Merge remote-tracking branch 'origin/dev' into PBM-1511-fallback-config-params

Author: boris-ilijic

README.md

@@ -1,7 +1,7 @@
 # Percona Backup for MongoDB
 [![Go Report Card](https://goreportcard.com/badge/github.com/percona/percona-backup-mongodb)](https://goreportcard.com/report/github.com/percona/percona-backup-mongodb) [![CLA assistant](https://cla-assistant.percona.com/readme/badge/percona/percona-backup-mongodb)](https://cla-assistant.percona.com/percona/percona-backup-mongodb)
 
-![PBM logo](backup-mongo.jpeg)
+![PBM logo](percona-backup-for-mongodb.png)
 
 Percona Backup for MongoDB (PBM) is a distributed, low-impact solution for achieving
 consistent backups of MongoDB sharded clusters and replica sets. Percona Backup for MongoDB supports Percona Server for MongoDB and MongoDB Community Edition v5.0 and higher.

backup-mongo.jpeg

@@ -5,6 +5,7 @@ import (
 	"maps"
 	"time"
 
+	"go.mongodb.org/mongo-driver/bson/primitive"
 	"go.mongodb.org/mongo-driver/mongo"
 
 	"github.com/percona/percona-backup-mongodb/pbm/backup"
@@ -110,6 +111,13 @@ func (a *Agent) PITR(ctx context.Context) {
 	l.Printf("starting PITR routine")
 
 	for {
+		select {
+		case <-a.closeCMD:
+			l.Debug(string(ctrl.CmdPITR), "", "", primitive.Timestamp{}, "stopping main loop")
+			return
+		default:
+		}
+
 		err := a.pitr(ctx)
 		if err != nil {
 			// we need epoch just to log pitr err with an extra context

cmd/pbm-agent/pitr.go

@@ -470,22 +470,35 @@ func (s storageStat) String() string {
 		return a.RestoreTS > b.RestoreTS
 	})
 
+	const (
+		statusSuccess = "success"
+		statusFailed  = "failed"
+		statusOngoing = "ongoing"
+	)
+
+	var printStatus string
+
 	for i := range s.Snapshot {
 		ss := &s.Snapshot[i]
 		var status string
 		switch ss.Status {
 		case defs.StatusDone:
 			status = fmt.Sprintf("[restore_to_time: %s]", fmtTS(ss.RestoreTS))
+			printStatus = statusSuccess
 		case defs.StatusCancelled:
 			status = fmt.Sprintf("[!canceled: %s]", fmtTS(ss.RestoreTS))
+			printStatus = statusFailed
 		case defs.StatusError:
 			if errors.Is(ss.Err, errIncompatible) {
 				status = fmt.Sprintf("[incompatible: %s] [%s]", ss.Err.Error(), fmtTS(ss.RestoreTS))
+				printStatus = statusSuccess
 			} else {
 				status = fmt.Sprintf("[ERROR: %s] [%s]", ss.Err.Error(), fmtTS(ss.RestoreTS))
+				printStatus = statusFailed
 			}
 		default:
 			status = fmt.Sprintf("[running: %s / %s]", ss.Status, fmtTS(ss.RestoreTS))
+			printStatus = statusOngoing
 		}
 
 		t := string(ss.Type)
@@ -497,7 +510,7 @@ func (s storageStat) String() string {
 		if ss.StoreName != "" {
 			t += ", *"
 		}
-		ret += fmt.Sprintf("    %s %s <%s> %s\n", ss.Name, storage.PrettySize(ss.Size), t, status)
+		ret += fmt.Sprintf("    %s %s <%s> %s %s\n", ss.Name, storage.PrettySize(ss.Size), t, printStatus, status)
 	}
 
 	if len(s.PITR.Ranges) == 0 {

cmd/pbm/status.go

@@ -55,17 +55,17 @@ func startOplogSlicer(
 				return
 			}
 
-			majority, err := topo.IsWriteMajorityRequested(ctx, m, writeConcern)
-			if err != nil {
-				l.Error("failed to inspect requested majority: %v", err)
+			majority, e := topo.IsWriteMajorityRequested(ctx, m, writeConcern)
+			if e != nil {
+				l.Error("failed to inspect requested majority: %v", e)
 			}
-			currOpTime, err := topo.GetLastWrite(ctx, m, majority)
-			if err != nil {
-				if errors.Is(err, context.Canceled) {
+			currOpTime, e := topo.GetLastWrite(ctx, m, majority)
+			if e != nil {
+				if errors.Is(e, context.Canceled) {
 					return
 				}
 
-				l.Error("failed to get last write: %v", err)
+				l.Error("failed to get last write: %v", e)
 				continue
 			}
 			if !currOpTime.After(startOpTime) {

pbm/backup/slicer.go

@@ -1134,7 +1134,7 @@ func (r *PhysRestore) Snapshot(
 	// the cmd stream anymore and will flood logs with errors on that.
 	l.Info("send to stopAgent chan")
 	if stopAgentC != nil {
-		stopAgentC <- struct{}{}
+		close(stopAgentC)
 	}
 	// anget will be stopped only after we exit this func
 	// so stop heartbeats not to spam logs while the restore is running