minio/cmd/callhome.go
Shireesh Anjal fc4561c64c
Start callhome immediately after enabling (#19764)
Currently, on enabling callhome (or restarting the server), the callhome
job gets scheduled. This means that one has to wait for 24hrs (the
default frequency duration) to see it in action and to figure out if it
is working as expected.

It will be a better user experience to perform the first callhome
execution immediately after enabling it (or on server start if already
enabled).

Also, generate audit event on callhome execution, setting the error
field in case the execution has failed.
2024-05-17 09:53:34 -07:00

202 lines
5.3 KiB
Go

// Copyright (c) 2015-2022 MinIO, Inc.
//
// This file is part of MinIO Object Storage stack
//
// This program is free software: you can redistribute it and/or modify
// it under the terms of the GNU Affero General Public License as published by
// the Free Software Foundation, either version 3 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU Affero General Public License for more details.
//
// You should have received a copy of the GNU Affero General Public License
// along with this program. If not, see <http://www.gnu.org/licenses/>.
package cmd
import (
"bytes"
"compress/gzip"
"context"
"encoding/json"
"errors"
"fmt"
"math/rand"
"net/url"
"time"
"github.com/minio/madmin-go/v3"
)
var callhomeLeaderLockTimeout = newDynamicTimeout(30*time.Second, 10*time.Second)
// initCallhome will start the callhome task in the background.
func initCallhome(ctx context.Context, objAPI ObjectLayer) {
if !globalCallhomeConfig.Enabled() {
return
}
go func() {
r := rand.New(rand.NewSource(time.Now().UnixNano()))
// Leader node (that successfully acquires the lock inside runCallhome)
// will keep performing the callhome. If the leader goes down for some reason,
// the lock will be released and another node will acquire it and take over
// because of this loop.
for {
if !globalCallhomeConfig.Enabled() {
return
}
if !runCallhome(ctx, objAPI) {
// callhome was disabled or context was canceled
return
}
// callhome running on a different node.
// sleep for some time and try again.
duration := time.Duration(r.Float64() * float64(globalCallhomeConfig.FrequencyDur()))
if duration < time.Second {
// Make sure to sleep at least a second to avoid high CPU ticks.
duration = time.Second
}
time.Sleep(duration)
}
}()
}
func runCallhome(ctx context.Context, objAPI ObjectLayer) bool {
// Make sure only 1 callhome is running on the cluster.
locker := objAPI.NewNSLock(minioMetaBucket, "callhome/runCallhome.lock")
lkctx, err := locker.GetLock(ctx, callhomeLeaderLockTimeout)
if err != nil {
// lock timedout means some other node is the leader,
// cycle back return 'true'
return true
}
ctx = lkctx.Context()
defer locker.Unlock(lkctx)
// Perform callhome once and then keep running it at regular intervals.
performCallhome(ctx)
callhomeTimer := time.NewTimer(globalCallhomeConfig.FrequencyDur())
defer callhomeTimer.Stop()
for {
if !globalCallhomeConfig.Enabled() {
// Stop the processing as callhome got disabled
return false
}
select {
case <-ctx.Done():
// indicates that we do not need to run callhome anymore
return false
case <-callhomeTimer.C:
if !globalCallhomeConfig.Enabled() {
// Stop the processing as callhome got disabled
return false
}
performCallhome(ctx)
// Reset the timer for next cycle.
callhomeTimer.Reset(globalCallhomeConfig.FrequencyDur())
}
}
}
func performCallhome(ctx context.Context) {
deadline := 10 * time.Second // Default deadline is 10secs for callhome
objectAPI := newObjectLayerFn()
if objectAPI == nil {
internalLogIf(ctx, errors.New("Callhome: object layer not ready"))
return
}
healthCtx, healthCancel := context.WithTimeout(ctx, deadline)
defer healthCancel()
healthInfoCh := make(chan madmin.HealthInfo)
query := url.Values{}
for _, k := range madmin.HealthDataTypesList {
query.Set(string(k), "true")
}
healthInfo := madmin.HealthInfo{
TimeStamp: time.Now().UTC(),
Version: madmin.HealthInfoVersion,
Minio: madmin.MinioHealthInfo{
Info: madmin.MinioInfo{
DeploymentID: globalDeploymentID(),
},
},
}
go fetchHealthInfo(healthCtx, objectAPI, &query, healthInfoCh, healthInfo)
for {
select {
case hi, hasMore := <-healthInfoCh:
if !hasMore {
auditOptions := AuditLogOptions{Event: "callhome:diagnostics"}
// Received all data. Send to SUBNET and return
err := sendHealthInfo(ctx, healthInfo)
if err != nil {
internalLogIf(ctx, fmt.Errorf("Unable to perform callhome: %w", err))
auditOptions.Error = err.Error()
}
auditLogInternal(ctx, auditOptions)
return
}
healthInfo = hi
case <-healthCtx.Done():
return
}
}
}
const (
subnetHealthPath = "/api/health/upload"
)
func sendHealthInfo(ctx context.Context, healthInfo madmin.HealthInfo) error {
url := globalSubnetConfig.BaseURL + subnetHealthPath
filename := fmt.Sprintf("health_%s.json.gz", UTCNow().Format("20060102150405"))
url += "?filename=" + filename
_, err := globalSubnetConfig.Upload(url, filename, createHealthJSONGzip(ctx, healthInfo))
return err
}
func createHealthJSONGzip(ctx context.Context, healthInfo madmin.HealthInfo) []byte {
var b bytes.Buffer
gzWriter := gzip.NewWriter(&b)
header := struct {
Version string `json:"version"`
}{Version: healthInfo.Version}
enc := json.NewEncoder(gzWriter)
if e := enc.Encode(header); e != nil {
internalLogIf(ctx, fmt.Errorf("Could not encode health info header: %w", e))
return nil
}
if e := enc.Encode(healthInfo); e != nil {
internalLogIf(ctx, fmt.Errorf("Could not encode health info: %w", e))
return nil
}
gzWriter.Flush()
gzWriter.Close()
return b.Bytes()
}