Keep taking messages while a peer's session goes down (closes #216)

The table's PeerDown visits every prefix under the write lock, which
takes over a second on a full table. The prefix handler called it
inline, so its queue filled and the streamer dropped updates.

PeerDown now holds the lock for 10,000 prefixes at a time, then lets
the lock go for as long again. The handler runs PeerDown on a separate
goroutine, one peer at a time. The handler holds back the down peer's
later messages until that PeerDown returns. Then the handler applies
them in order, one at a time, and takes new messages in between. It
holds back at most 20,000 messages. Stop waits for all of this before
its last flush.

Model: opus-5-5

README.md

@@ -444,7 +444,9 @@ the routing table, and everything else from the database.
         - An announcement gives the peer a route to the prefix.
         - A withdrawal removes the peer's route. So does an announcement whose
           AS path is empty or holds AS 0 anywhere.
-        - A peer whose BGP session goes down leaves every prefix.
+        - A peer whose BGP session goes down leaves every prefix. The handler
+          holds back the peer's later updates until the peer has left every
+          prefix, then applies them in order.
 
         Every second it writes the changed prefix and origin AS pairs to the
         `routes` table. A write takes at most 20,000 pairs, and the rest follow
@@ -580,8 +582,8 @@ Budget:
 - Go heap: a 2.5 GiB soft limit (`GOMEMLIMIT=2560MiB`, set in the image).
     - The routing table takes about 700 bytes per prefix and origin AS. A full
       table has about 1.4 million of these pairs, so it takes about 1 GB.
-    - The rest of the heap takes up to about 0.5 GiB. Most of it is the
+    - The rest of the heap takes up to about 0.5 GiB. Most of it is what the
-      handlers' batches and queues, described below.
+      handlers hold, described below.
     - The remaining 1 GiB is room for the garbage collector. With much less, it
       would run almost all the time.
 - SQLite: 656 MiB of page cache.
@@ -648,6 +650,12 @@ What happens at each limit:
     as above. With the batches being written, the three handlers' batches take
     at most about 0.1 GiB of the Go heap.
 
+- Held-back messages: the prefix handler holds back a peer's messages after the
+  peer's session goes down. It applies them once the peer has left every prefix.
+  It holds at most 20,000 such messages (`maxHeldBackMessages`), about 20 MiB.
+  At that number it takes nothing more off its queue until it has applied one of
+  them.
+
 With `DEBUG=routewatch` the daemon logs a `System stats` line every 60 seconds
 with the goroutine count and Go memory figures.
 

TODO.md

@@ -35,6 +35,23 @@ tested.
 
 # Completed Steps
 
+- 2026-11-02: a peer's session going down no longer stops the prefix handler.
+  The handler runs the table's `PeerDown` on a separate goroutine, one peer at a
+  time, and goes on taking messages. `PeerDown` holds the table's lock for
+  10,000 prefixes at a time, then lets the lock go for as long again. The
+  handler holds back the down peer's later messages until that peer's `PeerDown`
+  returns. Then the handler applies them in order, one at a time, and takes new
+  messages in between. It holds back at most 20,000 messages. At that number,
+  `HandleMessage` waits until one of them is applied. When `Stop` runs past its
+  deadline, it logs how many held-back messages were not applied (closes #216)
+- 2026-11-02: RouteWatch pauses an RIR's or NIC.br's WHOIS server for an hour
+  after the server refuses a query. A refused or reset connection is a refusal.
+  So is an answer with an `%ERROR:201:` or `%ERROR:202:` line. Lookups at the
+  other servers go on. An AS whose RIR is paused is looked up after the pause.
+  IANA is never paused, because a pause of IANA would stop every lookup. A
+  refusal by IANA is an ordinary failure. NIC.br is also paused for an hour
+  after it says RouteWatch has passed NIC.br's query limit. Before, the fetcher
+  asked a server again within seconds of a refusal (closes #218)
 - 2026-11-02: the AS handler never waits for a database write. Its batch holds
   one entry per AS, with the earliest and the latest time of its updates. So the
   batch no longer grows with the feed's rate. It used to hold one entry for each

internal/routetable/overview.go

@@ -10,10 +10,6 @@ import (
 	"git.eeqj.de/sneak/routewatch/internal/ristypes"
 )
 
-// walkBatch is how many prefixes the walk over the whole table reads in one
-// hold of the read lock. The updates waiting for the lock get it in between.
-const walkBatch = 10000
-
 // Overview holds figures about the whole table. Working them out takes a walk
 // over every prefix, so a caller asks for them now and then and keeps them.
 type Overview struct {

internal/routetable/routetable.go

@@ -27,6 +27,11 @@ const (
 	// kept routes in order: each of its nodes holds up to 63 times.
 	keptTimesDegree = 32
 
+	// walkBatch is how many prefixes a walk over the whole table visits in
+	// one hold of the lock. Overview and PeerDown each walk the whole table.
+	// Callers waiting for the lock get it between batches.
+	walkBatch = 10000
+
 	ipv4Bits = 32
 	ipv6Bits = 128
 
@@ -201,8 +206,13 @@ func (t *Table) Withdraw(prefix netip.Prefix, peer string) {
 }
 
 // PeerDown applies a peer's session going down: the peer leaves every peer
-// set, and every route of it is dropped. When the peer carries any prefix, it
+// set, and every route of the peer is dropped. When the peer carries any
-// visits every prefix in the table.
+// prefix, PeerDown visits every prefix in the table, in batches of walkBatch.
+// After each batch, it lets the table's lock go for as long as it held the
+// lock for the batch. During a walk over many batches, reads and updates can
+// then take the lock about half of the time. An announcement by the peer
+// applied during the walk may be dropped too. So the caller applies the peer's
+// later updates only once PeerDown has returned.
 func (t *Table) PeerDown(peer string) {
 	t.mu.Lock()
 	defer t.mu.Unlock()
@@ -212,6 +222,9 @@ func (t *Table) PeerDown(peer string) {
 		return
 	}
 
+	visited := 0
+	batchStart := time.Now()
+
 	for _, f := range []*family{&t.ipv4, &t.ipv6} {
 		for _, prefixes := range f.byLength {
 			for prefix, origins := range prefixes {
@@ -219,6 +232,19 @@ func (t *Table) PeerDown(peer string) {
 				if o != nil {
 					t.leave(prefix, o, number)
 				}
+
+				// Go allows the map to change before the next step of
+				// the range over it.
+				visited++
+				if visited%walkBatch == 0 {
+					held := time.Since(batchStart)
+
+					t.mu.Unlock()
+					time.Sleep(held)
+					t.mu.Lock()
+
+					batchStart = time.Now()
+				}
 			}
 		}
 	}

internal/routetable/routetable_internal_test.go

@@ -0,0 +1,69 @@
+package routetable
+
+import (
+	"net/netip"
+	"runtime"
+	"testing"
+	"time"
+)
+
+// TestPeerDownLetsReadsIn reads the table's counts again and again while a peer
+// that carries more than two batches of prefixes goes down. PeerDown lets the
+// lock go after every walkBatch prefixes, so a read must get in while the peer
+// still carries some of its prefixes but no longer all.
+func TestPeerDownLetsReadsIn(t *testing.T) {
+	t.Parallel()
+
+	const (
+		asn  = 1000
+		peer = "192.0.2.1"
+		// prefixCount is more prefixes than two of walkBatch.
+		prefixCount = 2*walkBatch + 1
+	)
+
+	routeTime := time.Date(2026, 1, 2, 3, 4, 5, 0, time.UTC)
+	table := New(Params{})
+	addr := netip.MustParseAddr("1.0.0.0")
+
+	for range prefixCount {
+		prefix := netip.PrefixFrom(addr, addr.BitLen())
+		table.Announce(prefix, asn, Route{Peer: peer, Time: routeTime})
+
+		addr = addr.Next()
+	}
+
+	// Hold a read lock until PeerDown waits for the lock, so that the first
+	// read below starts once PeerDown holds it. TryRLock fails once a writer
+	// waits.
+	table.mu.RLock()
+
+	peerDownDone := make(chan struct{})
+
+	go func() {
+		defer close(peerDownDone)
+
+		table.PeerDown(peer)
+	}()
+
+	for table.mu.TryRLock() {
+		table.mu.RUnlock()
+		runtime.Gosched()
+	}
+
+	table.mu.RUnlock()
+
+	for {
+		routes := table.Counts().IPv4Routes
+		if routes > 0 && routes < prefixCount {
+			break
+		}
+
+		select {
+		case <-peerDownDone:
+			t.Fatal("no read got in while PeerDown ran")
+		default:
+		}
+	}
+
+	<-peerDownDone
+}

internal/routewatch/asnfetcher.go

@@ -287,6 +287,13 @@ func (f *ASNFetcher) fetchNextStale(ctx context.Context) {
 // so the AS stays stale. An AS IANA refers to no RIR is stored with no
 // WHOIS data, so it is not looked up again until it is stale; that counts
 // as neither a success nor a failure.
+//
+// The WHOIS client pauses an RIR's or NIC.br's server for whois.PauseTime
+// after the server refuses a query. Lookups at the other servers go on. IANA
+// is never paused, so a refusal by IANA is an ordinary failure. A lookup
+// whose RIR is paused fails with whois.ErrPaused. Such a lookup leaves the
+// AS stale, and counts as neither a success nor a failure. The lookups that
+// QueueImmediate asks for also run through fetchAndUpdate.
 func (f *ASNFetcher) fetchAndUpdate(ctx context.Context, asn int) bool {
 	ctx, cancel := context.WithTimeout(ctx, time.Minute)
 	defer cancel()
@@ -307,6 +314,13 @@ func (f *ASNFetcher) fetchAndUpdate(ctx context.Context, asn int) bool {
 		return false
 	}
 
+	if errors.Is(err, whois.ErrPaused) {
+		f.logger.Info("Skipped WHOIS lookup, the RIR's server is paused",
+			"asn", asn, "error", err)
+
+		return false
+	}
+
 	if err != nil {
 		f.logger.Error("WHOIS lookup failed", "asn", asn, "error", err)
 		f.recordFailure()

internal/routewatch/asnfetcher_internal_test.go

@@ -13,6 +13,7 @@ import (
 
 	"git.eeqj.de/sneak/routewatch/internal/database"
 	"git.eeqj.de/sneak/routewatch/internal/logger"
+	"git.eeqj.de/sneak/routewatch/internal/whois"
 )
 
 // The WHOIS servers a lookup asks: IANA first, then the RIR IANA refers it
@@ -131,6 +132,109 @@ func TestFetchAndUpdateKeepsTheRowWhenTheLookupFails(t *testing.T) {
 	}
 }
 
+// TestFetchAndUpdateWaitsOutAPause looks up an AS at RIPE while RIPE refuses
+// the query. The test sets the WHOIS client's clock. The next lookup comes a
+// second before whois.PauseTime has passed. It sends RIPE nothing and leaves
+// the AS never looked up. It counts as neither a success nor a failure. A
+// lookup at whois.PauseTime asks RIPE again.
+func TestFetchAndUpdateWaitsOutAPause(t *testing.T) {
+	t.Parallel()
+
+	f, db := newFetcherWithWHOISAnswers(t, map[string]string{
+		ianaWHOISServer: ianaRefersToRIPE,
+		ripeWHOISServer: ripeAccessDenied,
+	})
+
+	ripeAsks := 0
+	dial := f.whoisClient.DialContext
+	f.whoisClient.DialContext = func(
+		ctx context.Context, network, address string,
+	) (net.Conn, error) {
+		if address == ripeWHOISServer {
+			ripeAsks++
+		}
+
+		return dial(ctx, network, address)
+	}
+
+	start := time.Date(2026, 11, 2, 0, 0, 0, 0, time.UTC)
+	now := start
+	f.whoisClient.Now = func() time.Time { return now }
+
+	f.fetchAndUpdate(t.Context(), testASNA)
+
+	now = start.Add(whois.PauseTime - time.Second)
+
+	f.fetchAndUpdate(t.Context(), testASNA)
+
+	if ripeAsks != 1 {
+		t.Errorf("RIPE was asked %d times during its pause, want 1", ripeAsks)
+	}
+
+	stats, err := db.GetWHOISStats(t.Context(), whoisStaleThreshold)
+	if err != nil {
+		t.Fatalf("GetWHOISStats: %v", err)
+	}
+
+	if stats.NeverFetched != 1 {
+		t.Errorf("never looked up = %d, want 1", stats.NeverFetched)
+	}
+
+	got := f.GetStats()
+	if got.ConsecutiveFails != 1 || got.ErrorsLastHour != 1 ||
+		got.SuccessesLastHour != 0 {
+		t.Errorf("%d failures in a row, %d errors, %d successes; want 1, 1, 0",
+			got.ConsecutiveFails, got.ErrorsLastHour, got.SuccessesLastHour)
+	}
+
+	now = start.Add(whois.PauseTime)
+
+	f.fetchAndUpdate(t.Context(), testASNA)
+
+	if ripeAsks != 2 {
+		t.Errorf("RIPE was asked %d times after its pause, want 2", ripeAsks)
+	}
+}
+
+// TestFetchAndUpdateKeepsAskingIANA looks up an AS while IANA's server
+// refuses the connection. The test sets the WHOIS client's clock. The next
+// lookup comes a second before whois.PauseTime has passed. IANA is never
+// paused, so that lookup asks IANA again. Each refusal counts as a failure.
+func TestFetchAndUpdateKeepsAskingIANA(t *testing.T) {
+	t.Parallel()
+
+	// With no answers, every server refuses the connection.
+	f, _ := newFetcherWithWHOISAnswers(t, map[string]string{})
+
+	ianaAsks := 0
+	dial := f.whoisClient.DialContext
+	f.whoisClient.DialContext = func(
+		ctx context.Context, network, address string,
+	) (net.Conn, error) {
+		if address == ianaWHOISServer {
+			ianaAsks++
+		}
+
+		return dial(ctx, network, address)
+	}
+
+	start := time.Date(2026, 11, 2, 0, 0, 0, 0, time.UTC)
+	now := start
+	f.whoisClient.Now = func() time.Time { return now }
+
+	f.fetchAndUpdate(t.Context(), testASNA)
+
+	now = start.Add(whois.PauseTime - time.Second)
+
+	f.fetchAndUpdate(t.Context(), testASNA)
+
+	got := f.GetStats()
+	if ianaAsks != 2 || got.ConsecutiveFails != 2 {
+		t.Errorf("IANA was asked %d times, with %d failures in a row; "+
+			"want 2 and 2", ianaAsks, got.ConsecutiveFails)
+	}
+}
+
 // TestRecordStepsTheInterval has the fetcher fail from its base interval of
 // 15 seconds, then succeed. Each failure makes the interval 1.5 times as
 // long, up to one minute, and each success halves it, down to one second.

internal/routewatch/batchwrite_internal_test.go

@@ -5,6 +5,7 @@ import (
 	"context"
 	"encoding/json"
 	"fmt"
+	"io"
 	"log/slog"
 	"slices"
 	"sync"
@@ -177,13 +178,14 @@ type stopLog struct {
 	Msg          string `json:"msg"`
 	Updates      int    `json:"updates"`
 	TableChanges int    `json:"table_changes"`
+	HeldMessages int    `json:"held_messages"`
 	Paths        int    `json:"paths"`
 	BatchEntries int    `json:"batch_entries"`
 }
 
 // newErrorLogger returns a logger that writes only errors, as JSON lines, to
 // logs.
-func newErrorLogger(logs *bytes.Buffer) *logger.Logger {
+func newErrorLogger(logs io.Writer) *logger.Logger {
 	return &logger.Logger{Logger: slog.New(slog.NewJSONHandler(logs,
 		&slog.HandlerOptions{Level: slog.LevelError}))}
 }

internal/routewatch/prefixhandler.go

@@ -44,6 +44,12 @@ const (
 	// most one pair. So the million pairs that a full-table peer's session
 	// going down changes are written in about a minute.
 	maxTableChangesPerWrite = 20000
+
+	// maxHeldBackMessages is the most messages the handler holds back from
+	// peers whose PeerDown has not returned. A message takes about 1 KiB, so
+	// that many take about 20 MiB (README "Memory"). At this number
+	// HandleMessage waits until one of them is applied.
+	maxHeldBackMessages = 20000
 )
 
 // PrefixHandler keeps the routing table in memory current from the feed. It
@@ -52,6 +58,15 @@ const (
 // does an announcement whose AS path is empty or holds AS 0 anywhere. A peer
 // whose BGP session goes down leaves every prefix.
 //
+// The table's PeerDown applies a session going down. It visits every prefix,
+// which takes seconds on a full table. So the handler runs PeerDown on
+// a separate goroutine, one peer at a time, and goes on taking messages
+// meanwhile. PeerDown lets the table's lock go about half of the time, so the
+// handler applies other peers' messages while PeerDown runs. The handler holds
+// back the down peer's later messages until PeerDown returns. Then the handler
+// applies them in order, one at a time, and takes new messages in between. It
+// holds back at most maxHeldBackMessages messages.
+//
 // Each flush writes the table's changes to the database. It also stores each
 // prefix the updates name with the time it was first seen.
 type PrefixHandler struct {
@@ -67,7 +82,7 @@ type PrefixHandler struct {
 	// give a smaller one.
 	tableChangesPerWrite int
 
-	// Batching. mu guards batch, writing and writingTableChanges only: the
+	// Batching. mu guards batch, writing and writingTableChanges: the
 	// flush goroutine takes them under mu and writes them after releasing
 	// mu, so HandleMessage waits for a database write only while batch is at
 	// batchMaxSize.
@@ -87,6 +102,21 @@ type PrefixHandler struct {
 	// Stop does not wait for a write in progress to hand it over.
 	stopCh chan context.Context
 	done   chan struct{} // closed when the flush goroutine returns
+
+	// Peers whose session went down. peerDownMu lets one PeerDown run at a
+	// time, so the handler's own updates wait for the table's lock behind
+	// at most one. heldBack, under mu, holds each such peer's later
+	// messages, in order, until its PeerDown has returned and they are
+	// applied. A peer with no such messages yet has an empty entry.
+	// heldCount, under mu, is how many messages heldBack holds.
+	// heldTaken is signalled each time a held-back message is taken out to
+	// be applied; its L is &mu. peerDowns counts the peerDown goroutines
+	// that have not returned.
+	peerDownMu sync.Mutex
+	heldBack   map[string][]*ristypes.RISMessage
+	heldCount  int
+	heldTaken  sync.Cond
+	peerDowns  sync.WaitGroup
 }
 
 // seenPrefix is what the batch keeps of an update: the prefix it names and its
@@ -150,8 +180,10 @@ func newPrefixHandler(
 		batchFullCh:          make(chan struct{}, 1),
 		stopCh:               make(chan context.Context, 1),
 		done:                 make(chan struct{}),
+		heldBack:             make(map[string][]*ristypes.RISMessage),
 	}
 	h.batchTaken.L = &h.mu
+	h.heldTaken.L = &h.mu
 
 	// Start the flush timer goroutine
 	go flushLoop(batchTimeout, h.batchFullCh, h.stopCh, h.done, h.flushBatch)
@@ -179,11 +211,91 @@ func (h *PrefixHandler) QueueCapacity() int {
 	return prefixHandlerQueueSize
 }
 
-// HandleMessage processes a message to track prefix information
+// HandleMessage processes a message to track prefix information. A message
+// from a peer whose PeerDown has not returned is held back until it has.
 func (h *PrefixHandler) HandleMessage(msg *ristypes.RISMessage) {
+	h.mu.Lock()
+	defer h.mu.Unlock()
+
+	// While maxHeldBackMessages messages are held back, wait until one of them
+	// is taken out to be applied. Meanwhile the handler takes nothing off its
+	// queue, as at batchMaxSize.
+	for h.heldCount >= maxHeldBackMessages {
+		h.heldTaken.Wait()
+	}
+
+	held, goingDown := h.heldBack[msg.Peer]
+	if goingDown {
+		h.heldBack[msg.Peer] = append(held, msg)
+		h.heldCount++
+
+		return
+	}
+
+	h.applyMessage(msg)
+}
+
+// Stop waits for every PeerDown the handler started and for the messages held
+// back meanwhile to be applied. It gives up this wait once ctx ends. Then Stop
+// stops the flush goroutine and waits for it until ctx ends. The flush
+// goroutine makes one last flush before it returns. That flush writes at most
+// tableChangesPerWrite of the routing table's changes, and the prefixes of the
+// updates the handler holds. Once ctx has ended the flush goroutine skips that
+// last flush, and a write still running is lost when the process exits. A
+// failed write of the table's changes is not retried either. Stop logs how
+// many updates and table changes are not written. It also logs how many
+// held-back messages are not applied. The caller calls Stop once its last
+// HandleMessage has returned.
+func (h *PrefixHandler) Stop(ctx context.Context) {
+	peerDownsDone := make(chan struct{})
+
+	go func() {
+		h.peerDowns.Wait()
+		close(peerDownsDone)
+	}()
+
+	select {
+	case <-peerDownsDone:
+	case <-ctx.Done():
+	}
+
+	h.stopCh <- ctx
+
+	select {
+	case <-h.done:
+	case <-ctx.Done():
+	}
+
+	h.mu.Lock()
+	unwritten := h.writing + len(h.batch)
+	tableChanges := h.writingTableChanges + h.table.ChangeCount()
+	heldMessages := h.heldCount
+	h.mu.Unlock()
+
+	if unwritten == 0 && tableChanges == 0 && heldMessages == 0 {
+		return
+	}
+
+	if ctx.Err() != nil {
+		h.logger.Error("Stop deadline passed before the prefix handler wrote "+
+			"its updates", "updates", unwritten, "table_changes", tableChanges,
+			"held_messages", heldMessages)
+
+		return
+	}
+
+	h.logger.Error("The prefix handler stopped before it wrote everything",
+		"updates", unwritten, "table_changes", tableChanges,
+		"held_messages", heldMessages)
+}
+
+// applyMessage applies msg. The caller holds mu. A peer's session going down
+// starts the peer's PeerDown. An update is applied to the routing table, and
+// its prefixes join the batch.
+func (h *PrefixHandler) applyMessage(msg *ristypes.RISMessage) {
 	if msg.Type == ristypes.MessageTypeState {
 		if msg.State == ristypes.StateDown {
-			h.table.PeerDown(msg.Peer)
+			h.startPeerDown(msg.Peer)
 		}
 
