logs

Co-authored-by: Matt Van Horn <455140+mvanhorn@users.noreply.github.com>
Co-authored-by: Quentin McGaw <quentin.mcgaw@gmail.com>

Thanks also to @arturhoo for also trying to resolve this issue in #3410

.github/workflows/ci.yml

@@ -85,13 +85,13 @@ jobs:
     steps:
       - uses: actions/checkout@v6
 
-      - uses: actions/setup-go@v7
+      - uses: actions/setup-go@v6
         with:
           go-version-file: ./devrun/go.mod
       - run: go test ./...
         working-directory: ./devrun
 
-      - uses: actions/setup-go@v7
+      - uses: actions/setup-go@v6
         with:
           go-version-file: ./ci/go.mod
       - run: go test ./...
@@ -114,7 +114,7 @@ jobs:
       - run: docker build -t qmcgaw/gluetun .
 
       - name: Setup Go for CI utility
-        uses: actions/setup-go@v7
+        uses: actions/setup-go@v6
         with:
           go-version-file: ci/go.mod
 
@@ -162,7 +162,7 @@ jobs:
       security-events: write
     steps:
       - uses: actions/checkout@v6
-      - uses: actions/setup-go@v7
+      - uses: actions/setup-go@v6
         with:
           go-version-file: go.mod
       - uses: github/codeql-action/init@v4
@@ -241,3 +241,5 @@ jobs:
             VERSION=${{ fromJSON(steps.meta.outputs.json).labels['org.opencontainers.image.version'] }}
           tags: ${{ steps.meta.outputs.tags }}
           push: true
+          provenance: mode=max
+          sbom: true

internal/healthcheck/checker.go

@@ -8,7 +8,6 @@ import (
 	"net"
 	"net/netip"
 	"strings"
-	"sync"
 	"time"
 
 	"github.com/qdm12/gluetun/internal/healthcheck/dns"
@@ -24,7 +23,6 @@ type Checker struct {
 	icmpTargetIPs  []netip.Addr
 	smallCheckType string
 	startupOnFail  bool
-	configMutex    sync.Mutex
 
 	icmpNotPermitted *bool
 
@@ -55,8 +53,6 @@ func NewChecker(logger Logger) *Checker {
 func (c *Checker) SetConfig(tlsDialAddrs []string, icmpTargets []netip.Addr,
 	smallCheckType string, startupOnFail bool,
 ) {
-	c.configMutex.Lock()
-	defer c.configMutex.Unlock()
 	c.tlsDialAddrs = tlsDialAddrs
 	c.icmpTargetIPs = icmpTargets
 	c.smallCheckType = smallCheckType
@@ -166,10 +162,8 @@ func (c *Checker) Stop() error {
 }
 
 func (c *Checker) smallPeriodicCheck(ctx context.Context) error {
-	c.configMutex.Lock()
 	icmpTargetIPs := make([]netip.Addr, len(c.icmpTargetIPs))
 	copy(icmpTargetIPs, c.icmpTargetIPs)
-	c.configMutex.Unlock()
 	tryTimeouts := []time.Duration{
 		5 * time.Second,
 		5 * time.Second,

internal/portforward/loop.go

@@ -99,6 +99,7 @@ func (l *Loop) run(runCtx context.Context, runDone chan<- struct{},
 		case partialUpdate := <-updateTrigger:
 			updatedSettings, err := l.settings.updateWith(partialUpdate, *l.settings.VPNIsUp)
 			if err != nil {
+				l.logger.Debug(fmt.Sprintf("port forward loop: run: updateWith error %v", err))
 				updateResult <- err
 				continue
 			}
@@ -136,6 +137,7 @@ func (l *Loop) run(runCtx context.Context, runDone chan<- struct{},
 			if err != nil {
 				err = fmt.Errorf("starting port forwarding service: %w", err)
 			}
+			l.logger.Debug(fmt.Sprintf("port forward loop: Start: error is %v", err))
 			updateResult <- err
 		} else if err != nil {
 			// Log the error and schedule a retry
@@ -150,6 +152,7 @@ func (l *Loop) UpdateWith(partialUpdate Settings) (err error) {
 	case l.updateTrigger <- partialUpdate:
 		select {
 		case err = <-l.updatedResult:
+			l.logger.Debug(fmt.Sprintf("port forward loop: UpdateWith: received error %v", err))
 			return err
 		case <-l.runCtx.Done():
 			return l.runCtx.Err()

internal/portforward/service/start.go

@@ -123,7 +123,9 @@ func (s *Service) onNewPorts(ctx context.Context, internalToExternalPorts map[ui
 		}
 	}
 
+	s.logger.Debug(fmt.Sprintf("writePortForwardedFile with externalPorts: %v", externalPorts))
 	err = s.writePortForwardedFile(externalPorts)
+	s.logger.Debug(fmt.Sprintf("writing port file: error is %v", err))
 	if err != nil {
 		_ = s.cleanup()
 		return fmt.Errorf("writing port file: %w", err)

internal/portforward/service/stop.go

@@ -58,6 +58,7 @@ func (s *Service) cleanup() (err error) {
 
 	s.ports = nil
 
+	s.logger.Debug("writePortForwardedFile in cleanup")
 	err = s.writePortForwardedFile(nil)
 	if err != nil {
 		return fmt.Errorf("clearing port file: %w", err)

internal/provider/protonvpn/portforward.go

@@ -133,6 +133,20 @@ func (p *Provider) KeepPortForward(ctx context.Context,
 		case <-timer.C:
 		}
 
+		_, externalIPv4Address, err := client.ExternalAddress(ctx, objects.Gateway)
+		if err != nil {
+			switch {
+			case strings.HasSuffix(err.Error(), "connection refused"):
+				err = fmt.Errorf("%w - make sure you have +pmp at the end of your OpenVPN username "+
+					"or that your Wireguard key is set to work with PMP", err)
+			case strings.Contains(err.Error(), "i/o timeout"):
+				err = fmt.Errorf("%w - make sure FIREWALL_OUTBOUND_SUBNETS does not conflict with "+
+					"the VPN gateway ip address %s", err, objects.Gateway)
+			}
+			logger.Error(fmt.Sprintf("getting external IPv4 address: %v", err))
+		}
+		logger.Debug("gateway external IPv4 address is " + externalIPv4Address.String())
+
 		objects.Logger.Debug("refreshing forwarded ports since 45 seconds have elapsed")
 		const lifetime = 60 * time.Second
 		for internalPort, externalPort := range p.internalToExternalPorts {

internal/vpn/loop.go

@@ -45,6 +45,7 @@ type Loop struct {
 	start       <-chan struct{}
 	running     chan<- models.LoopStatus
 	userTrigger bool
+	healthDone  <-chan struct{}
 	// Internal constant values
 	backoffTime time.Duration
 }
@@ -70,6 +71,11 @@ func NewLoop(vpnSettings settings.VPN, ipv6SupportLevel netlink.IPv6SupportLevel
 	statusManager := loopstate.New(constants.Stopped, start, running, stop, stopped)
 	state := state.New(statusManager, vpnSettings)
 
+	// Initialize healthDone channel to a closed channel so that the first time
+	// we call <-l.healthDone it does not block
+	healthDone := make(chan struct{})
+	close(healthDone)
+
 	return &Loop{
 		statusManager:    statusManager,
 		state:            state,
@@ -98,6 +104,7 @@ func NewLoop(vpnSettings settings.VPN, ipv6SupportLevel netlink.IPv6SupportLevel
 		stop:             stop,
 		stopped:          stopped,
 		userTrigger:      true,
+		healthDone:       healthDone,
 		backoffTime:      defaultBackoffTime,
 	}
 }

internal/vpn/portforward.go

@@ -34,7 +34,9 @@ func (l *Loop) startPortForwarding(data tunnelUpData) (err error) {
 			Password:       data.password,
 		},
 	}
-	return l.portForward.UpdateWith(partialUpdate)
+	err = l.portForward.UpdateWith(partialUpdate)
+	l.logger.Debug(fmt.Sprintf("vpn loop: start port forwarding: error is %v", err))
+	return err
 }
 
 func (l *Loop) stopPortForwarding() (err error) {

internal/vpn/tunnelup.go

@@ -81,6 +81,7 @@ func (l *Loop) onTunnelUp(ctx, loopCtx context.Context, data tunnelUpData) {
 
 	_, _ = l.dnsLooper.ApplyStatus(ctx, constants.Running)
 
+	<-l.healthDone // make sure the health checker is stopped before restarting it
 	icmpTargetIPs := l.healthSettings.ICMPTargetIPs
 	if len(icmpTargetIPs) == 1 && icmpTargetIPs[0].IsUnspecified() {
 		icmpTargetIPs = []netip.Addr{data.serverIP}
@@ -104,7 +105,12 @@ func (l *Loop) onTunnelUp(ctx, loopCtx context.Context, data tunnelUpData) {
 	// Start collecting health errors asynchronously, since
 	// we should not wait for the code below to complete
 	// to start monitoring health and auto-healing.
-	go l.collectHealthErrors(ctx, loopCtx, healthErrCh)
+	// We keep track of when this goroutine is done with the healthDone
+	// channel to avoid a race condition where the health checker is stopped
+	// after being reconfigured and restarted, instead of before.
+	healthDone := make(chan struct{})
+	l.healthDone = healthDone
+	go l.collectHealthErrors(ctx, loopCtx, healthDone, healthErrCh)
 
 	err = l.publicip.RunOnce(ctx)
 	if err != nil {
@@ -140,7 +146,10 @@ func (l *Loop) onTunnelUp(ctx, loopCtx context.Context, data tunnelUpData) {
 	}
 }
 
-func (l *Loop) collectHealthErrors(ctx, loopCtx context.Context, healthErrCh <-chan error) {
+func (l *Loop) collectHealthErrors(ctx, loopCtx context.Context,
+	done chan<- struct{}, healthErrCh <-chan error,
+) {
+	defer close(done)
 	var previousHealthErr error
 	for {
 		select {