fix: classify WAL restore errors with precise gRPC statuses

Map each restorer sentinel to the gRPC status that best reflects whether
the operator should retry:

  - ErrWALNotFound      -> NotFound        (terminal)
  - ErrInvalidWalName   -> InvalidArgument (terminal)
  - ErrConnectivity     -> Unavailable     (retry)
  - ErrGeneric          -> Unavailable     (retry)
  - anything else       -> Internal        (terminal)

Previously every non-NotFound failure was returned verbatim, leaving the
operator unable to tell transient blips apart from terminal conditions
like a malformed WAL name.

Signed-off-by: Leonardo Cecchi <leonardo.cecchi@enterprisedb.com>

go.mod

@@ -9,7 +9,7 @@ require (
 	github.com/cloudnative-pg/cloudnative-pg v1.29.1
 	github.com/cloudnative-pg/cnpg-i v0.5.0
 	github.com/cloudnative-pg/cnpg-i-machinery v0.4.2
-	github.com/cloudnative-pg/machinery v0.4.0
+	github.com/cloudnative-pg/machinery v0.5.0
 	github.com/onsi/ginkgo/v2 v2.29.0
 	github.com/onsi/gomega v1.41.0
 	github.com/spf13/cobra v1.10.2
@@ -114,7 +114,7 @@ require (
 	golang.org/x/net v0.53.0 // indirect
 	golang.org/x/oauth2 v0.36.0 // indirect
 	golang.org/x/sync v0.20.0 // indirect
-	golang.org/x/sys v0.43.0 // indirect
+	golang.org/x/sys v0.45.0 // indirect
 	golang.org/x/term v0.42.0 // indirect
 	golang.org/x/text v0.36.0 // indirect
 	golang.org/x/time v0.15.0 // indirect
@@ -137,3 +137,5 @@ require (
 	sigs.k8s.io/structured-merge-diff/v6 v6.4.0 // indirect
 	sigs.k8s.io/yaml v1.6.0 // indirect
 )
+
+replace github.com/cloudnative-pg/barman-cloud => github.com/leonardoce/barman-cloud v0.0.0-20260527084420-d2ef1c114ae8

go.sum

@@ -20,16 +20,14 @@ github.com/cespare/xxhash/v2 v2.3.0 h1:UL815xU9SqsFlibzuggzjXhog7bL6oX9BbNZnL2UF
 github.com/cespare/xxhash/v2 v2.3.0/go.mod h1:VGX0DQ3Q6kWi7AoAeZDth3/j3BFtOZR5XLFGgcrjCOs=
 github.com/cloudnative-pg/api v1.29.1 h1:FWr8S7EQeOfdhYXyr2cof8wUXLARZiiQt5Qa6ltED7w=
 github.com/cloudnative-pg/api v1.29.1/go.mod h1:QtWF3yzSvIfORMHaSkPAk/o3bhCJwEHJgN3riyRiz3o=
-github.com/cloudnative-pg/barman-cloud v0.5.1 h1:vjkXrrxo2DQXHT9u9usqhtaHiPZ/lTfDVs/pIWYTepQ=
-github.com/cloudnative-pg/barman-cloud v0.5.1/go.mod h1:XPc5IUFP1y4cZX1sg+Pd8j9V4tmUEVnv3BGCpfQOOg8=
 github.com/cloudnative-pg/cloudnative-pg v1.29.1 h1:ZNEt1TMlnQKXI1kho2UqQuqdfvIvjGln4kN7C1lsmGA=
 github.com/cloudnative-pg/cloudnative-pg v1.29.1/go.mod h1:Sbgx9jVmkle4/gR2U5JHrzDd74sRPOBHDtPkvncg5v8=
 github.com/cloudnative-pg/cnpg-i v0.5.0 h1:/TOzpNT6cwNgrpftTtrnLKdoHgMwd+88vZgXjlVgXeE=
 github.com/cloudnative-pg/cnpg-i v0.5.0/go.mod h1:7Gh4+UzhBpGhr4DreB1GN9wGYfvxwXCXZUyVt3zE/3I=
 github.com/cloudnative-pg/cnpg-i-machinery v0.4.2 h1:0reS9MtyLYINHXQ/MfxJ9jp39hhBf8e3Qdj+T5Nsq6I=
 github.com/cloudnative-pg/cnpg-i-machinery v0.4.2/go.mod h1:gvrKabgxXq0zGthXGucemDdsxakLEQDMxn43M4HLW30=
-github.com/cloudnative-pg/machinery v0.4.0 h1:3sfqrBptH4QQSVB4g10Z+7aiQRnh4g+6AsqsK0ibKaQ=
+github.com/cloudnative-pg/machinery v0.5.0 h1:hhTnkzn+AiN3NmbjCQ6RXj5rfqV3K6arzq6kdXAzcnQ=
-github.com/cloudnative-pg/machinery v0.4.0/go.mod h1:OIwaTYAnLv8PmBmBvEf0BvMK2JBX6J+naTMg9UgV1FQ=
+github.com/cloudnative-pg/machinery v0.5.0/go.mod h1:uuFjqBUjWn0a9uvAk1ixTSzPM0PrjaS+QiKLOIBqLm4=
 github.com/cpuguy83/go-md2man/v2 v2.0.6/go.mod h1:oOW0eioCTA6cOiMLiUPZOpcVxMig6NIQQ7OS05n1F4g=
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
@@ -143,6 +141,8 @@ github.com/kubernetes-csi/external-snapshotter/client/v8 v8.4.0 h1:bMqrb3UHgHbP+
 github.com/kubernetes-csi/external-snapshotter/client/v8 v8.4.0/go.mod h1:E3vdYxHj2C2q6qo8/Da4g7P+IcwqRZyy3gJBzYybV9Y=
 github.com/kylelemons/godebug v1.1.0 h1:RPNrshWIDI6G2gRW9EHilWtl7Z6Sb1BR0xunSBf0SNc=
 github.com/kylelemons/godebug v1.1.0/go.mod h1:9/0rRGxNHcop5bhtWyNeEfOS8JIWk580+fNqagV/RAw=
+github.com/leonardoce/barman-cloud v0.0.0-20260527084420-d2ef1c114ae8 h1:svNVHqtp2a7FfxuLPTm7bs9esj/tAp4arYY/8hvK3no=
+github.com/leonardoce/barman-cloud v0.0.0-20260527084420-d2ef1c114ae8/go.mod h1:DACs5rdlMIQ7mUOpJVBRTrbf+d7OGyMOLsFUEMJ2JV8=
 github.com/lib/pq v1.12.3 h1:tTWxr2YLKwIvK90ZXEw8GP7UFHtcbTtty8zsI+YjrfQ=
 github.com/lib/pq v1.12.3/go.mod h1:/p+8NSbOcwzAEI7wiMXFlgydTwcgTr3OSKMsD2BitpA=
 github.com/maruel/natural v1.1.1 h1:Hja7XhhmvEFhcByqDoHz9QZbkWey+COd9xWfCfn1ioo=
@@ -275,8 +275,8 @@ golang.org/x/oauth2 v0.36.0 h1:peZ/1z27fi9hUOFCAZaHyrpWG5lwe0RJEEEeH0ThlIs=
 golang.org/x/oauth2 v0.36.0/go.mod h1:YDBUJMTkDnJS+A4BP4eZBjCqtokkg1hODuPjwiGPO7Q=
 golang.org/x/sync v0.20.0 h1:e0PTpb7pjO8GAtTs2dQ6jYa5BWYlMuX047Dco/pItO4=
 golang.org/x/sync v0.20.0/go.mod h1:9xrNwdLfx4jkKbNva9FpL6vEN7evnE43NNNJQ2LF3+0=
-golang.org/x/sys v0.43.0 h1:Rlag2XtaFTxp19wS8MXlJwTvoh8ArU6ezoyFsMyCTNI=
+golang.org/x/sys v0.45.0 h1:dO4czNzziLiiXplLQgBCEpCvXQ3dnkn0SdaZSYdQ+FY=
-golang.org/x/sys v0.43.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
+golang.org/x/sys v0.45.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
 golang.org/x/term v0.42.0 h1:UiKe+zDFmJobeJ5ggPwOshJIVt6/Ft0rcfrXZDLWAWY=
 golang.org/x/term v0.42.0/go.mod h1:Dq/D+snpsbazcBG5+F9Q1n2rXV8Ma+71xEjTRufARgY=
 golang.org/x/text v0.36.0 h1:JfKh3XmcRPqZPKevfXVpI1wXPTqbkE5f7JA92a55Yxg=