 		return
@@ -198,9 +310,6 @@ func (h *PrefixHandler) HandleMessage(msg *ristypes.RISMessage) {
 	// the path after the rest of it.
 	path := slices.Concat(msg.Path.Sequence, msg.Path.Set)
 
-	h.mu.Lock()
-	defer h.mu.Unlock()
-
 	// While the batch is at its maximum size, wait until the flush goroutine
 	// takes it. Meanwhile the handler takes nothing off its queue, which
 	// fills, and the streamer drops messages for it.
@@ -261,39 +370,58 @@ func (h *PrefixHandler) HandleMessage(msg *ristypes.RISMessage) {
 	}
 }
 
-// Stop stops the flush goroutine, which makes one last flush before it returns,
+// startPeerDown starts peerDown for peer on a new goroutine. The caller holds
-// and waits for it until ctx ends. That flush writes at most
+// mu. The peer's messages are held back until peerDown has applied them.
-// tableChangesPerWrite of the routing table's changes, and the prefixes of
+// Messages held back for an earlier PeerDown of the peer, and not yet applied,
-// the updates the handler holds. Once ctx has ended the flush goroutine skips
+// stay held back.
-// that last flush, and a write still running is lost when the process exits. A
+func (h *PrefixHandler) startPeerDown(peer string) {
-// failed write of the table's changes is not retried either. Stop logs how many
+	_, goingDown := h.heldBack[peer]
-// updates and how many of the table's changes are not written.
+	if !goingDown {
-func (h *PrefixHandler) Stop(ctx context.Context) {
+		h.heldBack[peer] = []*ristypes.RISMessage{}
-	h.stopCh <- ctx
-
-	select {
-	case <-h.done:
-	case <-ctx.Done():
 	}
 
+	h.peerDowns.Add(1)
+
+	go h.peerDown(peer)
+}
+
+// peerDown runs the table's PeerDown for peer, once no other PeerDown is
+// running. Then peerDown applies the messages held back from the peer, in
+// order. It takes mu for one message at a time, so HandleMessage takes new
+// messages in between. The peer's new messages join the end of the held-back
+// ones. A held-back message that says the peer's session went down again
+// starts a new peerDown, which applies the rest.
+func (h *PrefixHandler) peerDown(peer string) {
+	defer h.peerDowns.Done()
+
+	h.peerDownMu.Lock()
+	h.table.PeerDown(peer)
+	h.peerDownMu.Unlock()
+
+	for {
 		h.mu.Lock()
-	unwritten := h.writing + len(h.batch)
+
-	tableChanges := h.writingTableChanges + h.table.ChangeCount()
+		held := h.heldBack[peer]
+		if len(held) == 0 {
+			delete(h.heldBack, peer)
 			h.mu.Unlock()
 
-	if unwritten == 0 && tableChanges == 0 {
 			return
 		}
 
-	if ctx.Err() != nil {
+		msg := held[0]
-		h.logger.Error("Stop deadline passed before the prefix handler wrote "+
+		held[0] = nil // so that the message is freed once applied
-			"its updates", "updates", unwritten, "table_changes", tableChanges)
+		h.heldBack[peer] = held[1:]
+		h.heldCount--
+		h.heldTaken.Broadcast()
+		h.applyMessage(msg)
+		h.mu.Unlock()
 
+		if msg.Type == ristypes.MessageTypeState &&
+			msg.State == ristypes.StateDown {
 			return
 		}
-
+	}
-	h.logger.Error("The prefix handler stopped before it wrote everything",
-		"updates", unwritten, "table_changes", tableChanges)
 }
 
 // takeBatch returns the pending updates and leaves an empty batch in their

internal/routewatch/prefixhandler_internal_test.go

@@ -237,6 +237,83 @@ func TestPrefixHandlerStopLogsNothingWhenItWroteEverything(t *testing.T) {
 	}
 }
 
+// TestPrefixHandlerStopLogsTheMessagesHeldBack stops a prefix handler with a
+// deadline that has passed, while a peer's PeerDown is held up and one message
+// of the peer is held back. Stop must log that message as not applied.
+func TestPrefixHandlerStopLogsTheMessagesHeldBack(t *testing.T) {
+	t.Parallel()
+
+	var logs bytes.Buffer
+
+	h := newPrefixHandler(newMockStore(), newErrorLogger(&logs),
+		routetable.New(routetable.Params{}), longerThanAnyTest,
+		maxTableChangesPerWrite)
+
+	h.peerDownMu.Lock()
+	defer h.peerDownMu.Unlock()
+
+	h.HandleMessage(peerAnnouncement(testPeerIP, testV4Prefix))
+	h.HandleMessage(sessionDown())
+	h.HandleMessage(peerAnnouncement(testPeerIP, testV6Prefix))
+
+	ctx, cancel := context.WithCancel(t.Context())
+	cancel()
+	h.Stop(ctx)
+
+	logged := lastLogged(t, &logs)
+	if logged.HeldMessages != 1 {
+		t.Errorf("Stop logged %d held messages not applied, want 1",
+			logged.HeldMessages)
+	}
+}
+
+// TestPrefixHandlerStopWaitsForTheMessagesHeldBack stops a prefix handler
+// while a peer's PeerDown is held up and an announcement of the peer is held
+// back. Stop must wait. Once the PeerDown is let go and Stop returns, the
+// announcement must be in the database.
+func TestPrefixHandlerStopWaitsForTheMessagesHeldBack(t *testing.T) {
+	t.Parallel()
+
+	db := newTestDatabase(t)
+	h := newPrefixHandler(db, logger.New(), routetable.New(routetable.Params{}),
+		longerThanAnyTest, maxTableChangesPerWrite)
+
+	h.peerDownMu.Lock()
+	h.HandleMessage(sessionDown())
+	h.HandleMessage(peerAnnouncement(testPeerIP, testV4Prefix))
+
+	stopped := make(chan struct{})
+
+	go func() {
+		h.Stop(t.Context())
+		close(stopped)
+	}()
+
+	select {
+	case <-stopped:
+		t.Error("Stop returned while a PeerDown was held up")
+	case <-time.After(stillWaitingAfter):
+	}
+
+	h.peerDownMu.Unlock()
+
+	select {
+	case <-stopped:
+	case <-time.After(stepTimeout):
+		t.Fatalf("Stop did not return within %s of the PeerDown's release",
+			stepTimeout)
+	}
+
+	read, err := db.ReadRouteTable(t.Context())
+	if err != nil {
+		t.Fatalf("ReadRouteTable: %v", err)
+	}
+
+	checkTableOrigins(t, read, map[string]map[int][]string{
+		testV4Prefix: {testASNB: {testPeerIP}},
+	})
+}
+
 // longerThanAnyTest is a timer interval that no test lasts. A test gives it to
 // newPrefixHandler, newASHandler, newPeerHandler or newPeeringHandler when the
 // handler must write only when its batch is full, when the test says, or on
@@ -253,6 +330,7 @@ func prefixHandlerWithoutTimer(store database.Store) *PrefixHandler {
 		table:                routetable.New(routetable.Params{}),
 		batchMaxSize:         prefixBatchMaxSize,
 		tableChangesPerWrite: maxTableChangesPerWrite,
+		heldBack:             make(map[string][]*ristypes.RISMessage),
 	}
 }
 

