cmd/cli: skip upstream.os healthcheck when WFP loopback protect enabled

Since the check will always be failed in this case, causing unnecessary
log spamming.
This commit is contained in:
Cuong Manh Le
2026-05-05 22:15:54 +07:00
committed by Cuong Manh Le
parent 2b27c148be
commit 5dd5846cca
4 changed files with 44 additions and 4 deletions
+4
View File
@@ -1816,3 +1816,7 @@ func (p *prog) forceReloadPFMainRuleset() {
mainLog.Load().Info().Msg("DNS intercept: force reload — pf ruleset and anchor reloaded successfully") mainLog.Load().Info().Msg("DNS intercept: force reload — pf ruleset and anchor reloaded successfully")
} }
// osHealthcheckSuppressed always returns false on darwin — WFP loopback
// protect (the trigger for suppression) is Windows-only.
func (p *prog) osHealthcheckSuppressed() bool { return false }
+4
View File
@@ -37,3 +37,7 @@ func (p *prog) scheduleDelayedRechecks() {}
// pfInterceptMonitor is a no-op on unsupported platforms. // pfInterceptMonitor is a no-op on unsupported platforms.
func (p *prog) pfInterceptMonitor() {} func (p *prog) pfInterceptMonitor() {}
// osHealthcheckSuppressed always returns false on non-Windows platforms —
// WFP loopback protect (the trigger for suppression) is Windows-only.
func (p *prog) osHealthcheckSuppressed() bool { return false }
+20
View File
@@ -1184,6 +1184,26 @@ func (p *prog) activateLoopbackWFPProtect(state *wfpState) error {
return nil return nil
} }
// osHealthcheckSuppressed reports whether the upstream.os healthcheck should
// be skipped because DNS intercept mode is active and the WFP loopback protect
// has been engaged. Loopback protect is only activated when an external WFP
// block filter (e.g. OpenVPN's block-outside-dns) is interfering with DNS,
// which is the same condition that makes the OS resolver healthcheck fail
// every 2s with i/o timeout — so suppressing the check avoids the log spam
// described in issue #526.
func (p *prog) osHealthcheckSuppressed() bool {
if !dnsIntercept || p.dnsInterceptState == nil {
return false
}
state, ok := p.dnsInterceptState.(*wfpState)
if !ok || state == nil {
return false
}
state.mu.Lock()
defer state.mu.Unlock()
return state.loopbackProtectActive
}
// deactivateLoopbackWFPProtectLocked is the lock-free inner implementation. // deactivateLoopbackWFPProtectLocked is the lock-free inner implementation.
// Caller must hold state.mu. // Caller must hold state.mu.
func (p *prog) deactivateLoopbackWFPProtectLocked(state *wfpState) { func (p *prog) deactivateLoopbackWFPProtectLocked(state *wfpState) {
+16 -4
View File
@@ -744,6 +744,7 @@ func (p *prog) proxy(ctx context.Context, req *proxyRequest) *proxyResponse {
var reason RecoveryReason var reason RecoveryReason
if upstreams[0] == upstreamOS { if upstreams[0] == upstreamOS {
reason = RecoveryReasonOSFailure reason = RecoveryReasonOSFailure
} else { } else {
reason = RecoveryReasonRegularFailure reason = RecoveryReasonRegularFailure
} }
@@ -1657,6 +1658,8 @@ func interfaceIPsEqual(a, b []netip.Prefix) bool {
return true return true
} }
var errOsHealthcheckSuppressed = errors.New("upstream os health check suppressed")
// checkUpstreamOnce sends a test query to the specified upstream. // checkUpstreamOnce sends a test query to the specified upstream.
// Returns nil if the upstream responds successfully. // Returns nil if the upstream responds successfully.
func (p *prog) checkUpstreamOnce(upstream string, uc *ctrld.UpstreamConfig) error { func (p *prog) checkUpstreamOnce(upstream string, uc *ctrld.UpstreamConfig) error {
@@ -1686,11 +1689,19 @@ func (p *prog) checkUpstreamOnce(upstream string, uc *ctrld.UpstreamConfig) erro
duration := time.Since(start) duration := time.Since(start)
if err != nil { if err != nil {
// Demote upstream.os check failures to debug while WFP loopback
// protect is active: an external WFP block filter is interfering
// with plain DNS so repeated failures here are expected. Other
// upstreams keep error level so real outages stay visible.
if upstream == upstreamOS && p.osHealthcheckSuppressed() {
mainLog.Load().Debug().Err(err).Msgf("Upstream %s check failed after %v (WFP loopback protect active)", upstream, duration)
return errOsHealthcheckSuppressed
}
mainLog.Load().Error().Err(err).Msgf("Upstream %s check failed after %v", upstream, duration) mainLog.Load().Error().Err(err).Msgf("Upstream %s check failed after %v", upstream, duration)
} else { return err
mainLog.Load().Debug().Msgf("Upstream %s responded successfully in %v", upstream, duration)
} }
return err mainLog.Load().Debug().Msgf("Upstream %s responded successfully in %v", upstream, duration)
return nil
} }
// recoveryDebounceWindow is the time to wait after the last network change // recoveryDebounceWindow is the time to wait after the last network change
@@ -1909,7 +1920,8 @@ func (p *prog) waitForUpstreamRecovery(ctx context.Context, upstreams map[string
default: default:
attempts++ attempts++
// checkUpstreamOnce will reset any failure counters on success. // checkUpstreamOnce will reset any failure counters on success.
if err := p.checkUpstreamOnce(name, uc); err == nil { err := p.checkUpstreamOnce(name, uc)
if err == nil || errors.Is(err, errOsHealthcheckSuppressed) {
mainLog.Load().Debug().Msgf("Upstream %s recovered successfully", name) mainLog.Load().Debug().Msgf("Upstream %s recovered successfully", name)
select { select {
case recoveredCh <- name: case recoveredCh <- name: