Write each handler batch from its flush goroutine, outside the mutex (closes #65)
check / check (push) Canceled after 0s

The AS, peer and prefix handlers wrote their batch to the database while
holding the mutex HandleMessage needs, so while a write ran nothing was
taken off their queues and the streamer dropped updates.

Each handler now writes only from its flush goroutine: it takes the batch
under the mutex, leaves an empty one in its place, and writes after
releasing it. A full batch wakes the flush goroutine. While a write runs the
next batch may grow to ten batch sizes; at that size HandleMessage waits, so
the queue fills and the streamer drops as before. README "Memory" gives the
numbers and their cost.

Model: opus-5-5
This commit was merged in pull request #83.
This commit is contained in:
2026-10-19 09:45:24 +02:00
parent f7c88c2496
commit 43c61ee0d0
6 changed files with 584 additions and 93 deletions
+66 -27
View File
@@ -19,6 +19,12 @@ const (
// asnBatchSize is the number of ASN operations to batch together
asnBatchSize = 30000
// asnBatchMaxSize is the most ASN operations the batch holds while the
// flush goroutine writes the previous one: ten batches, about 10 MiB
// (README "Memory"). At this size HandleMessage waits until the flush
// goroutine takes the batch.
asnBatchMaxSize = 300000
// asnBatchTimeout is the maximum time to wait before flushing a batch
// DO NOT reduce this timeout - larger batches are more efficient
asnBatchTimeout = 2 * time.Second
@@ -32,12 +38,16 @@ type ASHandler struct {
db database.Store
logger *logger.Logger
// Batching
mu sync.Mutex
batch []asnOp
lastFlush time.Time
stopCh chan struct{}
wg sync.WaitGroup
// Batching. mu guards batch only: the flush goroutine takes the batch
// under mu and writes it after releasing mu, so HandleMessage waits for
// a database write only while batch is at asnBatchMaxSize.
mu sync.Mutex
batch []asnOp
batchTaken sync.Cond // signalled when the flush goroutine takes batch; its L is &mu
batchFullCh chan struct{} // wakes the flush goroutine when batch is full
lastFlush time.Time // used only by the flush goroutine
stopCh chan struct{}
wg sync.WaitGroup
}
type asnOp struct {
@@ -52,12 +62,14 @@ type asnOp struct {
// background goroutine is terminated.
func NewASHandler(db database.Store, logger *logger.Logger) *ASHandler {
h := &ASHandler{
db: db,
logger: logger,
batch: make([]asnOp, 0, asnBatchSize),
lastFlush: time.Now(),
stopCh: make(chan struct{}),
db: db,
logger: logger,
batch: make([]asnOp, 0, asnBatchSize),
batchFullCh: make(chan struct{}, 1),
lastFlush: time.Now(),
stopCh: make(chan struct{}),
}
h.batchTaken.L = &h.mu
// Start the flush timer goroutine
h.wg.Add(1)
@@ -85,8 +97,8 @@ func (h *ASHandler) QueueCapacity() int {
// HandleMessage processes a RIS Live BGP message by extracting all ASNs from
// the AS path and queuing them for batch insertion into the database. The
// origin ASN (last element in the path) and all transit ASNs are recorded
// with their associated timestamps. The batch is automatically flushed when
// it reaches the configured size threshold.
// with their associated timestamps. When the batch reaches the configured size
// threshold, the flush goroutine is woken to write it.
func (h *ASHandler) HandleMessage(msg *ristypes.RISMessage) {
// Use the pre-parsed timestamp
timestamp := msg.ParsedTimestamp
@@ -100,6 +112,13 @@ func (h *ASHandler) HandleMessage(msg *ristypes.RISMessage) {
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.
for len(h.batch) >= asnBatchMaxSize {
h.batchTaken.Wait()
}
// Queue origin ASN operation
if originASN > 0 {
h.batch = append(h.batch, asnOp{
@@ -118,13 +137,19 @@ func (h *ASHandler) HandleMessage(msg *ristypes.RISMessage) {
}
}
// Check if we need to flush
// A full batch wakes the flush goroutine. If it is writing, the wake-up
// waits in batchFullCh and the batch keeps growing until the write ends
// or the batch reaches asnBatchMaxSize.
if len(h.batch) >= asnBatchSize {
h.flushBatchLocked()
select {
case h.batchFullCh <- struct{}{}:
default: // already woken
}
}
}
// flushLoop runs in a goroutine and periodically flushes batches
// flushLoop runs in a goroutine, the only one that writes the batch: it
// flushes when the batch is full, periodically, and once more on Stop.
func (h *ASHandler) flushLoop() {
defer h.wg.Done()
ticker := time.NewTicker(asnBatchTimeout)
@@ -133,31 +158,47 @@ func (h *ASHandler) flushLoop() {
for {
select {
case <-ticker.C:
h.mu.Lock()
if time.Since(h.lastFlush) >= asnBatchTimeout {
h.flushBatchLocked()
h.flushBatch()
}
h.mu.Unlock()
case <-h.batchFullCh:
h.flushBatch()
case <-h.stopCh:
// Final flush
h.mu.Lock()
h.flushBatchLocked()
h.mu.Unlock()
h.flushBatch()
return
}
}
}
// flushBatchLocked flushes the ASN batch to the database (must be called with mutex held)
func (h *ASHandler) flushBatchLocked() {
// takeBatch returns the pending operations and leaves an empty batch in their
// place for HandleMessage to fill.
func (h *ASHandler) takeBatch() []asnOp {
h.mu.Lock()
defer h.mu.Unlock()
if len(h.batch) == 0 {
return nil
}
batch := h.batch
h.batch = make([]asnOp, 0, asnBatchSize)
h.batchTaken.Broadcast()
return batch
}
// flushBatch takes the pending batch and writes it to the database, without
// holding the mutex during the write.
func (h *ASHandler) flushBatch() {
batch := h.takeBatch()
if len(batch) == 0 {
return
}
// Process ASNs first (deduped)
asnMap := make(map[int]time.Time)
for _, op := range h.batch {
for _, op := range batch {
if existing, ok := asnMap[op.number]; !ok || op.timestamp.After(existing) {
asnMap[op.number] = op.timestamp
}
@@ -168,8 +209,6 @@ func (h *ASHandler) flushBatchLocked() {
h.logger.Error("Failed to process ASN batch", "error", err, "count", len(asnMap))
}
// Clear batch
h.batch = h.batch[:0]
h.lastFlush = time.Now()
}