query.timeout", "Maximum time a query may take before being aborted.").
2017-06-20 09:48:17 -07:00
Default("2m").SetValue(&cfg.queryTimeout)
2017-06-20 08:38:01 -07:00
a.Flag("query.max-concurrency", "Maximum number of queries executed concurrently.").
Default("20").IntVar(&cfg.queryConcurrency)
a.Flag("query.max-samples", "Maximum number of samples a single query can load into memory. Note that queries will fail if they would load more samples than this into memory, so this also limits the number of samples a query can return.").
Default("50000000").IntVar(&cfg.queryMaxSamples)
2017-06-20 08:38:01 -07:00
promlogflag.AddFlags(a, &cfg.promlogConfig)
2017-06-20 09:48:17 -07:00
_, err := a.Parse(os.Args[1:])
if err != nil {
fmt.Fprintln(os.Stderr, errors.Wrapf(err, "Error parsing commandline arguments"))
2017-06-20 08:38:01 -07:00
a.Usage(os.Args[1:])
os.Exit(2)
}
2017-06-20 09:48:17 -07:00
cfg.web.ExternalURL, err = computeExternalURL(cfg.prometheusURL, cfg.web.ListenAddress)
if err != nil {
fmt.Fprintln(os.Stderr, errors.Wrapf(err, "parse external URL %q", cfg.prometheusURL))
2017-06-20 08:38:01 -07:00
os.Exit(2)
}
2017-06-20 09:48:17 -07:00
cfg.web.ReadTimeout = time.Duration(cfg.webTimeout)
// Default -web.route-prefix to path of -web.external-url.
if cfg.web.RoutePrefix == "" {
cfg.web.RoutePrefix = cfg.web.ExternalURL.Path
}
// RoutePrefix must always be at least '/'.
cfg.web.RoutePrefix = "/" + strings.Trim(cfg.web.RoutePrefix, "/")
2017-06-20 09:48:17 -07:00
if cfg.tsdb.MaxBlockDuration == 0 {
cfg.tsdb.MaxBlockDuration = cfg.tsdb.Retention / 10
}
2017-06-20 09:48:17 -07:00
promql.LookbackDelta = time.Duration(cfg.lookbackDelta)
promql.SetDefaultEvaluationInterval(time.Duration(config.DefaultGlobalConfig.EvaluationInterval))
2017-06-20 09:48:17 -07:00
logger := promlog.New(&cfg.promlogConfig)
2017-08-11 11:45:52 -07:00
// Above level 6, the k8s client would log bearer tokens in clear-text.
glog.ClampLevel(6)
glog.SetLogger(log.With(logger, "component", "k8s_client_runtime"))
level.Info(logger).Log("msg", "Starting Prometheus", "version", version.Info())
2017-08-11 11:45:52 -07:00
level.Info(logger).Log("build_context", version.BuildContext())
level.Info(logger).Log("host_details", prom_runtime.Uname())
level.Info(logger).Log("fd_limits", prom_runtime.FdLimits())
level.Info(logger).Log("vm_limits", prom_runtime.VmLimits())
var (
localStorage = &tsdb.ReadyStorage{}
remoteStorage = remote.NewStorage(log.With(logger, "component", "remote"), localStorage.StartTime, time.Duration(cfg.RemoteFlushDeadline))
fanoutStorage = storage.NewFanout(logger, localStorage, remoteStorage)
2017-08-11 11:45:52 -07:00
)
var (
ctxWeb, cancelWeb = context.WithCancel(context.Background())
ctxRule = context.Background()
notifierManager = notifier.NewManager(&cfg.notifier, log.With(logger, "component", "notifier"))
ctxScrape, cancelScrape = context.WithCancel(context.Background())
discoveryManagerScrape = discovery.NewManager(ctxScrape, log.With(logger, "component", "discovery manager scrape"), discovery.Name("scrape"))
ctxNotify, cancelNotify = context.WithCancel(context.Background())
discoveryManagerNotify = discovery.NewManager(ctxNotify, log.With(logger, "component", "discovery manager notify"), discovery.Name("notify"))
scrapeManager = scrape.NewManager(log.With(logger, "component", "scrape manager"), fanoutStorage)
opts = promql.EngineOpts{
Logger: log.With(logger, "component", "query engine"),
Reg: prometheus.DefaultRegisterer,
MaxConcurrent: cfg.queryConcurrency,
MaxSamples: cfg.queryMaxSamples,
Timeout: time.Duration(cfg.queryTimeout),
}
queryEngine = promql.NewEngine(opts)
ruleManager = rules.NewManager(&rules.ManagerOptions{
Appendable: fanoutStorage,
TSDB: localStorage,
QueryFunc: rules.EngineQueryFunc(queryEngine, fanoutStorage),
NotifyFunc: sendAlerts(notifierManager, cfg.web.ExternalURL.String()),
Context: ctxRule,
ExternalURL: cfg.web.ExternalURL,
Registerer: prometheus.DefaultRegisterer,
Logger: log.With(logger, "component", "rule manager"),
OutageTolerance: time.Duration(cfg.outageTolerance),
ForGracePeriod: time.Duration(cfg.forGracePeriod),
ResendDelay: time.Duration(cfg.resendDelay),
})
)
cfg.web.Context = ctxWeb
cfg.web.TSDB = localStorage.Get
cfg.web.Storage = fanoutStorage
cfg.web.QueryEngine = queryEngine
2017-11-25 05:13:54 -08:00
cfg.web.ScrapeManager = scrapeManager
cfg.web.RuleManager = ruleManager
cfg.web.Notifier = notifierManager
cfg.web.Version = &web.PrometheusVersion{
Version: version.Version,
Revision: version.Revision,
Branch: version.Branch,
BuildUser: version.BuildUser,
BuildDate: version.BuildDate,
GoVersion: version.GoVersion,
}
cfg.web.Flags = map[string]string{}
// Exclude kingpin default flags to expose only Prometheus ones.
boilerplateFlags := kingpin.New("", "").Version("")
2017-06-20 08:38:01 -07:00
for _, f := range a.Model().Flags {
if boilerplateFlags.GetFlag(f.Name) != nil {
continue
}
cfg.web.Flags[f.Name] = f.Value.String()
2017-06-20 08:38:01 -07:00
}
// Depends on cfg.web.ScrapeManager so needs to be after cfg.web.ScrapeManager = scrapeManager
webHandler := web.New(log.With(logger, "component", "web"), &cfg.web)
// Monitor outgoing connections on default transport with conntrack.
http.DefaultTransport.(*http.Transport).DialContext = conntrack.NewDialContextFunc(
conntrack.DialWithTracing(),
)
reloaders := []func(cfg *config.Config) error{
remoteStorage.ApplyConfig,
webHandler.ApplyConfig,
// The Scrape and notifier managers need to reload before the Discovery manager as
// they need to read the most updated config when receiving the new targets list.
notifierManager.ApplyConfig,
scrapeManager.ApplyConfig,
func(cfg *config.Config) error {
c := make(map[string]sd_config.ServiceDiscoveryConfig)
for _, v := range cfg.ScrapeConfigs {
c[v.JobName] = v.ServiceDiscoveryConfig
}
return discoveryManagerScrape.ApplyConfig(c)
},
func(cfg *config.Config) error {
c := make(map[string]sd_config.ServiceDiscoveryConfig)
for _, v := range cfg.AlertingConfig.AlertmanagerConfigs {
2018-01-16 03:10:54 -08:00
// AlertmanagerConfigs doesn't hold an unique identifier so we use the config hash as the identifier.
b, err := json.Marshal(v)
if err != nil {
return err
}
c[fmt.Sprintf("%x", md5.Sum(b))] = v.ServiceDiscoveryConfig
}
return discoveryManagerNotify.ApplyConfig(c)
},
func(cfg *config.Config) error {
// Get all rule files matching the configuration oaths.
var files []string
for _, pat := range cfg.RuleFiles {
fs, err := filepath.Glob(pat)
if err != nil {
// The only error can be a bad pattern.
return fmt.Errorf("error retrieving rule files for %s: %s", pat, err)
}
files = append(files, fs...)
}
return ruleManager.Update(time.Duration(cfg.GlobalConfig.EvaluationInterval), files)
},
}
prometheus.MustRegister(configSuccess)
prometheus.MustRegister(configSuccessTime)
// Start all components while we wait for TSDB to open but only load
// initial config and mark ourselves as ready after it completed.
dbOpen := make(chan struct{})
// sync.Once is used to make sure we can close the channel at different execution stages(SIGTERM or when the config is loaded).
type closeOnce struct {
C chan struct{}
once sync.Once
Close func()
}
// Wait until the server is ready to handle reloading.
reloadReady := &closeOnce{
C: make(chan struct{}),
}
reloadReady.Close = func() {
reloadReady.once.Do(func() {
close(reloadReady.C)
})
}
var g group.Group
{
2018-04-01 11:19:30 -07:00
// Termination handler.
term := make(chan os.Signal, 1)
signal.Notify(term, os.Interrupt, syscall.SIGTERM)
cancel := make(chan struct{})
g.Add(
func() error {
// Don't forget to release the reloadReady channel so that waiting blocks can exit normally.