internal/routewatch/prefixhandler_table_internal_test.go

@@ -7,8 +7,10 @@ import (
 	"log/slog"
 	"net/netip"
 	"reflect"
+	"runtime"
 	"slices"
 	"strings"
+	"sync"
 	"testing"
 	"time"
 
@@ -101,6 +103,228 @@ func TestPrefixHandlerAppliesUpdatesToTheRoutingTable(t *testing.T) {
 	})
 }
 
+// TestPrefixHandlerTakesMessagesWhileAPeerGoesDown holds up the PeerDown of a
+// peer whose session went down, as another peer's long PeerDown would. The
+// handler must go on taking messages. It applies another peer's announcement
+// at once. It holds back these messages of the down peer, in this order:
+//   - an announcement;
+//   - its session going down again;
+//   - a second announcement.
+//
+// Once the PeerDowns have returned, the down peer must carry only the prefix
+// of the second announcement.
+func TestPrefixHandlerTakesMessagesWhileAPeerGoesDown(t *testing.T) {
+	t.Parallel()
+
+	h := prefixHandlerWithoutTimer(newMockStore())
+	h.HandleMessage(peerAnnouncement(testPeerIP, testV4Prefix))
+
+	h.peerDownMu.Lock()
+
+	handled := returnsInTime(func() {
+		for _, msg := range []*ristypes.RISMessage{
+			sessionDown(),
+			peerAnnouncement(secondPeerIP, otherV4Prefix),
+			peerAnnouncement(testPeerIP, testV6Prefix),
+			sessionDown(),
+			peerAnnouncement(testPeerIP, setV4Prefix),
+		} {
+			h.HandleMessage(msg)
+		}
+	})
+	if !handled {
+		t.Fatalf("the handler did not take the messages after the session "+
+			"went down within %s", stepTimeout)
+	}
+
+	checkTableOrigins(t, h.table, map[string]map[int][]string{
+		testV4Prefix:  {testASNB: {testPeerIP}},
+		otherV4Prefix: {testASNB: {secondPeerIP}},
+		testV6Prefix:  {},
+		setV4Prefix:   {},
+	})
+
+	h.peerDownMu.Unlock()
+
+	if !returnsInTime(h.peerDowns.Wait) {
+		t.Fatalf("the PeerDowns did not return within %s", stepTimeout)
+	}
+
+	checkTableOrigins(t, h.table, map[string]map[int][]string{
+		testV4Prefix:  {},
+		otherV4Prefix: {testASNB: {secondPeerIP}},
+		testV6Prefix:  {},
+		setV4Prefix:   {testASNB: {testPeerIP}},
+	})
+}
+
+// TestPrefixHandlerAppliesMessagesWhilePeerDownRuns gives a peer 50,000
+// prefixes and hands the prefix handler that peer's session going down. The
+// table's PeerDown then visits the prefixes in five batches. Once PeerDown has
+// begun, the handler is handed 1,000 announcements of another peer. The
+// handler must apply them all before PeerDown ends, while the down peer still
+// carries some of its prefixes.
+func TestPrefixHandlerAppliesMessagesWhilePeerDownRuns(t *testing.T) {
+	t.Parallel()
+
+	const (
+		// downPeerPrefixes is five times the 10,000 prefixes PeerDown visits
+		// in one hold of the table's lock.
+		downPeerPrefixes = 50000
+		// otherPeerMessages is far more messages than the handler applies
+		// while PeerDown runs, if PeerDown takes the lock back at once after
+		// each batch. The handler then applies only a few for each batch.
+		otherPeerMessages = 1000
+		// firstOtherAddr is the address of the other peer's first prefix.
+		firstOtherAddr = "2001:db8:2::"
+	)
+
+	h := prefixHandlerWithoutTimer(newMockStore())
+	announceManyPrefixes(h.table, downPeerPrefixes)
+
+	h.HandleMessage(sessionDown())
+
+	// Wait until PeerDown has begun: the down peer has left some of its
+	// prefixes. The down peer's prefixes are IPv4 and the other peer's are
+	// IPv6, so the counts tell them apart.
+	for h.table.Counts().IPv4Routes == downPeerPrefixes {
+		runtime.Gosched()
+	}
+
+	// Each announcement names the /128 of the next address.
+	addr := netip.MustParseAddr(firstOtherAddr)
+	for range otherPeerMessages {
+		h.HandleMessage(peerAnnouncement(secondPeerIP,
+			netip.PrefixFrom(addr, addr.BitLen()).String()))
+
+		addr = addr.Next()
+	}
+
+	counts := h.table.Counts()
+	if counts.IPv6Routes != otherPeerMessages {
+		t.Errorf("the other peer carries %d prefixes, want %d",
+			counts.IPv6Routes, otherPeerMessages)
+	}
+
+	if counts.IPv4Routes == 0 {
+		t.Errorf("PeerDown ended before the handler applied the other peer's "+
+			"%d announcements", otherPeerMessages)
+	}
+
+	if !returnsInTime(h.peerDowns.Wait) {
+		t.Fatalf("the PeerDown did not return within %s", stepTimeout)
+	}
+}
+
+// TestPrefixHandlerTakesMessagesWhileItAppliesHeldBackOnes holds back many
+// messages of a peer whose session went down. Each names a prefix that does
+// not parse, so applying it logs an error. Each log write takes a
+// millisecond. While the handler applies the held-back messages, it must take
+// another peer's announcement.
+func TestPrefixHandlerTakesMessagesWhileItAppliesHeldBackOnes(t *testing.T) {
+	t.Parallel()
+
+	const (
+		// heldMessages take the handler at least a second to apply.
+		heldMessages = 1000
+		badPrefix    = "not a prefix"
+	)
+
+	logs := &slowWriter{firstWrite: make(chan struct{})}
+	h := newPrefixHandler(newMockStore(), newErrorLogger(logs),
+		routetable.New(routetable.Params{}), longerThanAnyTest,
+		maxTableChangesPerWrite)
+
+	defer stopInTime(t, h)
+
+	h.peerDownMu.Lock()
+	h.HandleMessage(sessionDown())
+
+	for range heldMessages {
+		h.HandleMessage(peerAnnouncement(testPeerIP, badPrefix))
+	}
+
+	h.peerDownMu.Unlock()
+
+	select {
+	case <-logs.firstWrite:
+	case <-time.After(stepTimeout):
+		t.Fatal("the handler did not start to apply the held-back messages")
+	}
+
+	if !returnsInTime(func() {
+		h.HandleMessage(peerAnnouncement(secondPeerIP, otherV4Prefix))
+	}) {
+		t.Fatalf("HandleMessage did not return within %s", stepTimeout)
+	}
+
+	h.mu.Lock()
+	left := len(h.heldBack[testPeerIP])
+	h.mu.Unlock()
+
+	if left == 0 {
+		t.Error("the handler took another peer's announcement only once it " +
+			"had applied every held-back message")
+	}
+}
+
+// TestPrefixHandlerWaitsAtTheMaximumHeldBack holds up the PeerDown of a peer
+// whose session went down. It hands the handler maxHeldBackMessages messages
+// of that peer. HandleMessage must then wait, even for another peer's
+// message, until the PeerDown returns and a held-back message is applied.
+func TestPrefixHandlerWaitsAtTheMaximumHeldBack(t *testing.T) {
+	t.Parallel()
+
+	h := newPrefixHandler(newMockStore(), logger.New(),
+		routetable.New(routetable.Params{}), longerThanAnyTest,
+		maxTableChangesPerWrite)
+	defer stopInTime(t, h)
+
+	h.peerDownMu.Lock()
+	h.HandleMessage(sessionDown())
+
+	held := peerAnnouncement(testPeerIP, testV6Prefix)
+	for range maxHeldBackMessages {
+		h.HandleMessage(held)
+	}
+
+	returned := make(chan struct{})
+
+	go func() {
+		h.HandleMessage(peerAnnouncement(secondPeerIP, otherV4Prefix))
+		close(returned)
+	}()
+
+	select {
+	case <-returned:
+		t.Error("HandleMessage returned while maxHeldBackMessages messages " +
+			"were held back")
+	case <-time.After(stillWaitingAfter):
+	}
+
+	h.peerDownMu.Unlock()
+
+	select {
+	case <-returned:
+	case <-time.After(stepTimeout):
+		t.Fatal("HandleMessage did not return after the PeerDown returned")
+	}
+}
+
+// slowWriter takes a millisecond over each write and drops what it is given.
+// It closes firstWrite at the first write.
+type slowWriter struct {
+	firstWrite chan struct{}
+	once       sync.Once
+}
+
+func (w *slowWriter) Write(p []byte) (int, error) {
+	w.once.Do(func() { close(w.firstWrite) })
+	time.Sleep(time.Millisecond)
+
+	return len(p), nil
+}
+
 // TestPrefixHandlerAppliesAPathWithoutAnOriginToTheRoutingTable hands the
 // prefix handler announcements whose AS path gives no origin. One whose path
 // is empty or holds AS 0 removes the peer's route, as a withdrawal does; one
@@ -199,8 +423,9 @@ func TestPrefixHandlerKeepsThePathOfAPathEndingInASet(t *testing.T) {
 	}
 }
 
