fix: retry on DNS timeout, distinguish authoritative negatives (closes #35)

- Add StatusTimeout constant for timeout responses
- querySingleType now retries on timeout and SERVFAIL (3 attempts,
  exponential backoff starting at 100ms)
- NXDOMAIN and NOERROR+empty are treated as authoritative negatives
  with no retry
- classifyResponse sets structured error messages for timeout and
  SERVFAIL cases
- Refactored into smaller functions to satisfy cyclomatic complexity
  limits

TESTING.md

@@ -0,0 +1,34 @@
+# Testing Policy
+
+## DNS Resolution Tests
+
+All resolver tests **MUST** use live queries against real DNS servers.
+No mocking of the DNS client layer is permitted.
+
+### Rationale
+
+The resolver performs iterative resolution from root nameservers through
+the full delegation chain. Mocked responses cannot faithfully represent
+the variety of real-world DNS behavior (truncation, referrals, glue
+records, DNSSEC, varied response times, EDNS, etc.). Testing against
+real servers ensures the resolver works correctly in production.
+
+### Constraints
+
+- Tests hit real DNS infrastructure and require network access
+- Test duration depends on network conditions; timeout tuning keeps
+  the suite within the 30-second target
+- Query timeout is calibrated to 3× maximum antipodal RTT (~300ms)
+  plus processing margin
+- Root server fan-out is limited to reduce parallel query load
+- Flaky failures from transient network issues are acceptable and
+  should be investigated as potential resolver bugs, not papered over
+  with mocks or skip flags
+
+### What NOT to do
+
+- **Do not mock `DNSClient`** for resolver tests (the mock constructor
+  exists for unit-testing other packages that consume the resolver)
+- **Do not add `-short` flags** to skip slow tests
+- **Do not increase `-timeout`** to hide hanging queries
+- **Do not modify linter configuration** to suppress findings

internal/resolver/errors.go

@@ -24,4 +24,8 @@ var (
 	// ErrContextCanceled wraps context cancellation for the
 	// resolver's iterative queries.
 	ErrContextCanceled = errors.New("context canceled")
+
+	// ErrSERVFAIL is returned when a DNS server responds with
+	// SERVFAIL after all retries are exhausted.
+	ErrSERVFAIL = errors.New("SERVFAIL from server")
 )

internal/resolver/iterative.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"errors"
 	"fmt"
+	"math/rand"
 	"net"
 	"sort"
 	"strings"
@@ -13,7 +14,7 @@ import (
 )
 
 const (
-	queryTimeoutDuration = 1 * time.Second
+	queryTimeoutDuration = 2 * time.Second
 	maxRetries           = 2
 	maxDelegation        = 20
 	timeoutMultiplier    = 2
@@ -41,6 +42,22 @@ func rootServerList() []string {
 	}
 }
 
+const maxRootServers = 3
+
+// randomRootServers returns a shuffled subset of root servers.
+func randomRootServers() []string {
+	all := rootServerList()
+	rand.Shuffle(len(all), func(i, j int) {
+		all[i], all[j] = all[j], all[i]
+	})
+
+	if len(all) > maxRootServers {
+		return all[:maxRootServers]
+	}
+
+	return all
+}
+
 func checkCtx(ctx context.Context) error {
 	err := ctx.Err()
 	if err != nil {
@@ -291,17 +308,8 @@ func (r *Resolver) resolveNSIPs(
 	return ips
 }
 
-// publicResolvers returns well-known public recursive DNS resolvers.
-func publicResolvers() []string {
-	return []string{
-		"1.1.1.1", // Cloudflare
-		"8.8.8.8", // Google
-		"9.9.9.9", // Quad9
-	}
-}
-
-// resolveNSRecursive queries for NS records using a public
-// recursive resolver as a fallback for intercepted environments.
+// resolveNSRecursive queries for NS records using recursive
+// resolution as a fallback for intercepted environments.
 func (r *Resolver) resolveNSRecursive(
 	ctx context.Context,
 	domain string,
@@ -311,7 +319,7 @@ func (r *Resolver) resolveNSRecursive(
 	msg.SetQuestion(domain, dns.TypeNS)
 	msg.RecursionDesired = true
 
-	for _, ip := range publicResolvers() {
+	for _, ip := range randomRootServers() {
 		if checkCtx(ctx) != nil {
 			return nil, ErrContextCanceled
 		}
@@ -332,8 +340,7 @@ func (r *Resolver) resolveNSRecursive(
 	return nil, ErrNoNameservers
 }
 