select {
case <-term:
level.Warn(logger).Log("msg", "Received SIGTERM, exiting gracefully...")
reloadReady.Close()
case <-webHandler.Quit():
level.Warn(logger).Log("msg", "Received termination request via web service, exiting gracefully...")
case <-cancel:
reloadReady.Close()
break
}
return nil
},
func(err error) {
close(cancel)
},
)
}
{
2018-04-01 11:19:30 -07:00
// Scrape discovery manager.
g.Add(
func() error {
err := discoveryManagerScrape.Run()
level.Info(logger).Log("msg", "Scrape discovery manager stopped")
return err
},
func(err error) {
level.Info(logger).Log("msg", "Stopping scrape discovery manager...")
cancelScrape()
},
)
}
{
2018-04-01 11:19:30 -07:00
// Notify discovery manager.
g.Add(
func() error {
err := discoveryManagerNotify.Run()
level.Info(logger).Log("msg", "Notify discovery manager stopped")
return err
},
func(err error) {
level.Info(logger).Log("msg", "Stopping notify discovery manager...")
cancelNotify()
},
)
}
{
2018-04-01 11:19:30 -07:00
// Scrape manager.
g.Add(
func() error {
// When the scrape manager receives a new targets list
// it needs to read a valid config for each job.
// It depends on the config being in sync with the discovery manager so
// we wait until the config is fully loaded.
<-reloadReady.C
err := scrapeManager.Run(discoveryManagerScrape.SyncCh())
level.Info(logger).Log("msg", "Scrape manager stopped")
return err
},
func(err error) {
// Scrape manager needs to be stopped before closing the local TSDB
// so that it doesn't try to write samples to a closed storage.
level.Info(logger).Log("msg", "Stopping scrape manager...")
scrapeManager.Stop()
},
)
}
{
2018-04-01 11:19:30 -07:00
// Reload handler.
// Make sure that sighup handler is registered with a redirect to the channel before the potentially
// long and synchronous tsdb init.
hup := make(chan os.Signal, 1)
signal.Notify(hup, syscall.SIGHUP)
cancel := make(chan struct{})
g.Add(
func() error {
<-reloadReady.C
for {
select {
case <-hup:
if err := reloadConfig(cfg.configFile, logger, reloaders...); err != nil {
level.Error(logger).Log("msg", "Error reloading config", "err", err)
}
case rc := <-webHandler.Reload():
if err := reloadConfig(cfg.configFile, logger, reloaders...); err != nil {
level.Error(logger).Log("msg", "Error reloading config", "err", err)
rc <- err
} else {
rc <- nil
}
case <-cancel:
return nil
}
}
},
func(err error) {
// Wait for any in-progress reloads to complete to avoid
// reloading things after they have been shutdown.
cancel <- struct{}{}
},
)
}
{
2018-04-01 11:19:30 -07:00
// Initial configuration loading.
cancel := make(chan struct{})
g.Add(
func() error {
select {
case <-dbOpen:
break
// In case a shutdown is initiated before the dbOpen is released
case <-cancel:
reloadReady.Close()
return nil
}
if err := reloadConfig(cfg.configFile, logger, reloaders...); err != nil {
return fmt.Errorf("error loading config from %q: %s", cfg.configFile, err)
}
reloadReady.Close()
webHandler.Ready()
level.Info(logger).Log("msg", "Server is ready to receive web requests.")
<-cancel
return nil
},
func(err error) {
close(cancel)
},
)
}
{
// Rule manager.
// TODO(krasi) refactor ruleManager.Run() to be blocking to avoid using an extra blocking channel.
cancel := make(chan struct{})
g.Add(
func() error {
<-reloadReady.C
ruleManager.Run()
<-cancel
return nil
},
func(err error) {
ruleManager.Stop()
close(cancel)
},
)
}
{
2018-04-01 11:19:30 -07:00
// TSDB.
cancel := make(chan struct{})
g.Add(
func() error {
level.Info(logger).Log("msg", "Starting TSDB ...")
db, err := tsdb.Open(
cfg.localStoragePath,
log.With(logger, "component", "tsdb"),
prometheus.DefaultRegisterer,
&cfg.tsdb,
)
if err != nil {
return fmt.Errorf("opening storage failed: %s", err)
}
level.Info(logger).Log("msg", "TSDB started")
startTimeMargin := int64(2 * time.Duration(cfg.tsdb.MinBlockDuration).Seconds() * 1000)
localStorage.Set(db, startTimeMargin)
close(dbOpen)
<-cancel
return nil
},
func(err error) {
if err := fanoutStorage.Close(); err != nil {
level.Error(logger).Log("msg", "Error stopping storage", "err", err)
}
close(cancel)
},
)
}
{
2018-04-01 11:19:30 -07:00
// Web handler.
g.Add(
func() error {
2017-11-25 05:13:54 -08:00
if err := webHandler.Run(ctxWeb); err != nil {
return fmt.Errorf("error starting web server: %s", err)
}
return nil
},
func(err error) {
2017-11-25 05:13:54 -08:00
cancelWeb()
},
)
}
{
2018-04-01 11:19:30 -07:00
// Notifier.
// Calling notifier.Stop() before ruleManager.Stop() will cause a panic if the ruleManager isn't running,
// so keep this interrupt after the ruleManager.Stop().
g.Add(
func() error {
// When the notifier manager receives a new targets list
// it needs to read a valid config for each job.
// It depends on the config being in sync with the discovery manager
// so we wait until the config is fully loaded.
<-reloadReady.C
notifierManager.Run(discoveryManagerNotify.SyncCh())
level.Info(logger).Log("msg", "Notifier manager stopped")
return nil
},
func(err error) {
notifierManager.Stop()
},
)
}
if err := g.Run(); err != nil {
level.Error(logger).Log("err", err)
os.Exit(1)
}
2017-08-11 11:45:52 -07:00
level.Info(logger).Log("msg", "See you next time!")
}
func reloadConfig(filename string, logger log.Logger, rls ...func(*config.Config) error) (err error) {
2017-08-11 11:45:52 -07:00
level.Info(logger).Log("msg", "Loading configuration file", "filename", filename)
defer func() {
if err == nil {
configSuccess.Set(1)
configSuccessTime.SetToCurrentTime()
} else {
configSuccess.Set(0)
}
}()
conf, err := config.LoadFile(filename)
if err != nil {
return fmt.Errorf("couldn't load configuration (--config.file=%q): %v", filename, err)
}
failed := false
for _, rl := range rls {
if err := rl(conf); err != nil {
2017-08-11 11:45:52 -07:00
level.Error(logger).Log("msg", "Failed to apply configuration", "err", err)
failed = true
}
}
if failed {
return fmt.Errorf("one or more errors occurred while applying the new configuration (--config.file=%q)", filename)
}
promql.SetDefaultEvaluationInterval(time.Duration(conf.GlobalConfig.EvaluationInterval))
level.Info(logger).Log("msg", "Completed loading of configuration file", "filename", filename)
return nil
}
2017-06-20 09:48:17 -07:00
2017-10-05 03:16:15 -07:00
func startsOrEndsWithQuote(s string) bool {
return strings.HasPrefix(s, "\"") || strings.HasPrefix(s, "'") ||
strings.HasSuffix(s, "\"") || strings.HasSuffix(s, "'")
}
2017-06-20 09:48:17 -07:00
// computeExternalURL computes a sanitized external URL from a raw input. It infers unset
// URL parts from the OS and the given listen address.
func computeExternalURL(u, listenAddr string) (*url.URL, error) {
if u == "" {
hostname, err := os.Hostname()
if err != nil {
return nil, err
}
_, port, err := net.SplitHostPort(listenAddr)
if err != nil {
return nil, err
}
u = fmt.Sprintf("http://%s:%s/", hostname, port)
}
2017-10-05 03:16:15 -07:00
if startsOrEndsWithQuote(u) {
return nil, fmt.Errorf("URL must not begin or end with quotes")
2017-06-20 09:48:17 -07:00
}
eu, err := url.Parse(u)
if err != nil {
return nil, err
}
ppref := strings.TrimRight(eu.Path, "/")
if ppref != "" && !strings.HasPrefix(ppref, "/") {
ppref = "/" + ppref
}
eu.Path = ppref
return eu, nil
}
type sender interface {
Send(alerts ...*notifier.Alert)
}
// sendAlerts implements the rules.NotifyFunc for a Notifier.
func sendAlerts(s sender, externalURL string) rules.NotifyFunc {
return func(ctx context.Context, expr string, alerts ...*rules.Alert) {
var res []*notifier.Alert
for _, alert := range alerts {
a := &notifier.Alert{
StartsAt: alert.FiredAt,
Labels: alert.Labels,
Annotations: alert.Annotations,
GeneratorURL: externalURL + strutil.TableLinkForExpression(expr),
}
if !alert.ResolvedAt.IsZero() {
a.EndsAt = alert.ResolvedAt
} else {
a.EndsAt = alert.ValidUntil
}
res = append(res, a)
}
if len(alerts) > 0 {
s.Send(res...)
}
}
}