Serve /api/v1/stats from a cache and index-scan the route timestamps (closes #27)
check / check (push) Failing after 0s
check / check (push) Failing after 0s
Once the database passed about 4.5 GiB every stats request ran a COUNT(*) over each table plus a MIN/MAX union scan of both route tables, took the full timeout and returned HTTP 500, so the status page went blank. The server now keeps the last database statistics in memory and recomputes them at most once every 30 seconds; requests serve the cached copy and a stale copy triggers a single background refresh, so no request runs the scans. The route-count split is folded into the cached stats, removing the separate per-request live-route count query. The oldest/newest route timestamps now read one row from each end of the last_updated index instead of scanning both tables, and select the column directly so the driver parses it into time.Time; the old aggregate returned an untyped string that failed to scan and logged a warning every call. Model: opus-4-8
This commit is contained in:
+12
-71
@@ -179,37 +179,14 @@ func (s *Server) handleStatusJSON() http.HandlerFunc {
|
||||
|
||||
metrics := s.streamer.GetMetrics()
|
||||
|
||||
// Get database stats with timeout. The channels are buffered so the
|
||||
// goroutine's send never blocks if the timeout wins and nothing here
|
||||
// receives; otherwise it would block forever and leak.
|
||||
statsChan := make(chan database.Stats, 1)
|
||||
errChan := make(chan error, 1)
|
||||
|
||||
go func() {
|
||||
dbStats, err := s.db.GetStatsContext(ctx)
|
||||
if err != nil {
|
||||
s.logger.Debug("Database stats query failed", "error", err)
|
||||
errChan <- err
|
||||
|
||||
return
|
||||
}
|
||||
statsChan <- dbStats
|
||||
}()
|
||||
|
||||
var dbStats database.Stats
|
||||
select {
|
||||
case <-ctx.Done():
|
||||
s.logger.Error("Database stats timeout in status.json")
|
||||
writeJSONError(w, http.StatusRequestTimeout, "Database timeout")
|
||||
|
||||
return
|
||||
case err := <-errChan:
|
||||
// Serve database statistics from the cache, which runs the table scans at
|
||||
// most once per interval so this request does not.
|
||||
dbStats, err := s.stats.get()
|
||||
if err != nil {
|
||||
s.logger.Error("Failed to get database stats", "error", err)
|
||||
writeJSONError(w, http.StatusInternalServerError, err.Error())
|
||||
|
||||
return
|
||||
case dbStats = <-statsChan:
|
||||
// Success
|
||||
}
|
||||
|
||||
uptime := time.Since(metrics.ConnectedSince).Truncate(time.Second).String()
|
||||
@@ -219,13 +196,6 @@ func (s *Server) handleStatusJSON() http.HandlerFunc {
|
||||
|
||||
const bitsPerMegabit = 1000000.0
|
||||
|
||||
// Get route counts from database
|
||||
ipv4Routes, ipv6Routes, err := s.db.GetLiveRouteCountsContext(ctx)
|
||||
if err != nil {
|
||||
s.logger.Warn("Failed to get live route counts", "error", err)
|
||||
// Continue with zero counts
|
||||
}
|
||||
|
||||
// Get route update metrics
|
||||
routeMetrics := s.streamer.GetMetricsTracker().GetRouteMetrics()
|
||||
|
||||
@@ -259,8 +229,8 @@ func (s *Server) handleStatusJSON() http.HandlerFunc {
|
||||
Peers: dbStats.Peers,
|
||||
DatabaseSizeBytes: dbStats.FileSizeBytes,
|
||||
LiveRoutes: dbStats.LiveRoutes,
|
||||
IPv4Routes: ipv4Routes,
|
||||
IPv6Routes: ipv6Routes,
|
||||
IPv4Routes: dbStats.IPv4Routes,
|
||||
IPv6Routes: dbStats.IPv6Routes,
|
||||
OldestRoute: dbStats.OldestRoute,
|
||||
NewestRoute: dbStats.NewestRoute,
|
||||
IPv4UpdatesPerSec: routeMetrics.IPv4UpdatesPerSec,
|
||||
@@ -400,36 +370,14 @@ func (s *Server) handleStats() http.HandlerFunc {
|
||||
|
||||
metrics := s.streamer.GetMetrics()
|
||||
|
||||
// Get database stats with timeout. The channels are buffered so the
|
||||
// goroutine's send never blocks if the timeout wins and nothing here
|
||||
// receives; otherwise it would block forever and leak.
|
||||
statsChan := make(chan database.Stats, 1)
|
||||
errChan := make(chan error, 1)
|
||||
|
||||
go func() {
|
||||
dbStats, err := s.db.GetStatsContext(ctx)
|
||||
if err != nil {
|
||||
s.logger.Debug("Database stats query failed", "error", err)
|
||||
errChan <- err
|
||||
|
||||
return
|
||||
}
|
||||
statsChan <- dbStats
|
||||
}()
|
||||
|
||||
var dbStats database.Stats
|
||||
select {
|
||||
case <-ctx.Done():
|
||||
s.logger.Error("Database stats timeout")
|
||||
// Don't write response here - timeout middleware already handles it
|
||||
return
|
||||
case err := <-errChan:
|
||||
// Serve database statistics from the cache, which runs the table scans at
|
||||
// most once per interval so this request does not.
|
||||
dbStats, err := s.stats.get()
|
||||
if err != nil {
|
||||
s.logger.Error("Failed to get database stats", "error", err)
|
||||
writeJSONError(w, http.StatusInternalServerError, err.Error())
|
||||
|
||||
return
|
||||
case dbStats = <-statsChan:
|
||||
// Success
|
||||
}
|
||||
|
||||
uptime := time.Since(metrics.ConnectedSince).Truncate(time.Second).String()
|
||||
@@ -439,13 +387,6 @@ func (s *Server) handleStats() http.HandlerFunc {
|
||||
|
||||
const bitsPerMegabit = 1000000.0
|
||||
|
||||
// Get route counts from database
|
||||
ipv4Routes, ipv6Routes, err := s.db.GetLiveRouteCountsContext(ctx)
|
||||
if err != nil {
|
||||
s.logger.Warn("Failed to get live route counts", "error", err)
|
||||
// Continue with zero counts
|
||||
}
|
||||
|
||||
// Get route update metrics
|
||||
routeMetrics := s.streamer.GetMetricsTracker().GetRouteMetrics()
|
||||
|
||||
@@ -537,8 +478,8 @@ func (s *Server) handleStats() http.HandlerFunc {
|
||||
Peers: dbStats.Peers,
|
||||
DatabaseSizeBytes: dbStats.FileSizeBytes,
|
||||
LiveRoutes: dbStats.LiveRoutes,
|
||||
IPv4Routes: ipv4Routes,
|
||||
IPv6Routes: ipv6Routes,
|
||||
IPv4Routes: dbStats.IPv4Routes,
|
||||
IPv6Routes: dbStats.IPv6Routes,
|
||||
OldestRoute: dbStats.OldestRoute,
|
||||
NewestRoute: dbStats.NewestRoute,
|
||||
IPv4UpdatesPerSec: routeMetrics.IPv4UpdatesPerSec,
|
||||
|
||||
Reference in New Issue
Block a user