feat: make pulse-sensor-proxy rate limits configurable

Add support for configuring rate limits via config.yaml to allow
administrators to tune the proxy for different deployment sizes.

Changes:
- Add RateLimitConfig struct to config.go with per_peer_interval_ms and per_peer_burst
- Update newRateLimiter() to accept optional RateLimitConfig parameter
- Load rate limit config from YAML and apply overrides to defaults
- Update tests to pass nil for default behavior
- Add comprehensive config.example.yaml with documentation

Configuration examples:
- Small (1-3 nodes): 1000ms interval, burst 5 (default)
- Medium (4-10 nodes): 500ms interval, burst 10
- Large (10+ nodes): 250ms interval, burst 20

Defaults remain conservative (1 req/sec, burst 5) to support most
deployments while allowing customization for larger environments.

Related: #46b8b8d08 (rate limit fix for multi-node support)
This commit is contained in:
rcourtman
2025-10-21 11:25:21 +00:00
parent 46b8b8d08f
commit ca534e2b6f
5 changed files with 76 additions and 5 deletions

View File

@@ -0,0 +1,39 @@
# Pulse Sensor Proxy Configuration
# This file is optional. If not provided, the proxy will use sensible defaults.
# Network Configuration
# Specify which networks are allowed to connect to the proxy
# If not specified, the proxy will auto-detect host IP addresses
allowed_source_subnets:
- "127.0.0.1/32" # Localhost
- "192.168.0.0/24" # Local network
# Peer Authorization
# Specify which UIDs/GIDs are allowed to connect
# Required when running Pulse in a container (use mapped UID/GID from container)
allowed_peer_uids: [100999] # Container pulse user UID
allowed_peer_gids: [100996] # Container pulse group GID
# ID-Mapped Root Authentication
# Allow connections from ID-mapped root users (for LXC containers)
allow_idmapped_root: true
allowed_idmap_users:
- root
# Metrics Server
# Address for Prometheus metrics endpoint
metrics_address: "127.0.0.1:9127"
# Rate Limiting (Optional)
# Control how frequently peers can make requests to prevent abuse
# Adjust these values based on your deployment size:
# - Small (1-3 nodes): Use defaults (1000ms, burst 5)
# - Medium (4-10 nodes): 500ms, burst 10
# - Large (10+ nodes): 250ms, burst 20
rate_limit:
per_peer_interval_ms: 1000 # Minimum milliseconds between requests per peer (1000ms = 1 qps)
per_peer_burst: 5 # Number of requests allowed in a burst (supports up to 5 simultaneous requests)
# Default values if not specified:
# per_peer_interval_ms: 1000 (1 second = 1 qps = 60 requests/min)
# per_peer_burst: 5

View File

@@ -11,6 +11,12 @@ import (
"gopkg.in/yaml.v3"
)
// RateLimitConfig holds rate limiting configuration
type RateLimitConfig struct {
PerPeerIntervalMs int `yaml:"per_peer_interval_ms"` // Milliseconds between requests per peer
PerPeerBurst int `yaml:"per_peer_burst"` // Number of requests allowed in a burst
}
// Config holds proxy configuration
type Config struct {
AllowedSourceSubnets []string `yaml:"allowed_source_subnets"`
@@ -20,6 +26,8 @@ type Config struct {
AllowedPeerUIDs []uint32 `yaml:"allowed_peer_uids"`
AllowedPeerGIDs []uint32 `yaml:"allowed_peer_gids"`
AllowedIDMapUsers []string `yaml:"allowed_idmap_users"`
RateLimit *RateLimitConfig `yaml:"rate_limit,omitempty"`
}
// loadConfig loads configuration from file and environment variables
@@ -150,6 +158,14 @@ func loadConfig(configPath string) (*Config, error) {
}
}
// Log rate limit configuration if provided
if cfg.RateLimit != nil {
log.Info().
Int("per_peer_interval_ms", cfg.RateLimit.PerPeerIntervalMs).
Int("per_peer_burst", cfg.RateLimit.PerPeerBurst).
Msg("Rate limit configuration loaded from config file")
}
return cfg, nil
}

View File

@@ -351,7 +351,7 @@ func runProxy() {
socketPath: socketPath,
sshKeyPath: sshKeyPath,
knownHosts: knownHostsManager,
rateLimiter: newRateLimiter(metrics),
rateLimiter: newRateLimiter(metrics, cfg.RateLimit),
nodeGate: newNodeGate(),
config: cfg,
metrics: metrics,

View File

@@ -50,14 +50,30 @@ var (
)
// newRateLimiter creates a new rate limiter with cleanup loop
func newRateLimiter(metrics *ProxyMetrics) *rateLimiter {
// If rateLimitCfg is provided, it overrides the default rate limit settings
func newRateLimiter(metrics *ProxyMetrics, rateLimitCfg *RateLimitConfig) *rateLimiter {
// Use defaults
perPeerLimit := defaultPerPeerLimit
perPeerBurst := defaultPerPeerBurst
// Override with config if provided
if rateLimitCfg != nil {
if rateLimitCfg.PerPeerIntervalMs > 0 {
interval := time.Duration(rateLimitCfg.PerPeerIntervalMs) * time.Millisecond
perPeerLimit = rate.Every(interval)
}
if rateLimitCfg.PerPeerBurst > 0 {
perPeerBurst = rateLimitCfg.PerPeerBurst
}
}
rl := &rateLimiter{
entries: make(map[peerID]*limiterEntry),
quitChan: make(chan struct{}),
globalSem: make(chan struct{}, defaultGlobalConcurrency),
policy: limiterPolicy{
perPeerLimit: defaultPerPeerLimit,
perPeerBurst: defaultPerPeerBurst,
perPeerLimit: perPeerLimit,
perPeerBurst: perPeerBurst,
perPeerConcurrency: defaultPerPeerConcurrency,
globalConcurrency: defaultGlobalConcurrency,
penaltyDuration: defaultPenaltyDuration,

View File

@@ -7,7 +7,7 @@ import (
func TestRateLimiterPenalizeMetrics(t *testing.T) {
metrics := NewProxyMetrics("test")
rl := newRateLimiter(metrics)
rl := newRateLimiter(metrics, nil)
rl.policy.penaltyDuration = 10 * time.Millisecond
start := time.Now()