fix: fix a deadlock issue when it occurs simultaneously: a goroutine is trying to lock, and another goroutine waits since full of channel buffers go-ldap#440

Author: t2y

conn.go

@@ -327,8 +327,6 @@ func (l *Conn) nextMessageID() int64 {
 // GetLastError returns the last recorded error from goroutines like processMessages and reader.
 // Only the last recorded error will be returned.
 func (l *Conn) GetLastError() error {
-	l.messageMutex.Lock()
-	defer l.messageMutex.Unlock()
 	return l.err
 }
 

response.go

@@ -60,10 +60,6 @@ func (r *searchResponse) Next() bool {
 	if r.err != nil {
 		return false
 	}
-	r.err = r.conn.GetLastError()
-	if r.err != nil {
-		return false
-	}
 	r.entry = res.Entry
 	r.referral = res.Referral
 	r.controls = res.Controls

v3/conn.go

@@ -327,8 +327,6 @@ func (l *Conn) nextMessageID() int64 {
 // GetLastError returns the last recorded error from goroutines like processMessages and reader.
 // // Only the last recorded error will be returned.
 func (l *Conn) GetLastError() error {
-	l.messageMutex.Lock()
-	defer l.messageMutex.Unlock()
 	return l.err
 }
 

v3/response.go

@@ -60,10 +60,6 @@ func (r *searchResponse) Next() bool {
 	if r.err != nil {
 		return false
 	}
-	r.err = r.conn.GetLastError()
-	if r.err != nil {
-		return false
-	}
 	r.entry = res.Entry
 	r.referral = res.Referral
 	r.controls = res.Controls