Skip to content

fix: recover zombie connections on KeepAliveTimeout #126

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
FlavioPulli wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix: recover zombie connections on KeepAliveTimeout #126

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
35 changes: 32 additions & 3 deletions pkg/whatsmeow/service/whatsmeow.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1988,6 +1988,35 @@ func (mycli *MyClient) myEventHandler(rawEvt interface{}) {
mycli.loggerWrapper.GetLogger(instanceID).LogError("[%s] Failed to restart instance: %v", instanceID, err)
}
}(mycli.userID)
case *events.KeepAliveTimeout:
doWebhook = true
postMap["event"] = "KeepAliveTimeout"
postMap["data"] = map[string]interface{}{
"ErrorCount": evt.ErrorCount,
"LastSuccess": evt.LastSuccess,
}
mycli.loggerWrapper.GetLogger(mycli.userID).LogWarn("[%s] Keepalive ping timed out (%d consecutive, last success: %s)", mycli.userID, evt.ErrorCount, evt.LastSuccess.Format(time.RFC3339))

// With EnableAutoReconnect disabled, a TCP connection that silently dies
// keeps timing out keepalives forever without ever emitting
// events.Disconnected — the instance stays "connected" but is a zombie.
// whatsmeow's own docs suggest using this event to force a faster
// disconnect+reconnect, so after 3 consecutive timeouts restart the
// instance through the same path the Disconnected handler uses. Exact
// match keeps counts 4, 5, ... from piling up restarts while one is
// already underway.
if evt.ErrorCount == 3 {
go func(instanceID string) {
mycli.loggerWrapper.GetLogger(instanceID).LogWarn("[%s] 3 consecutive keepalive timeouts, restarting instance", instanceID)
if err := mycli.service.ReconnectClient(instanceID); err != nil {
mycli.loggerWrapper.GetLogger(instanceID).LogError("[%s] Failed to restart instance: %v", instanceID, err)
}
}(mycli.userID)
}
case *events.KeepAliveRestored:
doWebhook = true
postMap["event"] = "KeepAliveRestored"
mycli.loggerWrapper.GetLogger(mycli.userID).LogInfo("[%s] Keepalive pings restored", mycli.userID)
case *events.LabelEdit:
doWebhook = true
postMap["event"] = "LabelEdit"
Expand Down Expand Up @@ -2222,7 +2251,7 @@ func (w *whatsmeowService) CallWebhook(instance *instance_model.Instance, queueN
w.loggerWrapper.GetLogger(instance.Id).LogInfo("[%s] Event received of type %s", instance.Id, eventType)
w.sendToQueueOrWebhook(instance, queueName, jsonData)
}
case "Connected", "PairSuccess", "TemporaryBan", "LoggedOut", "ConnectFailure", "Disconnected":
case "Connected", "PairSuccess", "TemporaryBan", "LoggedOut", "ConnectFailure", "Disconnected", "KeepAliveTimeout", "KeepAliveRestored":
if contains(subscriptions, "CONNECTION") {
w.loggerWrapper.GetLogger(instance.Id).LogInfo("[%s] Event received of type %s", instance.Id, eventType)
w.sendToQueueOrWebhook(instance, queueName, jsonData)
Expand Down Expand Up @@ -2510,7 +2539,7 @@ func (w *whatsmeowService) SendToGlobalQueues(eventType string, payload []byte,
globalEventType = "CHAT_PRESENCE"
case "CallOffer", "CallAccept", "CallTerminate", "CallOfferNotice", "CallRelayLatency":
globalEventType = "CALL"
case "Connected", "PairSuccess", "TemporaryBan", "LoggedOut", "ConnectFailure", "Disconnected":
case "Connected", "PairSuccess", "TemporaryBan", "LoggedOut", "ConnectFailure", "Disconnected", "KeepAliveTimeout", "KeepAliveRestored":
globalEventType = "CONNECTION"
case "LabelEdit", "LabelAssociationChat", "LabelAssociationMessage":
globalEventType = "LABEL"
Expand Down Expand Up @@ -2572,7 +2601,7 @@ func (w *whatsmeowService) SendToGlobalQueues(eventType string, payload []byte,
globalEventType = "CHAT_PRESENCE"
case "CallOffer", "CallAccept", "CallTerminate", "CallOfferNotice", "CallRelayLatency":
globalEventType = "CALL"
case "Connected", "PairSuccess", "TemporaryBan", "LoggedOut", "ConnectFailure", "Disconnected":
case "Connected", "PairSuccess", "TemporaryBan", "LoggedOut", "ConnectFailure", "Disconnected", "KeepAliveTimeout", "KeepAliveRestored":
globalEventType = "CONNECTION"
case "LabelEdit", "LabelAssociationChat", "LabelAssociationMessage":
globalEventType = "LABEL"
Expand Down