mirror of
https://github.com/VictoriaMetrics/VictoriaMetrics.git
synced 2024-12-29 23:30:04 +01:00
264 lines
8.9 KiB
Go
264 lines
8.9 KiB
Go
package vmselect
|
|
|
|
import (
|
|
"flag"
|
|
"fmt"
|
|
"net/http"
|
|
"runtime"
|
|
"strings"
|
|
"time"
|
|
|
|
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmselect/netstorage"
|
|
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmselect/prometheus"
|
|
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmselect/promql"
|
|
"github.com/VictoriaMetrics/VictoriaMetrics/app/vmstorage"
|
|
"github.com/VictoriaMetrics/VictoriaMetrics/lib/fs"
|
|
"github.com/VictoriaMetrics/VictoriaMetrics/lib/httpserver"
|
|
"github.com/VictoriaMetrics/VictoriaMetrics/lib/logger"
|
|
"github.com/VictoriaMetrics/VictoriaMetrics/lib/timerpool"
|
|
"github.com/VictoriaMetrics/metrics"
|
|
)
|
|
|
|
var (
|
|
deleteAuthKey = flag.String("deleteAuthKey", "", "authKey for metrics' deletion via /api/v1/admin/tsdb/delete_series")
|
|
maxConcurrentRequests = flag.Int("search.maxConcurrentRequests", getDefaultMaxConcurrentRequests(), "The maximum number of concurrent search requests. "+
|
|
"It shouldn't be high, since a single request can saturate all the CPU cores. See also `-search.maxQueueDuration`")
|
|
maxQueueDuration = flag.Duration("search.maxQueueDuration", 10*time.Second, "The maximum time the request waits for execution when `-search.maxConcurrentRequests` limit is reached")
|
|
)
|
|
|
|
func getDefaultMaxConcurrentRequests() int {
|
|
n := runtime.GOMAXPROCS(-1)
|
|
if n <= 4 {
|
|
n *= 2
|
|
}
|
|
if n > 16 {
|
|
// A single request can saturate all the CPU cores, so there is no sense
|
|
// in allowing higher number of concurrent requests - they will just contend
|
|
// for unavailable CPU time.
|
|
n = 16
|
|
}
|
|
return n
|
|
}
|
|
|
|
// Init initializes vmselect
|
|
func Init() {
|
|
tmpDirPath := *vmstorage.DataPath + "/tmp"
|
|
fs.RemoveDirContents(tmpDirPath)
|
|
netstorage.InitTmpBlocksDir(tmpDirPath)
|
|
promql.InitRollupResultCache(*vmstorage.DataPath + "/cache/rollupResult")
|
|
|
|
concurrencyCh = make(chan struct{}, *maxConcurrentRequests)
|
|
}
|
|
|
|
// Stop stops vmselect
|
|
func Stop() {
|
|
promql.StopRollupResultCache()
|
|
}
|
|
|
|
var concurrencyCh chan struct{}
|
|
|
|
var (
|
|
concurrencyLimitReached = metrics.NewCounter(`vm_concurrent_select_limit_reached_total`)
|
|
concurrencyLimitTimeout = metrics.NewCounter(`vm_concurrent_select_limit_timeout_total`)
|
|
|
|
_ = metrics.NewGauge(`vm_concurrent_select_capacity`, func() float64 {
|
|
return float64(cap(concurrencyCh))
|
|
})
|
|
_ = metrics.NewGauge(`vm_concurrent_select_current`, func() float64 {
|
|
return float64(len(concurrencyCh))
|
|
})
|
|
)
|
|
|
|
// RequestHandler handles remote read API requests for Prometheus
|
|
func RequestHandler(w http.ResponseWriter, r *http.Request) bool {
|
|
// Limit the number of concurrent queries.
|
|
select {
|
|
case concurrencyCh <- struct{}{}:
|
|
defer func() { <-concurrencyCh }()
|
|
default:
|
|
// Sleep for a while until giving up. This should resolve short bursts in requests.
|
|
concurrencyLimitReached.Inc()
|
|
t := timerpool.Get(*maxQueueDuration)
|
|
select {
|
|
case concurrencyCh <- struct{}{}:
|
|
timerpool.Put(t)
|
|
defer func() { <-concurrencyCh }()
|
|
case <-t.C:
|
|
timerpool.Put(t)
|
|
concurrencyLimitTimeout.Inc()
|
|
err := &httpserver.ErrorWithStatusCode{
|
|
Err: fmt.Errorf("cannot handle more than %d concurrent search requests during %s; possible solutions: "+
|
|
"increase `-search.maxQueueDuration`, increase `-search.maxConcurrentRequests`, increase server capacity",
|
|
*maxConcurrentRequests, *maxQueueDuration),
|
|
StatusCode: http.StatusServiceUnavailable,
|
|
}
|
|
httpserver.Errorf(w, "%s", err)
|
|
return true
|
|
}
|
|
}
|
|
|
|
path := strings.Replace(r.URL.Path, "//", "/", -1)
|
|
if strings.HasPrefix(path, "/api/v1/label/") {
|
|
s := r.URL.Path[len("/api/v1/label/"):]
|
|
if strings.HasSuffix(s, "/values") {
|
|
labelValuesRequests.Inc()
|
|
labelName := s[:len(s)-len("/values")]
|
|
httpserver.EnableCORS(w, r)
|
|
if err := prometheus.LabelValuesHandler(labelName, w, r); err != nil {
|
|
labelValuesErrors.Inc()
|
|
sendPrometheusError(w, r, err)
|
|
return true
|
|
}
|
|
return true
|
|
}
|
|
}
|
|
|
|
switch path {
|
|
case "/api/v1/query":
|
|
queryRequests.Inc()
|
|
httpserver.EnableCORS(w, r)
|
|
if err := prometheus.QueryHandler(w, r); err != nil {
|
|
queryErrors.Inc()
|
|
sendPrometheusError(w, r, err)
|
|
return true
|
|
}
|
|
return true
|
|
case "/api/v1/query_range":
|
|
queryRangeRequests.Inc()
|
|
httpserver.EnableCORS(w, r)
|
|
if err := prometheus.QueryRangeHandler(w, r); err != nil {
|
|
queryRangeErrors.Inc()
|
|
sendPrometheusError(w, r, err)
|
|
return true
|
|
}
|
|
return true
|
|
case "/api/v1/series":
|
|
seriesRequests.Inc()
|
|
httpserver.EnableCORS(w, r)
|
|
if err := prometheus.SeriesHandler(w, r); err != nil {
|
|
seriesErrors.Inc()
|
|
sendPrometheusError(w, r, err)
|
|
return true
|
|
}
|
|
return true
|
|
case "/api/v1/series/count":
|
|
seriesCountRequests.Inc()
|
|
httpserver.EnableCORS(w, r)
|
|
if err := prometheus.SeriesCountHandler(w, r); err != nil {
|
|
seriesCountErrors.Inc()
|
|
sendPrometheusError(w, r, err)
|
|
return true
|
|
}
|
|
return true
|
|
case "/api/v1/labels":
|
|
labelsRequests.Inc()
|
|
httpserver.EnableCORS(w, r)
|
|
if err := prometheus.LabelsHandler(w, r); err != nil {
|
|
labelsErrors.Inc()
|
|
sendPrometheusError(w, r, err)
|
|
return true
|
|
}
|
|
return true
|
|
case "/api/v1/labels/count":
|
|
labelsCountRequests.Inc()
|
|
httpserver.EnableCORS(w, r)
|
|
if err := prometheus.LabelsCountHandler(w, r); err != nil {
|
|
labelsCountErrors.Inc()
|
|
sendPrometheusError(w, r, err)
|
|
return true
|
|
}
|
|
return true
|
|
case "/api/v1/export":
|
|
exportRequests.Inc()
|
|
if err := prometheus.ExportHandler(w, r); err != nil {
|
|
exportErrors.Inc()
|
|
httpserver.Errorf(w, "error in %q: %s", r.URL.Path, err)
|
|
return true
|
|
}
|
|
return true
|
|
case "/federate":
|
|
federateRequests.Inc()
|
|
if err := prometheus.FederateHandler(w, r); err != nil {
|
|
federateErrors.Inc()
|
|
httpserver.Errorf(w, "error int %q: %s", r.URL.Path, err)
|
|
return true
|
|
}
|
|
return true
|
|
case "/api/v1/rules":
|
|
// Return dumb placeholder
|
|
rulesRequests.Inc()
|
|
w.Header().Set("Content-Type", "application/json")
|
|
fmt.Fprintf(w, "%s", `{"status":"success","data":{"groups":[]}}`)
|
|
return true
|
|
case "/api/v1/alerts":
|
|
// Return dumb placehloder
|
|
alertsRequests.Inc()
|
|
w.Header().Set("Content-Type", "application/json")
|
|
fmt.Fprintf(w, "%s", `{"status":"success","data":{"alerts":[]}}`)
|
|
return true
|
|
case "/api/v1/admin/tsdb/delete_series":
|
|
deleteRequests.Inc()
|
|
authKey := r.FormValue("authKey")
|
|
if authKey != *deleteAuthKey {
|
|
httpserver.Errorf(w, "invalid authKey %q. It must match the value from -deleteAuthKey command line flag", authKey)
|
|
return true
|
|
}
|
|
if err := prometheus.DeleteHandler(r); err != nil {
|
|
deleteErrors.Inc()
|
|
httpserver.Errorf(w, "error in %q: %s", r.URL.Path, err)
|
|
return true
|
|
}
|
|
w.WriteHeader(http.StatusNoContent)
|
|
return true
|
|
default:
|
|
return false
|
|
}
|
|
}
|
|
|
|
func sendPrometheusError(w http.ResponseWriter, r *http.Request, err error) {
|
|
logger.Errorf("error in %q: %s", r.RequestURI, err)
|
|
|
|
w.Header().Set("Content-Type", "application/json")
|
|
statusCode := http.StatusUnprocessableEntity
|
|
if esc, ok := err.(*httpserver.ErrorWithStatusCode); ok {
|
|
statusCode = esc.StatusCode
|
|
}
|
|
w.WriteHeader(statusCode)
|
|
prometheus.WriteErrorResponse(w, statusCode, err)
|
|
}
|
|
|
|
var (
|
|
labelValuesRequests = metrics.NewCounter(`vm_http_requests_total{path="/api/v1/label/{}/values"}`)
|
|
labelValuesErrors = metrics.NewCounter(`vm_http_request_errors_total{path="/api/v1/label/{}/values"}`)
|
|
|
|
queryRequests = metrics.NewCounter(`vm_http_requests_total{path="/api/v1/query"}`)
|
|
queryErrors = metrics.NewCounter(`vm_http_request_errors_total{path="/api/v1/query"}`)
|
|
|
|
queryRangeRequests = metrics.NewCounter(`vm_http_requests_total{path="/api/v1/query_range"}`)
|
|
queryRangeErrors = metrics.NewCounter(`vm_http_request_errors_total{path="/api/v1/query_range"}`)
|
|
|
|
seriesRequests = metrics.NewCounter(`vm_http_requests_total{path="/api/v1/series"}`)
|
|
seriesErrors = metrics.NewCounter(`vm_http_request_errors_total{path="/api/v1/series"}`)
|
|
|
|
seriesCountRequests = metrics.NewCounter(`vm_http_requests_total{path="/api/v1/series/count"}`)
|
|
seriesCountErrors = metrics.NewCounter(`vm_http_request_errors_total{path="/api/v1/series/count"}`)
|
|
|
|
labelsRequests = metrics.NewCounter(`vm_http_requests_total{path="/api/v1/labels"}`)
|
|
labelsErrors = metrics.NewCounter(`vm_http_request_errors_total{path="/api/v1/labels"}`)
|
|
|
|
labelsCountRequests = metrics.NewCounter(`vm_http_requests_total{path="/api/v1/labels/count"}`)
|
|
labelsCountErrors = metrics.NewCounter(`vm_http_request_errors_total{path="/api/v1/labels/count"}`)
|
|
|
|
deleteRequests = metrics.NewCounter(`vm_http_requests_total{path="/api/v1/admin/tsdb/delete_series"}`)
|
|
deleteErrors = metrics.NewCounter(`vm_http_request_errors_total{path="/api/v1/admin/tsdb/delete_series"}`)
|
|
|
|
exportRequests = metrics.NewCounter(`vm_http_requests_total{path="/api/v1/export"}`)
|
|
exportErrors = metrics.NewCounter(`vm_http_request_errors_total{path="/api/v1/export"}`)
|
|
|
|
federateRequests = metrics.NewCounter(`vm_http_requests_total{path="/federate"}`)
|
|
federateErrors = metrics.NewCounter(`vm_http_request_errors_total{path="/federate"}`)
|
|
|
|
rulesRequests = metrics.NewCounter(`vm_http_requests_total{path="/api/v1/rules"}`)
|
|
alertsRequests = metrics.NewCounter(`vm_http_requests_total{path="/api/v1/alerts"}`)
|
|
)
|