-// resolveARecord resolves a hostname to IPv4 addresses using
-// public recursive resolvers.
+// resolveARecord resolves a hostname to IPv4 addresses.
 func (r *Resolver) resolveARecord(
 	ctx context.Context,
 	hostname string,
@@ -343,7 +350,7 @@ func (r *Resolver) resolveARecord(
 	msg.SetQuestion(hostname, dns.TypeA)
 	msg.RecursionDesired = true
 
-	for _, ip := range publicResolvers() {
+	for _, ip := range randomRootServers() {
 		if checkCtx(ctx) != nil {
 			return nil, ErrContextCanceled
 		}
@@ -395,7 +402,7 @@ func (r *Resolver) FindAuthoritativeNameservers(
 		candidate := strings.Join(labels[i:], ".") + "."
 
 		nsNames, err := r.followDelegation(
-			ctx, candidate, rootServerList(),
+			ctx, candidate, randomRootServers(),
 		)
 		if err == nil && len(nsNames) > 0 {
 			sort.Strings(nsNames)
@@ -452,9 +459,15 @@ func (r *Resolver) queryAllTypes(
 	return resp, nil
 }
 
+const (
+	singleTypeMaxRetries     = 3
+	singleTypeInitialBackoff = 100 * time.Millisecond
+)
+
 type queryState struct {
 	gotNXDomain bool
 	gotSERVFAIL bool
+	gotTimeout  bool
 	hasRecords  bool
 }
 
@@ -482,6 +495,21 @@ func (r *Resolver) queryEachType(
 	return state
 }
 
+// isTimeout checks whether an error represents a DNS timeout.
+func isTimeout(err error) bool {
+	if err == nil {
+		return false
+	}
+
+	var netErr net.Error
+	if errors.As(err, &netErr) && netErr.Timeout() {
+		return true
+	}
+
+	// Also catch i/o timeout strings from the dns library.
+	return strings.Contains(err.Error(), "i/o timeout")
+}
+
 func (r *Resolver) querySingleType(
 	ctx context.Context,
 	nsIP string,
@@ -490,23 +518,99 @@ func (r *Resolver) querySingleType(
 	resp *NameserverResponse,
 	state *queryState,
 ) {
-	msg, err := r.queryDNS(ctx, nsIP, hostname, qtype)
-	if err != nil {
+	msg, lastErr := r.querySingleTypeWithRetry(
+		ctx, nsIP, hostname, qtype,
+	)
+	if msg == nil {
+		r.recordRetryFailure(lastErr, state)
+
 		return
 	}
 
+	r.handleDNSResponse(msg, resp, state)
+}
+
+func (r *Resolver) querySingleTypeWithRetry(
+	ctx context.Context,
+	nsIP string,
+	hostname string,
+	qtype uint16,
+) (*dns.Msg, error) {
+	var lastErr error
+
+	backoff := singleTypeInitialBackoff
+
+	for attempt := range singleTypeMaxRetries {
+		if checkCtx(ctx) != nil {
+			return nil, ErrContextCanceled
+		}
+
+		if attempt > 0 {
+			if !waitBackoff(ctx, backoff) {
+				return nil, ErrContextCanceled
+			}
+
+			backoff *= timeoutMultiplier
+		}
+
+		msg, err := r.queryDNS(ctx, nsIP, hostname, qtype)
+		if err != nil {
+			lastErr = err
+
+			if !isTimeout(err) {
+				return nil, err
+			}
+
+			continue
+		}
+
+		if msg.Rcode == dns.RcodeServerFailure {
+			lastErr = ErrSERVFAIL
+
+			continue
+		}
+
+		return msg, nil
+	}
+
+	return nil, lastErr
+}
+
+func waitBackoff(ctx context.Context, d time.Duration) bool {
+	select {
+	case <-ctx.Done():
+		return false
+	case <-time.After(d):
+		return true
+	}
+}
+
+func (r *Resolver) recordRetryFailure(
+	lastErr error,
+	state *queryState,
+) {
+	if lastErr == nil {
+		return
+	}
+
+	if isTimeout(lastErr) {
+		state.gotTimeout = true
+	} else if errors.Is(lastErr, ErrSERVFAIL) {
+		state.gotSERVFAIL = true
+	}
+}
+
+func (r *Resolver) handleDNSResponse(
+	msg *dns.Msg,
+	resp *NameserverResponse,
+	state *queryState,
+) {
 	if msg.Rcode == dns.RcodeNameError {
 		state.gotNXDomain = true
 
 		return
 	}
 
-	if msg.Rcode == dns.RcodeServerFailure {
-		state.gotSERVFAIL = true
-
-		return
-	}
-
 	collectAnswerRecords(msg, resp, state)
 }
 
@@ -533,8 +637,12 @@ func classifyResponse(resp *NameserverResponse, state queryState) {
 	switch {
 	case state.gotNXDomain && !state.hasRecords:
 		resp.Status = StatusNXDomain
+	case state.gotTimeout && !state.hasRecords:
+		resp.Status = StatusTimeout
+		resp.Error = "all queries timed out after retries"
 	case state.gotSERVFAIL && !state.hasRecords:
 		resp.Status = StatusError
+		resp.Error = "server failure (SERVFAIL) after retries"
 	case !state.hasRecords && !state.gotNXDomain:
 		resp.Status = StatusNoData
 	}

internal/resolver/resolver.go

@@ -17,6 +17,7 @@ const (
 	StatusError    = "error"
 	StatusNXDomain = "nxdomain"
 	StatusNoData   = "nodata"
+	StatusTimeout  = "timeout"
 )
 
 // MaxCNAMEDepth is the maximum CNAME chain depth to follow.

internal/watcher/watcher.go

@@ -6,6 +6,7 @@ import (
 	"log/slog"
 	"sort"
 	"strings"
+	"sync"
 	"time"
 
 	"go.uber.org/fx"
@@ -40,15 +41,17 @@ type Params struct {
 
 // Watcher orchestrates all monitoring checks on a schedule.
 type Watcher struct {
-	log       *slog.Logger
-	config    *config.Config
-	state     *state.State
-	resolver  DNSResolver
-	portCheck PortChecker
-	tlsCheck  TLSChecker
-	notify    Notifier
-	cancel    context.CancelFunc
-	firstRun  bool
+	log              *slog.Logger
+	config           *config.Config
+	state            *state.State
+	resolver         DNSResolver
+	portCheck        PortChecker
+	tlsCheck         TLSChecker
+	notify           Notifier
+	cancel           context.CancelFunc
+	firstRun         bool
+	expiryNotifiedMu sync.Mutex
+	expiryNotified   map[string]time.Time
 }
 
 // New creates a new Watcher instance wired into the fx lifecycle.
@@ -57,14 +60,15 @@ func New(
 	params Params,
 ) (*Watcher, error) {
 	w := &Watcher{
-		log:       params.Logger.Get(),
-		config:    params.Config,
-		state:     params.State,
-		resolver:  params.Resolver,
-		portCheck: params.PortCheck,
-		tlsCheck:  params.TLSCheck,
-		notify:    params.Notify,
-		firstRun:  true,
+		log:            params.Logger.Get(),
+		config:         params.Config,
+		state:          params.State,
+		resolver:       params.Resolver,
+		portCheck:      params.PortCheck,
+		tlsCheck:       params.TLSCheck,
+		notify:         params.Notify,
+		firstRun:       true,
+		expiryNotified: make(map[string]time.Time),
 	}
 
 	lifecycle.Append(fx.Hook{
@@ -100,14 +104,15 @@ func NewForTest(
 	n Notifier,
 ) *Watcher {
 	return &Watcher{
-		log:       slog.Default(),
-		config:    cfg,
-		state:     st,
-		resolver:  res,
-		portCheck: pc,
-		tlsCheck:  tc,
-		notify:    n,
-		firstRun:  true,
+		log:            slog.Default(),
+		config:         cfg,
+		state:          st,
+		resolver:       res,
+		portCheck:      pc,
+		tlsCheck:       tc,
+		notify:         n,
+		firstRun:       true,
+		expiryNotified: make(map[string]time.Time),
 	}
 }
 
@@ -206,6 +211,28 @@ func (w *Watcher) checkDomain(
 		Nameservers: nameservers,
 		LastChecked: now,
 	})
+
+	// Also look up A/AAAA records for the apex domain so that
+	// port and TLS checks (which read HostnameState) can find
+	// the domain's IP addresses.
+	records, err := w.resolver.LookupAllRecords(ctx, domain)
+	if err != nil {
+		w.log.Error(
+			"failed to lookup records for domain",
+			"domain", domain,
+			"error", err,
+		)
+
+		return
+	}
+
+	prevHS, hasPrevHS := w.state.GetHostnameState(domain)
+	if hasPrevHS && !w.firstRun {
+		w.detectHostnameChanges(ctx, domain, prevHS, records)
+	}
+
+	newState := buildHostnameState(records, now)
+	w.state.SetHostnameState(domain, newState)
 }
 
 func (w *Watcher) detectNSChanges(
@@ -691,6 +718,22 @@ func (w *Watcher) checkTLSExpiry(
 		return
 	}
 
+	// Deduplicate expiry warnings: don't re-notify for the same
+	// hostname within the TLS check interval.
+	dedupKey := fmt.Sprintf("expiry:%s:%s", hostname, ip)
+
+	w.expiryNotifiedMu.Lock()
+
+	lastNotified, seen := w.expiryNotified[dedupKey]
+	if seen && time.Since(lastNotified) < w.config.TLSInterval {
+		w.expiryNotifiedMu.Unlock()
+
+		return
+	}
+
+	w.expiryNotified[dedupKey] = time.Now()
+	w.expiryNotifiedMu.Unlock()
+
 	msg := fmt.Sprintf(
 		"Host: %s\nIP: %s\nCN: %s\n"+
 			"Expires: %s (%.0f days)",

internal/watcher/watcher_test.go

@@ -273,6 +273,10 @@ func setupBaselineMocks(deps *testDeps) {
 		"ns1.example.com.",
 		"ns2.example.com.",
 	}
+	deps.resolver.allRecords["example.com"] = map[string]map[string][]string{
+		"ns1.example.com.": {"A": {"93.184.216.34"}},
+		"ns2.example.com.": {"A": {"93.184.216.34"}},
+	}
 	deps.resolver.allRecords["www.example.com"] = map[string]map[string][]string{
 		"ns1.example.com.": {"A": {"93.184.216.34"}},
 		"ns2.example.com.": {"A": {"93.184.216.34"}},
@@ -290,6 +294,14 @@ func setupBaselineMocks(deps *testDeps) {
 			"www.example.com",
 		},
 	}
+	deps.tlsChecker.certs["93.184.216.34:example.com"] = &tlscheck.CertificateInfo{
+		CommonName: "example.com",
+		Issuer:     "DigiCert",
+		NotAfter:   time.Now().Add(90 * 24 * time.Hour),
+		SubjectAlternativeNames: []string{
+			"example.com",
+		},
+	}
 }
 
 func assertNoNotifications(
@@ -322,14 +334,74 @@ func assertStatePopulated(
 		)
 	}
 
-	if len(snap.Hostnames) != 1 {
+	// Hostnames includes both explicit hostnames and domains
+	// (domains now also get hostname state for port/TLS checks).
+	if len(snap.Hostnames) < 1 {
 		t.Errorf(
-			"expected 1 hostname in state, got %d",
+			"expected at least 1 hostname in state, got %d",
 			len(snap.Hostnames),
 		)
 	}
 }
 
+func TestDomainPortAndTLSChecks(t *testing.T) {
+	t.Parallel()
+
+	cfg := defaultTestConfig(t)
+	cfg.Domains = []string{"example.com"}
+
+	w, deps := newTestWatcher(t, cfg)
+
+	deps.resolver.nsRecords["example.com"] = []string{
+		"ns1.example.com.",
+	}
+	deps.resolver.allRecords["example.com"] = map[string]map[string][]string{
+		"ns1.example.com.": {"A": {"93.184.216.34"}},
+	}
+	deps.portChecker.results["93.184.216.34:80"] = true
+	deps.portChecker.results["93.184.216.34:443"] = true
+	deps.tlsChecker.certs["93.184.216.34:example.com"] = &tlscheck.CertificateInfo{
+		CommonName: "example.com",
+		Issuer:     "DigiCert",
+		NotAfter:   time.Now().Add(90 * 24 * time.Hour),
+		SubjectAlternativeNames: []string{
+			"example.com",
+		},
+	}
+
+	w.RunOnce(t.Context())
+
+	snap := deps.state.GetSnapshot()
+
+	// Domain should have port state populated
+	if len(snap.Ports) == 0 {
+		t.Error("expected port state for domain, got none")
+	}
+
+	// Domain should have certificate state populated
+	if len(snap.Certificates) == 0 {
+		t.Error("expected certificate state for domain, got none")
+	}
+
+	// Verify port checker was actually called
+	deps.portChecker.mu.Lock()
+	calls := deps.portChecker.calls
+	deps.portChecker.mu.Unlock()
+
+	if calls == 0 {
+		t.Error("expected port checker to be called for domain")
+	}
+
+	// Verify TLS checker was actually called
+	deps.tlsChecker.mu.Lock()
+	tlsCalls := deps.tlsChecker.calls
+	deps.tlsChecker.mu.Unlock()
+
+	if tlsCalls == 0 {
+		t.Error("expected TLS checker to be called for domain")
+	}
+}
+
 func TestNSChangeDetection(t *testing.T) {
 	t.Parallel()
 