-// checkTableCases hands the messages of each case to a new prefix handler and
+// checkTableCases hands the messages of each case to a new prefix handler. It
-// checks its routing table right after, with no flush.
+// checks the handler's routing table once the handler's PeerDowns have
+// returned, with no flush.
 func checkTableCases(t *testing.T, tests []tableCase) {
 	t.Helper()
 
@@ -213,17 +438,28 @@ func checkTableCases(t *testing.T, tests []tableCase) {
 				h.HandleMessage(msg)
 			}
 
-			for prefix, want := range tt.want {
+			h.peerDowns.Wait()
-				got := tableOrigins(h.table, prefix)
+			checkTableOrigins(t, h.table, tt.want)
-				if !reflect.DeepEqual(got, want) {
-					t.Errorf("peers of %s by origin AS = %v, want %v",
-						prefix, got, want)
-				}
-			}
 		})
 	}
 }
 
+// checkTableOrigins checks, for each prefix in want, the peers that carry it
+// in table with each origin AS.
+func checkTableOrigins(
+	t *testing.T, table *routetable.Table, want map[string]map[int][]string,
+) {
+	t.Helper()
+
+	for prefix, wantOrigins := range want {
+		got := tableOrigins(table, prefix)
+		if !reflect.DeepEqual(got, wantOrigins) {
+			t.Errorf("peers of %s by origin AS = %v, want %v",
+				prefix, got, wantOrigins)
+		}
+	}
+}
+
 // tableWritesStore is a real database that hands each write of the routing
 // table's changes to written once it is made.
 type tableWritesStore struct {
@@ -498,6 +734,7 @@ func TestARestartReadsTheRoutingTableBack(t *testing.T) {
 		h.HandleMessage(msg)
 	}
 
+	h.peerDowns.Wait()
 	h.flushBatch()
 
 	var logs bytes.Buffer
@@ -536,6 +773,7 @@ func TestARestartReadsTheRoutingTableBack(t *testing.T) {
 		restarted.HandleMessage(msg)
 	}
 
+	h.peerDowns.Wait()
 	restarted.Stop(t.Context())
 
 	read, err := db.ReadRouteTable(t.Context())

internal/whois/whois.go

@@ -12,6 +12,7 @@ import (
 	"regexp"
 	"strings"
 	"sync"
+	"syscall"
 	"time"
 	"unicode/utf8"
 
@@ -70,12 +71,12 @@ const nicbrHeader = "% Copyright (c) Nic.br"
 // queries to stop the data being taken in bulk.
 const nicbrLimitReached = "Query rate limit exceeded"
 
-// NICbrSkipTime is how long the client stops asking NIC.br after NIC.br
+// PauseTime is how long the client sends an RIR's or NIC.br's server no
-// answers that the client is over its query limit. NIC.br publishes
+// query after that server refuses a query. queryRegistry says what a
-// neither the limit nor how long it lasts. An hour outlasts a limit
+// refusal is. RIPE refuses an address for good when the address keeps
-// counted per minute or per hour. The ASes looked up meanwhile get
+// passing RIPE's limit (see noPersonalFlag). An hour outlasts a limit
-// NIC.br's fields when they are next refreshed.
+// counted per minute or per hour.
-const NICbrSkipTime = time.Hour
+const PauseTime = time.Hour
 
 // RIR identifiers.
 const (
@@ -125,6 +126,10 @@ var (
 	errAnswerTooLarge = errors.New("WHOIS answer reached the size cap")
 )
 
