Files
dnswatcher/internal/resolver/resolver_test.go
sneak 9cb2c2b7e0
All checks were successful
check / check (push) Successful in 1m18s
test: make live DNS tests robust instead of gated (closes #93)
The resolver's live-DNS tests failed nondeterministically, a different
subset each run. Three structural causes, all test-side:

- Burst fan-out. Every test in the package is parallel and the build
  hosts have many cores, so all ~35 iterative resolutions started at
  the same instant and, because queryServers walks rootServerList() in
  fixed order, hit the same root server within milliseconds. Root
  servers rate-limit that.
- No retry anywhere. One dropped UDP packet in a delegation chain
  failed a test outright.
- Unanimity assertions. TestQueryAllNameservers_AllReturnOK and
  _NXDomainFromAllNS required every one of a domain's nameservers to
  answer, with no tolerance for one being slow.

New internal/resolver/livedns_test.go addresses each: a package-wide
gate bounds how many live resolutions are in flight at once, every
live operation gets three attempts with exponential backoff and its
own deadline, and multi-nameserver assertions now need a strict
majority rather than unanimity. The retry predicate is deliberately
transport-level -- "did a nameserver answer at all" -- never the
assertion under test, so a resolver that answers incorrectly still
fails on the first attempt. A nameserver that stays silent is
tolerated; one that answers wrongly is not.

livedns_harness_test.go tests that machinery directly: quorum
arithmetic, status counting, the gate's concurrency bound, per-attempt
deadlines, and recovery from a transient failure. It touches no DNS.

Nothing is mocked, faked, stubbed, recorded, skipped or build-tagged,
and production resolver behaviour is unchanged.

Test caps move to the new org-wide values ruled at prompts issue 41:
60s hard cap, 20s target, 90s -timeout backstop. REPO_POLICIES.md is
re-vendored byte-identical from sneak/prompts rather than hand-edited,
which also picks up the golangci-lint paragraph this copy had drifted
behind on. TESTING.md's stale 30-second target follows to 60.

#93
2026-08-10 13:09:17 +00:00

568 lines
12 KiB
Go

package resolver_test
import (
"context"
"log/slog"
"net"
"os"
"sort"
"strings"
"testing"
"time"
"github.com/miekg/dns"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
"sneak.berlin/go/dnswatcher/internal/resolver"
)
// ----------------------------------------------------------------
// Test helpers
// ----------------------------------------------------------------
func newTestResolver(t *testing.T) *resolver.Resolver {
t.Helper()
log := slog.New(slog.NewTextHandler(
os.Stderr,
&slog.HandlerOptions{Level: slog.LevelDebug},
))
return resolver.NewFromLogger(log)
}
// findOneNSForDomain picks one authoritative nameserver to aim a
// test at. Live-DNS retry, concurrency and quorum handling live in
// livedns_test.go.
func findOneNSForDomain(
t *testing.T,
r *resolver.Resolver,
domain string,
) string {
t.Helper()
return liveFindAuthoritative(t, r, domain)[0]
}
// ----------------------------------------------------------------
// FindAuthoritativeNameservers tests
// ----------------------------------------------------------------
func TestFindAuthoritativeNameservers_ValidDomain(
t *testing.T,
) {
t.Parallel()
r := newTestResolver(t)
nameservers := liveFindAuthoritative(t, r, "google.com")
hasGoogleNS := false
for _, ns := range nameservers {
if strings.Contains(ns, "google") {
hasGoogleNS = true
break
}
}
assert.True(t, hasGoogleNS,
"expected google nameservers, got: %v", nameservers,
)
}
func TestFindAuthoritativeNameservers_Subdomain(
t *testing.T,
) {
t.Parallel()
r := newTestResolver(t)
nameservers := liveFindAuthoritative(t, r, "www.google.com")
assert.NotEmpty(t, nameservers)
}
func TestFindAuthoritativeNameservers_ReturnsSorted(
t *testing.T,
) {
t.Parallel()
r := newTestResolver(t)
nameservers := liveFindAuthoritative(t, r, "google.com")
assert.True(
t,
sort.StringsAreSorted(nameservers),
"nameservers should be sorted, got: %v", nameservers,
)
}
func TestFindAuthoritativeNameservers_Deterministic(
t *testing.T,
) {
t.Parallel()
r := newTestResolver(t)
first := liveFindAuthoritative(t, r, "google.com")
second := liveFindAuthoritative(t, r, "google.com")
assert.Equal(t, first, second)
}
func TestFindAuthoritativeNameservers_TrailingDot(
t *testing.T,
) {
t.Parallel()
r := newTestResolver(t)
ns1 := liveFindAuthoritative(t, r, "google.com")
ns2 := liveFindAuthoritative(t, r, "google.com.")
assert.Equal(t, ns1, ns2)
}
func TestFindAuthoritativeNameservers_CloudflareDomain(
t *testing.T,
) {
t.Parallel()
r := newTestResolver(t)
nameservers := liveFindAuthoritative(t, r, "cloudflare.com")
for _, ns := range nameservers {
assert.True(t, strings.HasSuffix(ns, "."),
"NS should be FQDN with trailing dot: %s", ns,
)
}
}
// ----------------------------------------------------------------
// QueryNameserver tests
// ----------------------------------------------------------------
func TestQueryNameserver_BasicA(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ns := findOneNSForDomain(t, r, "google.com")
resp := liveQueryNameserver(t, r, ns, "www.google.com")
require.NotNil(t, resp)
assert.Equal(t, resolver.StatusOK, resp.Status)
assert.Equal(t, ns, resp.Nameserver)
hasRecords := len(resp.Records["A"]) > 0 ||
len(resp.Records["CNAME"]) > 0
assert.True(t, hasRecords,
"expected A or CNAME records for www.google.com",
)
}
func TestQueryNameserver_AAAA(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ns := findOneNSForDomain(t, r, "cloudflare.com")
resp := liveQueryNameserver(t, r, ns, "cloudflare.com")
aaaaRecords := resp.Records["AAAA"]
require.NotEmpty(t, aaaaRecords,
"cloudflare.com should have AAAA records",
)
for _, ip := range aaaaRecords {
parsed := net.ParseIP(ip)
require.NotNil(t, parsed,
"should be valid IP: %s", ip,
)
}
}
func TestQueryNameserver_MX(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ns := findOneNSForDomain(t, r, "google.com")
resp := liveQueryNameserver(t, r, ns, "google.com")
mxRecords := resp.Records["MX"]
require.NotEmpty(t, mxRecords,
"google.com should have MX records",
)
}
func TestQueryNameserver_TXT(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ns := findOneNSForDomain(t, r, "google.com")
resp := liveQueryNameserver(t, r, ns, "google.com")
txtRecords := resp.Records["TXT"]
require.NotEmpty(t, txtRecords,
"google.com should have TXT records",
)
hasSPF := false
for _, txt := range txtRecords {
if strings.Contains(txt, "v=spf1") {
hasSPF = true
break
}
}
assert.True(t, hasSPF,
"google.com should have SPF TXT record",
)
}
func TestQueryNameserver_NXDomain(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ns := findOneNSForDomain(t, r, "google.com")
resp := liveQueryNameserver(
t, r, ns, "this-surely-does-not-exist-xyz.google.com",
)
assert.Equal(t, resolver.StatusNXDomain, resp.Status)
}
func TestQueryNameserver_RecordsSorted(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ns := findOneNSForDomain(t, r, "google.com")
resp := liveQueryNameserver(t, r, ns, "google.com")
for recordType, values := range resp.Records {
assert.True(
t,
sort.StringsAreSorted(values),
"%s records should be sorted", recordType,
)
}
}
func TestQueryNameserver_ResponseIncludesNameserver(
t *testing.T,
) {
t.Parallel()
r := newTestResolver(t)
ns := findOneNSForDomain(t, r, "cloudflare.com")
resp := liveQueryNameserver(t, r, ns, "cloudflare.com")
assert.Equal(t, ns, resp.Nameserver)
}
func TestQueryNameserver_EmptyRecordsOnNXDomain(
t *testing.T,
) {
t.Parallel()
r := newTestResolver(t)
ns := findOneNSForDomain(t, r, "google.com")
resp := liveQueryNameserver(
t, r, ns, "this-surely-does-not-exist-xyz.google.com",
)
totalRecords := 0
for _, values := range resp.Records {
totalRecords += len(values)
}
assert.Zero(t, totalRecords)
}
func TestQueryNameserver_TrailingDotHandling(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ns := findOneNSForDomain(t, r, "google.com")
resp1 := liveQueryNameserver(t, r, ns, "google.com")
resp2 := liveQueryNameserver(t, r, ns, "google.com.")
assert.Equal(t, resp1.Status, resp2.Status)
}
// ----------------------------------------------------------------