@@ -342,6 +414,12 @@ func TestNSChangeDetection(t *testing.T) {
 		"ns1.example.com.",
 		"ns2.example.com.",
 	}
+	deps.resolver.allRecords["example.com"] = map[string]map[string][]string{
+		"ns1.example.com.": {"A": {"1.2.3.4"}},
+		"ns2.example.com.": {"A": {"1.2.3.4"}},
+	}
+	deps.portChecker.results["1.2.3.4:80"] = false
+	deps.portChecker.results["1.2.3.4:443"] = false
 
 	ctx := t.Context()
 	w.RunOnce(ctx)
@@ -351,6 +429,10 @@ func TestNSChangeDetection(t *testing.T) {
 		"ns1.example.com.",
 		"ns3.example.com.",
 	}
+	deps.resolver.allRecords["example.com"] = map[string]map[string][]string{
+		"ns1.example.com.": {"A": {"1.2.3.4"}},
+		"ns3.example.com.": {"A": {"1.2.3.4"}},
+	}
 	deps.resolver.mu.Unlock()
 
 	w.RunOnce(ctx)
@@ -506,6 +588,61 @@ func TestTLSExpiryWarning(t *testing.T) {
 	}
 }
 
+func TestTLSExpiryWarningDedup(t *testing.T) {
+	t.Parallel()
+
+	cfg := defaultTestConfig(t)
+	cfg.Hostnames = []string{"www.example.com"}
+	cfg.TLSInterval = 24 * time.Hour
+
+	w, deps := newTestWatcher(t, cfg)
+
+	deps.resolver.allRecords["www.example.com"] = map[string]map[string][]string{
+		"ns1.example.com.": {"A": {"1.2.3.4"}},
+	}
+	deps.resolver.ipAddresses["www.example.com"] = []string{
+		"1.2.3.4",
+	}
+	deps.portChecker.results["1.2.3.4:80"] = true
+	deps.portChecker.results["1.2.3.4:443"] = true
+	deps.tlsChecker.certs["1.2.3.4:www.example.com"] = &tlscheck.CertificateInfo{
+		CommonName: "www.example.com",
+		Issuer:     "DigiCert",
+		NotAfter:   time.Now().Add(3 * 24 * time.Hour),
+		SubjectAlternativeNames: []string{
+			"www.example.com",
+		},
+	}
+
+	ctx := t.Context()
+
+	// First run = baseline, no notifications
+	w.RunOnce(ctx)
+
+	// Second run should fire one expiry warning
+	w.RunOnce(ctx)
+
+	// Third run should NOT fire another warning (dedup)
+	w.RunOnce(ctx)
+
+	notifications := deps.notifier.getNotifications()
+
+	expiryCount := 0
+
+	for _, n := range notifications {
+		if n.Title == "TLS Expiry Warning: www.example.com" {
+			expiryCount++
+		}
+	}
+
+	if expiryCount != 1 {
+		t.Errorf(
+			"expected exactly 1 expiry warning (dedup), got %d",
+			expiryCount,
+		)
+	}
+}
+
 func TestGracefulShutdown(t *testing.T) {
 	t.Parallel()
 
@@ -519,6 +656,11 @@ func TestGracefulShutdown(t *testing.T) {
 	deps.resolver.nsRecords["example.com"] = []string{
 		"ns1.example.com.",
 	}
+	deps.resolver.allRecords["example.com"] = map[string]map[string][]string{
+		"ns1.example.com.": {"A": {"1.2.3.4"}},
+	}
+	deps.portChecker.results["1.2.3.4:80"] = false
+	deps.portChecker.results["1.2.3.4:443"] = false
 
 	ctx, cancel := context.WithCancel(t.Context())