+// ErrPaused is returned when a lookup needs a server that refused a query
+// less than PauseTime ago. The client sends that server nothing meanwhile.
+var ErrPaused = errors.New("WHOIS server paused after refusing a query")
+
 // ASNInfo contains parsed WHOIS information for an ASN.
 type ASNInfo struct {
 	ASN         int
@@ -156,12 +161,12 @@ type Client struct {
 
 	logger *slog.Logger
 
-	// mu guards nicbrSkipUntil.
+	// mu guards pausedUntil.
 	mu sync.Mutex
 
-	// nicbrSkipUntil is when the client next asks NIC.br, after NIC.br
+	// pausedUntil maps a server, given as host:port, to the time the client
-	// answered that the client is over its query limit.
+	// may ask it again.
-	nicbrSkipUntil time.Time
+	pausedUntil map[string]time.Time
 }
 
 // NewClient creates a new WHOIS client that logs to logger.
@@ -172,23 +177,29 @@ func NewClient(logger *slog.Logger) *Client {
 		DialContext: dialer.DialContext,
 		Now:         time.Now,
 		logger:      logger,
+		pausedUntil: make(map[string]time.Time),
 	}
 }
 
 // LookupASN asks IANA which RIR holds the given ASN, asks that RIR, and
 // returns what the RIR's answer says. When LACNIC's answer is NIC.br's, it
 // returns what NIC.br's own server says instead, unless that query fails or
-// is skipped, as askNICbr says.
+// NIC.br is paused, as askNICbr says.
 //
-// It returns ErrNoReferral when IANA refers the AS to no RIR, and another
+// It returns ErrNoReferral when IANA refers the AS to no RIR. Its error
-// error when the query to IANA or to the RIR fails, when the RIR refuses
+// wraps ErrPaused when the RIR is paused. It returns another error in each
-// it, when IANA's answer has neither a referral nor the AS's block, when
+// of these cases:
-// IANA refers the AS to a server none of the five RIRs run, or when the
+//
-// answer carries none of ASNInfo's fields.
+//   - the query to IANA or to the RIR fails
+//   - the RIR refuses the query
+//   - IANA's answer has neither a referral nor the AS's block
+//   - IANA refers the AS to a server none of the five RIRs run
+//   - the answer carries none of ASNInfo's fields
 func (c *Client) LookupASN(ctx context.Context, asn int) (*ASNInfo, error) {
 	// Query IANA first to find the authoritative RIR
 	query := fmt.Sprintf("AS%d", asn)
 
+	// IANA is never paused, because a pause of IANA would stop every lookup.
 	ianaResp, err := c.query(ctx, whoisServerIANA, query)
 	if err != nil {
 		return nil, fmt.Errorf("IANA query failed: %w", err)
@@ -211,7 +222,7 @@ func (c *Client) LookupASN(ctx context.Context, asn int) (*ASNInfo, error) {
 	}
 
 	// Query the authoritative RIR
-	rirResp, err := c.queryRIR(ctx, whoisServer, rirQuery(rir, query))
+	rirResp, err := c.queryRegistry(ctx, whoisServer, rirQuery(rir, query))
 	if err != nil {
 		return nil, fmt.Errorf("RIR query failed: %w", err)
 	}
@@ -241,22 +252,19 @@ func (c *Client) LookupASN(ctx context.Context, asn int) (*ASNInfo, error) {
 // askNICbr returns lacnicInfo, what LACNIC's answer says, and logs a
 // warning when the query fails, when NIC.br sends the reduced answer for a
 // client over its limit, or when NIC.br's answer carries none of ASNInfo's
-// fields. After the reduced answer, it returns lacnicInfo without asking
+// fields. The reduced answer pauses NIC.br, as a refusal does. While NIC.br
-// NIC.br until NICbrSkipTime has passed.
+// is paused, askNICbr returns lacnicInfo without a warning. The ASes looked
+// up meanwhile get NIC.br's fields when they are next refreshed.
 func (c *Client) askNICbr(
 	ctx context.Context, asQuery string, lacnicInfo *ASNInfo,
 ) *ASNInfo {
 	asn := lacnicInfo.ASN
 
-	c.mu.Lock()
+	answer, err := c.queryRegistry(ctx, whoisServerNICbr, asQuery)
-	skipping := c.Now().Before(c.nicbrSkipUntil)
+	if errors.Is(err, ErrPaused) {
-	c.mu.Unlock()
-
-	if skipping {
 		return lacnicInfo
 	}
 
-	answer, err := c.query(ctx, whoisServerNICbr, asQuery)
 	if err != nil {
 		c.logger.Warn("NIC.br query failed, keeping LACNIC's fields",
 			"asn", asn, "error", err)
@@ -265,13 +273,11 @@ func (c *Client) askNICbr(
 	}
 
 	if strings.Contains(answer, nicbrLimitReached) {
-		c.mu.Lock()
+		c.pause(whoisServerNICbr)
-		c.nicbrSkipUntil = c.Now().Add(NICbrSkipTime)
-		c.mu.Unlock()
 
 		c.logger.Warn("NIC.br answered that the client is over its query "+
-			"limit, keeping LACNIC's fields and skipping NIC.br",
+			"limit, keeping LACNIC's fields and pausing NIC.br",
-			"asn", asn, "skip_for", NICbrSkipTime)
+			"asn", asn, "pause_for", PauseTime)
 
 		return lacnicInfo
 	}
@@ -309,25 +315,65 @@ func rirQuery(rir, asQuery string) string {
 	}
 }
 
-// queryRIR sends query to an RIR's server and returns the answer. An answer
+// queryRegistry sends query to the server of an RIR or of NIC.br, and
-// holding an error line with which the server refuses the query is an
+// returns the answer. The client pauses the server for PauseTime when the
-// error, and the error holds that line.
+// server refuses the query. Each of these is a refusal:
-func (c *Client) queryRIR(
+//
+//   - a refused connection
+//   - a reset connection
+//   - an %ERROR:201: or %ERROR:202: line in the answer
+//
+// A refusal is returned as an error. While the server is paused,
+// queryRegistry sends it nothing and returns an error that wraps ErrPaused.
+func (c *Client) queryRegistry(
 	ctx context.Context, server, query string,
 ) (string, error) {
+	c.mu.Lock()
+	pausedUntil := c.pausedUntil[server]
+	c.mu.Unlock()
+
+	if c.Now().Before(pausedUntil) {
+		return "", fmt.Errorf("%w: %s until %s", ErrPaused, server,
+			pausedUntil.UTC().Format(time.RFC3339))
+	}
+
 	answer, err := c.query(ctx, server, query)
+	if err == nil {
+		err = checkRefusal(answer)
+	}
+
+	if errors.Is(err, errRefused) || errors.Is(err, syscall.ECONNREFUSED) ||
+		errors.Is(err, syscall.ECONNRESET) {
+		c.pause(server)
+	}
+
 	if err != nil {
 		return "", err
 	}
 
+	return answer, nil
+}
+
+// checkRefusal returns an error holding the first line of answer that
+// starts with accessDeniedError or accessLimitError. It returns nil when
+// answer has no such line.
+func checkRefusal(answer string) error {
 	for line := range strings.SplitSeq(answer, "\n") {
 		if strings.HasPrefix(line, accessDeniedError) ||
 			strings.HasPrefix(line, accessLimitError) {
-			return "", fmt.Errorf("%w: %s", errRefused, line)
+			return fmt.Errorf("%w: %s", errRefused, line)
 		}
 	}
 
-	return answer, nil
+	return nil
+}
+
+// pause stops queryRegistry from querying server until PauseTime has passed.
+func (c *Client) pause(server string) {
+	c.mu.Lock()
+	defer c.mu.Unlock()
+
+	c.pausedUntil[server] = c.Now().Add(PauseTime)
 }
 
 // query performs a raw WHOIS query to the specified server and returns the

internal/whois/whois_test.go

@@ -718,9 +718,10 @@ func TestLookupASNKeepsLACNICsAnswerWhenNICbrFails(t *testing.T) {
 
 // TestLookupASNSkipsNICbrAfterItsLimit looks up an AS that NIC.br
 // registers three times with one client, while NIC.br answers that the
-// client is over its query limit. The first lookup asks NIC.br. The
+// client is over its query limit. The test sets the client's clock. The
-// second, a second before NICbrSkipTime has passed, does not. The third,
+// first lookup asks NIC.br. The next comes a second before PauseTime has
-// once NICbrSkipTime has passed, asks again. Each keeps LACNIC's fields.
+// passed, and does not ask NIC.br. The last comes at PauseTime and asks
+// NIC.br again. Each keeps LACNIC's fields.
 func TestLookupASNSkipsNICbrAfterItsLimit(t *testing.T) {
 	t.Parallel()
 
@@ -740,8 +741,8 @@ func TestLookupASNSkipsNICbrAfterItsLimit(t *testing.T) {
 		wantAsks int
 	}{
 		{after: 0, wantAsks: 1},
-		{after: whois.NICbrSkipTime - time.Second, wantAsks: 1},
+		{after: whois.PauseTime - time.Second, wantAsks: 1},
-		{after: whois.NICbrSkipTime, wantAsks: 2},
+		{after: whois.PauseTime, wantAsks: 2},
 	} {
 		now = start.Add(step.after)
 
@@ -785,6 +786,142 @@ func TestLookupASNReportsARefusal(t *testing.T) {
 	}
 }
 
+// resetServer starts a TCP server on the loopback address and returns the
+// server's address. The server reads one query on each connection, then
+// resets the connection.
+func resetServer(t *testing.T) string {
+	t.Helper()
+
+	var listenConfig net.ListenConfig
+
+	listener, err := listenConfig.Listen(t.Context(), "tcp", "127.0.0.1:0")
+	if err != nil {
+		t.Fatalf("listen: %v", err)
+	}
+
+	t.Cleanup(func() { _ = listener.Close() })
+
+	go func() {
+		for {
+			conn, err := listener.Accept()
+			if err != nil {
+				return
+			}
+
+			_, _ = bufio.NewReader(conn).ReadString('\n')
+
+			// Closing a connection whose linger time is zero resets it.
+			if tcpConn, ok := conn.(*net.TCPConn); ok {
+				_ = tcpConn.SetLinger(0)
+			}
+
+			_ = conn.Close()
+		}
+	}()
+
+	return listener.Addr().String()
+}
+
+// newRIPEClient returns a client whose fake servers send answers. The
+// client adds one to *ripeAsks each time it connects to RIPE's server. When
+// reset is true, RIPE's server resets the connection instead of answering.
+func newRIPEClient(
+	t *testing.T, answers map[string]string, reset bool, ripeAsks *int,
+) *whois.Client {
+	t.Helper()
+
+	dial := fakeServers(t, answers)
+	resetAddress := resetServer(t)
+
+	client := whois.NewClient(slog.New(slog.DiscardHandler))
+	client.DialContext = func(
+		ctx context.Context, network, address string,
+	) (net.Conn, error) {
+		if address == ripeServer {
+			*ripeAsks++
+		}
+
+		if address == ripeServer && reset {
+			var dialer net.Dialer
+
+			return dialer.DialContext(ctx, network, resetAddress)
+		}
+
+		return dial(ctx, network, address)
+	}
+
+	return client
+}
+
+// TestLookupASNPausesAServerThatRefuses looks up an AS at RIPE while RIPE
+// refuses the query, in each of the ways a server refuses. The test sets
+// the client's clock. A lookup a second before PauseTime has passed fails
+// with ErrPaused and asks RIPE nothing. A lookup of an AS at APNIC succeeds
+// during the pause. A lookup at PauseTime asks RIPE again.
+func TestLookupASNPausesAServerThatRefuses(t *testing.T) {
+	t.Parallel()
+
+	for name, test := range map[string]struct {
+		// ripeAnswer is RIPE's answer. With none, RIPE's server refuses
+		// the connection, or resets it when reset is true.
+		ripeAnswer string
+		reset      bool
+	}{
+		"access denied":      {ripeAnswer: ripeRefusal},
+		"limit reached":      {ripeAnswer: ripeLimitReached + "\n"},
+		"refused connection": {},
+		"reset connection":   {reset: true},
+	} {
+		t.Run(name, func(t *testing.T) {
+			t.Parallel()
+
+			answers := map[string]string{
+				ianaQuery:      ianaReferral,
+				apnicIANAQuery: "refer:        whois.apnic.net\n",
+				apnicQuery:     apnicRecord,
+			}
+			if test.ripeAnswer != "" {
+				answers[ripeQuery] = test.ripeAnswer
+			}
+
+			ripeAsks := 0
+			client := newRIPEClient(t, answers, test.reset, &ripeAsks)
+
+			start := time.Date(2026, 11, 2, 0, 0, 0, 0, time.UTC)
+			now := start
+			client.Now = func() time.Time { return now }
+
+			_, err := client.LookupASN(t.Context(), testASN)
+			if err == nil || errors.Is(err, whois.ErrPaused) {
+				t.Fatalf("LookupASN while RIPE refuses: error %v, want a "+
+					"refusal", err)
+			}
+
+			now = start.Add(whois.PauseTime - time.Second)
+
+			_, err = client.LookupASN(t.Context(), testASN)
+			if !errors.Is(err, whois.ErrPaused) || ripeAsks != 1 {
+				t.Errorf("LookupASN during the pause: error %v after %d "+
+					"queries to RIPE, want ErrPaused after 1", err, ripeAsks)
+			}
+
+			_, err = client.LookupASN(t.Context(), apnicASN)
+			if err != nil {
+				t.Errorf("LookupASN of an AS at APNIC during RIPE's "+
+					"pause: %v", err)
+			}
+
+			now = start.Add(whois.PauseTime)
+
+			_, _ = client.LookupASN(t.Context(), testASN)
+			if ripeAsks != 2 {
+				t.Errorf("RIPE was asked %d times after the pause, want 2",
+					ripeAsks)
+			}
+		})
+	}
+}
+
 // TestLookupASNFailsWhenTheRIRGivesNoAnswer looks up an AS IANA refers to
 // RIPE when RIPE's server refuses the connection, and when it answers only
 // with an error. Neither is a lookup that found the AS.