internal/cnpgi/common/errors.go

@@ -35,10 +35,50 @@ var ErrEndOfWALStreamReached = status.Errorf(codes.OutOfRange, "end of WAL reach
 var ErrMissingPermissions = status.Errorf(codes.FailedPrecondition,
 	"no permission to download the backup credentials, retrying")
 
-// newWALNotFoundError returns a error that states that a
+// newWALNotFoundError returns an error indicating that the
-// certain WAL file has not been found. This error is
+// requested WAL file is not present in the object store.
-// compatible with GRPC status codes, resulting in a 404
+// It carries a gRPC NotFound status so the operator can
-// being used as a response code.
+// treat it as a terminal condition.
 func newWALNotFoundError(walName string) error {
 	return status.Errorf(codes.NotFound, "wal %q not found", walName)
 }
+
+// newUnavailableError returns an error indicating that
+// downloading the given WAL file failed for a transient
+// reason (e.g. a connectivity blip). It carries a gRPC
+// Unavailable status so the operator will retry the
+// request.
+func newUnavailableError(walName string, err error) error {
+	return status.Errorf(
+		codes.Unavailable,
+		"transient error while downloading %q: %s",
+		walName,
+		err.Error(),
+	)
+}
+
+// newInvalidWALNameError returns an error indicating that
+// the requested WAL name is not valid. It carries a gRPC
+// InvalidArgument status so the operator treats it as a
+// terminal condition.
+func newInvalidWALNameError(walName string, err error) error {
+	return status.Errorf(
+		codes.InvalidArgument,
+		"invalid WAL name %q: %s",
+		walName,
+		err.Error(),
+	)
+}
+
+// newInternalWALRestoreError returns an error indicating
+// that downloading the given WAL file failed for an
+// unclassified reason. It carries a gRPC Internal status
+// so the operator treats it as a terminal condition.
+func newInternalWALRestoreError(walName string, err error) error {
+	return status.Errorf(
+		codes.Internal,
+		"internal error while downloading %q: %s",
+		walName,
+		err.Error(),
+	)
+}

internal/cnpgi/common/wal.go

@@ -365,11 +365,28 @@ func (w WALServiceImplementation) restoreFromBarmanObjectStore(
 	// is the one that PostgreSQL has requested to restore.
 	// The failure has already been logged in walRestorer.RestoreList method
 	if walStatus[0].Err != nil {
-		if errors.Is(walStatus[0].Err, barmanRestorer.ErrWALNotFound) {
+		walName := walStatus[0].WalName
-			return newWALNotFoundError(walStatus[0].WalName)
+		walErr := walStatus[0].Err
+		switch {
+		case errors.Is(walErr, barmanRestorer.ErrWALNotFound):
+			// A missing WAL is a terminal condition: surface it as a
+			// gRPC NotFound so the operator stops retrying.
+			return newWALNotFoundError(walName)
+		case errors.Is(walErr, barmanRestorer.ErrInvalidWalName):
+			// A malformed WAL name will never succeed on retry.
+			return newInvalidWALNameError(walName, walErr)
+		case errors.Is(walErr, barmanRestorer.ErrConnectivity),
+			errors.Is(walErr, barmanRestorer.ErrGeneric):
+			// Connectivity and generic barman-cloud failures are
+			// considered transient: surface them as gRPC Unavailable
+			// so the operator retries the download.
+			return newUnavailableError(walName, walErr)
+		default:
+			// Unrecognized exit codes and unexpected failures (e.g. the
+			// barman-cloud command could not be executed) are surfaced
+			// as gRPC Internal so the operator treats them as terminal.
+			return newInternalWALRestoreError(walName, walErr)
 		}
-
-		return walStatus[0].Err
 	}
 
 	// We skip this step if streaming connection is not available