XTLS/Xray-core · warning · errors.Error

unable to get logger instance

Error message

unable to get logger instance

What it means

Returned by the gRPC LoggerService RestartLogger handler when s.V.GetFeature((*log.Instance)(nil)) yields nil — the running core instance has no log app registered, so there is nothing to restart.

Source

Thrown at app/log/command/command.go:21

import (
	"context"

	"github.com/xtls/xray-core/app/log"
	"github.com/xtls/xray-core/common"
	"github.com/xtls/xray-core/common/errors"
	"github.com/xtls/xray-core/core"
	grpc "google.golang.org/grpc"
)

type LoggerServer struct {
	V *core.Instance
}

// RestartLogger implements LoggerService.
func (s *LoggerServer) RestartLogger(ctx context.Context, request *RestartLoggerRequest) (*RestartLoggerResponse, error) {
	logger := s.V.GetFeature((*log.Instance)(nil))
	if logger == nil {
		return nil, errors.New("unable to get logger instance")
	}
	if err := logger.Close(); err != nil {
		return nil, errors.New("failed to close logger").Base(err)
	}
	if err := logger.Start(); err != nil {
		return nil, errors.New("failed to start logger").Base(err)
	}
	return &RestartLoggerResponse{}, nil
}

func (s *LoggerServer) mustEmbedUnimplementedLoggerServiceServer() {}

type service struct {
	v *core.Instance
}

func (s *service) Register(server *grpc.Server) {
	ls := &LoggerServer{

View on GitHub (pinned to 7d214f8b09)

Solutions

1. Enable the log app in the instance config (a log block with a valid loglevel) before calling RestartLogger
2. Have API callers treat this as a benign no-op condition when logging is intentionally disabled
3. Verify with a GetStats/log status call that the log feature exists before restarting

// API clients should first confirm the log feature exists:
// (equivalent server-side check)
if s.V.GetFeature((*log.Instance)(nil)) == nil {
    return nil, errors.New("unable to get logger instance")
}

func hasLogger(v *core.Instance) bool {
    return v.GetFeature((*log.Instance)(nil)) != nil
}

resp, err := loggerClient.RestartLogger(ctx, &RestartLoggerRequest{})
if err != nil && strings.Contains(err.Error(), "unable to get logger instance") {
    // logging disabled on the instance: safe to ignore
}

When it happens

Trigger: Calling the xray API RestartLogger endpoint on an instance whose log app was never registered (log disabled in config or the app omitted from the build).

Common situations: Setting loglevel to 'none'/removing the log section and still issuing API restarts; third-party panels calling RestartLogger unconditionally against minimal instances.

Related errors

• failed to close logger
• failed to start logger
• unsupported router implementation
• AddRule: config type error
• Unable to locate a fake DNS Engine

AI-assisted analysis of XTLS/Xray-core@7d214f8b09 (2026-08-15). Data as JSON: /api/errors/03637cf059991cd6. Report an issue: GitHub.