// QueryAllNameservers tests
// ----------------------------------------------------------------
func TestQueryAllNameservers_ReturnsAllNS(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
results := liveQueryAllNameservers(t, r, "google.com")
assert.GreaterOrEqual(t, len(results), minNameservers)
for ns, resp := range results {
assert.Equal(t, ns, resp.Nameserver)
}
}
func TestQueryAllNameservers_AllReturnOK(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
results := liveQueryAllNameservers(t, r, "google.com")
// A quorum, not unanimity: one authoritative server being
// slow or rate-limiting us is a property of the live
// internet, not a resolver defect.
assert.GreaterOrEqual(
t,
countStatus(results, resolver.StatusOK),
liveQuorum(len(results)),
"a quorum of nameservers should answer OK: %s",
describeStatuses(results),
)
// Any nameserver claiming google.com does not exist is a
// real failure and is never tolerated.
assert.Zero(
t,
countStatus(results, resolver.StatusNXDomain),
"no nameserver should report NXDOMAIN: %s",
describeStatuses(results),
)
}
func TestQueryAllNameservers_NXDomainFromAllNS(
t *testing.T,
) {
t.Parallel()
r := newTestResolver(t)
results := liveQueryAllNameservers(
t, r, "this-surely-does-not-exist-xyz.google.com",
)
assert.GreaterOrEqual(
t,
countStatus(results, resolver.StatusNXDomain),
liveQuorum(len(results)),
"a quorum of nameservers should report NXDOMAIN: %s",
describeStatuses(results),
)
// Silence is tolerated; a positive answer for a name that
// does not exist is not.
assert.Zero(
t,
countStatus(results, resolver.StatusOK),
"no nameserver should answer OK: %s",
describeStatuses(results),
)
}
// ----------------------------------------------------------------
// LookupNS tests
// ----------------------------------------------------------------
func TestLookupNS_ValidDomain(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
nameservers := liveLookupNS(t, r, "google.com")
for _, ns := range nameservers {
assert.True(t, strings.HasSuffix(ns, "."),
"NS should have trailing dot: %s", ns,
)
}
}
func TestLookupNS_Sorted(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
nameservers := liveLookupNS(t, r, "google.com")
assert.True(t, sort.StringsAreSorted(nameservers))
}
func TestLookupNS_MatchesFindAuthoritative(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
fromLookup := liveLookupNS(t, r, "google.com")
fromFind := liveFindAuthoritative(t, r, "google.com")
assert.Equal(t, fromFind, fromLookup)
}
// ----------------------------------------------------------------
// ResolveIPAddresses tests
// ----------------------------------------------------------------
func TestResolveIPAddresses_ReturnsIPs(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ips := liveResolveIPs(t, r, "google.com")
for _, ip := range ips {
parsed := net.ParseIP(ip)
assert.NotNil(t, parsed,
"should be valid IP: %s", ip,
)
}
}
func TestResolveIPAddresses_Deduplicated(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ips := liveResolveIPs(t, r, "google.com")
seen := make(map[string]bool)
for _, ip := range ips {
assert.False(t, seen[ip], "duplicate IP: %s", ip)
seen[ip] = true
}
}
func TestResolveIPAddresses_Sorted(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ips := liveResolveIPs(t, r, "google.com")
assert.True(t, sort.StringsAreSorted(ips))
}
func TestResolveIPAddresses_NXDomainReturnsEmpty(
t *testing.T,
) {
t.Parallel()
r := newTestResolver(t)
ips := liveResolveIPsAllowingEmpty(
t, r, "this-surely-does-not-exist-xyz.google.com",
)
assert.Empty(t, ips)
}
func TestResolveIPAddresses_CloudflareDomain(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ips := liveResolveIPs(t, r, "cloudflare.com")
assert.NotEmpty(t, ips)
}
// ----------------------------------------------------------------
// Context cancellation tests
// ----------------------------------------------------------------
func TestFindAuthoritativeNameservers_ContextCanceled(
t *testing.T,
) {
t.Parallel()
r := newTestResolver(t)
ctx, cancel := context.WithCancel(context.Background())
cancel()
_, err := r.FindAuthoritativeNameservers(ctx, "google.com")
assert.Error(t, err)
}
func TestQueryNameserver_ContextCanceled(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ctx, cancel := context.WithCancel(context.Background())
cancel()
_, err := r.QueryNameserver(
ctx, "ns1.google.com.", "google.com",
)
assert.Error(t, err)
}
func TestQueryAllNameservers_ContextCanceled(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ctx, cancel := context.WithCancel(context.Background())
cancel()
_, err := r.QueryAllNameservers(ctx, "google.com")
assert.Error(t, err)
}
// ----------------------------------------------------------------
// Timeout tests
// ----------------------------------------------------------------
func TestQueryNameserverIP_Timeout(t *testing.T) {
t.Parallel()
log := slog.New(slog.NewTextHandler(
os.Stderr,
&slog.HandlerOptions{Level: slog.LevelDebug},
))
r := resolver.NewFromLoggerWithClient(
log, &timeoutClient{},
)
ctx, cancel := context.WithTimeout(
context.Background(), 10*time.Second,
)
t.Cleanup(cancel)
// Query any IP — the client always returns a timeout error.
resp, err := r.QueryNameserverIP(
ctx, "unreachable.test.", "192.0.2.1",
"example.com",
)
require.NoError(t, err)
assert.Equal(t, resolver.StatusTimeout, resp.Status)
assert.NotEmpty(t, resp.Error)
}
// timeoutClient simulates DNS timeout errors for testing.
type timeoutClient struct{}
func (c *timeoutClient) ExchangeContext(
_ context.Context,
_ *dns.Msg,
_ string,
) (*dns.Msg, time.Duration, error) {
return nil, 0, &net.OpError{
Op: "read",
Net: "udp",
Err: &timeoutError{},
}
}
type timeoutError struct{}
func (e *timeoutError) Error() string { return "i/o timeout" }
func (e *timeoutError) Timeout() bool { return true }
func (e *timeoutError) Temporary() bool { return true }
func TestResolveIPAddresses_ContextCanceled(t *testing.T) {
t.Parallel()
r := newTestResolver(t)
ctx, cancel := context.WithCancel(context.Background())
cancel()
_, err := r.ResolveIPAddresses(ctx, "google.com")
assert.Error(t, err)
}