revert(dingtalk): remove health monitoring to avoid conflict with SDK auto-reconnect
The custom health monitoring implementation conflicted with the SDK's built-in auto-reconnect mechanism, causing "panic: send on closed channel" errors after extended runtime. The dingtalk-stream-sdk-go already has: - AutoReconnect enabled by default - 120-second heartbeat ping/pong detection - Automatic reconnection on connection errors Having our own healthMonitor calling Close()/Start() while the SDK was also attempting reconnection created a race condition where the read goroutine tried to send to a closed channel. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
parent
fa68f1e12e
commit
c25ad85beb
1 changed files with 1 additions and 94 deletions
|
|
@ -21,14 +21,6 @@ import (
|
|||
"github.com/sipeed/picoclaw/pkg/utils"
|
||||
)
|
||||
|
||||
// Health check constants
|
||||
const (
|
||||
healthCheckInterval = 5 * time.Minute // Check every 5 minutes
|
||||
maxSilenceDuration = 30 * time.Minute // Max time without messages before recovery
|
||||
recoveryDelay = 2 * time.Second // Delay before reconnecting
|
||||
recoveryRetryDelay = 30 * time.Second // Delay before retrying failed recovery
|
||||
)
|
||||
|
||||
// DingTalkSession stores complete session information for proactive messaging
|
||||
type DingTalkSession struct {
|
||||
SessionWebhook string `json:"session_webhook"`
|
||||
|
|
@ -56,9 +48,6 @@ type DingTalkChannel struct {
|
|||
tokenExpiry time.Time
|
||||
tokenMu sync.RWMutex
|
||||
httpClient *http.Client
|
||||
// Health monitoring
|
||||
lastMessageTime time.Time
|
||||
mu sync.RWMutex
|
||||
}
|
||||
|
||||
// NewDingTalkChannel creates a new DingTalk channel instance
|
||||
|
|
@ -86,7 +75,6 @@ func (c *DingTalkChannel) Start(ctx context.Context) error {
|
|||
logger.InfoC("dingtalk", "Starting DingTalk channel (Stream Mode)...")
|
||||
|
||||
c.ctx, c.cancel = context.WithCancel(ctx)
|
||||
c.lastMessageTime = time.Now() // Initialize on start
|
||||
|
||||
// Initialize HTTP client for robot API
|
||||
c.httpClient = &http.Client{Timeout: 30 * time.Second}
|
||||
|
|
@ -106,9 +94,6 @@ func (c *DingTalkChannel) Start(ctx context.Context) error {
|
|||
return err
|
||||
}
|
||||
|
||||
// Start health monitoring goroutine
|
||||
go c.healthMonitor()
|
||||
|
||||
c.SetRunning(true)
|
||||
logger.InfoCF("dingtalk", "DingTalk channel started", map[string]any{
|
||||
"proactive_send": c.config.ProactiveSend,
|
||||
|
|
@ -138,81 +123,6 @@ func (c *DingTalkChannel) startStreamClient() error {
|
|||
return nil
|
||||
}
|
||||
|
||||
// healthMonitor periodically checks connection health and triggers recovery if needed
|
||||
func (c *DingTalkChannel) healthMonitor() {
|
||||
ticker := time.NewTicker(healthCheckInterval)
|
||||
defer ticker.Stop()
|
||||
|
||||
for {
|
||||
select {
|
||||
case <-c.ctx.Done():
|
||||
return
|
||||
case <-ticker.C:
|
||||
c.checkAndRecover()
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// checkAndRecover checks if connection is stale and triggers recovery
|
||||
func (c *DingTalkChannel) checkAndRecover() {
|
||||
c.mu.RLock()
|
||||
silenceDuration := time.Since(c.lastMessageTime)
|
||||
c.mu.RUnlock()
|
||||
|
||||
logger.DebugCF("dingtalk", "Health check: silence duration", map[string]any{"duration": silenceDuration})
|
||||
|
||||
if silenceDuration >= maxSilenceDuration {
|
||||
logger.InfoCF(
|
||||
"dingtalk",
|
||||
"Connection appears stale (no messages for configured duration), triggering recovery",
|
||||
map[string]any{"silenceDuration": silenceDuration},
|
||||
)
|
||||
c.recoverConnection()
|
||||
}
|
||||
}
|
||||
|
||||
// recoverConnection attempts to recover the stream connection
|
||||
func (c *DingTalkChannel) recoverConnection() {
|
||||
// Close old client
|
||||
if c.streamClient != nil {
|
||||
c.streamClient.Close()
|
||||
time.Sleep(recoveryDelay)
|
||||
}
|
||||
|
||||
// Attempt to reconnect
|
||||
for {
|
||||
select {
|
||||
case <-c.ctx.Done():
|
||||
logger.InfoC("dingtalk", "Recovery aborted: context canceled")
|
||||
return
|
||||
default:
|
||||
}
|
||||
|
||||
err := c.startStreamClient()
|
||||
if err == nil {
|
||||
logger.InfoCF("dingtalk", "Connection recovered successfully", map[string]any{})
|
||||
c.mu.Lock()
|
||||
c.lastMessageTime = time.Now()
|
||||
c.mu.Unlock()
|
||||
return
|
||||
}
|
||||
|
||||
logger.WarnCF(
|
||||
"dingtalk",
|
||||
"Recovery failed, retrying",
|
||||
map[string]any{"error": err, "retryDelay": recoveryRetryDelay},
|
||||
)
|
||||
time.Sleep(recoveryRetryDelay)
|
||||
}
|
||||
}
|
||||
|
||||
// updateLastMessageTime updates the last message timestamp
|
||||
func (c *DingTalkChannel) updateLastMessageTime() {
|
||||
c.mu.Lock()
|
||||
c.lastMessageTime = time.Now()
|
||||
c.mu.Unlock()
|
||||
}
|
||||
|
||||
// Stop gracefully stops the DingTalk channel
|
||||
func (c *DingTalkChannel) Stop(ctx context.Context) error {
|
||||
logger.InfoC("dingtalk", "Stopping DingTalk channel...")
|
||||
|
|
@ -293,9 +203,6 @@ func (c *DingTalkChannel) onChatBotMessageReceived(
|
|||
ctx context.Context,
|
||||
data *chatbot.BotCallbackDataModel,
|
||||
) ([]byte, error) {
|
||||
// Update last message time for health monitoring
|
||||
c.updateLastMessageTime()
|
||||
|
||||
// Extract message content from Text field
|
||||
content := data.Text.Content
|
||||
if content == "" {
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue