kafka-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From 隋玉增 <suiyuz...@163.com>
Subject add log//Fw:Temporary node of broke in zookeeper lost
Date Mon, 10 Oct 2016 03:29:59 GMT
   The image may not be displayed. I have added it to the attachment.  I had added some log
in the function handleStateChanged, such as follows:
def handleStateChanged() {
  info("re-registering broker info in ZK for broker " + brokerId)

 It was called when the session expired but was not called again when the session re-established.
And the function handleNewSession was also not called. It was probabilistic and the network
is not very stable. 


-------- Forwarding messages --------
From: "隋玉增" <suiyuzeng@163.com>
Date: 2016-10-09 21:06:58
To: users@kafka.apache.org,dev@kafka.apache.org
Subject: Temporary node of broke in zookeeper lost

   I meet a issue that the temporary node of broke in zookeeper was lost while the process
of the broker still exist. At this time, the controller would consider it to be offline. 
According to zkClient log, I find the session is timeout, but handleStateChanged and handleNewSession(in
KafkaHealthcheck) is not called after the session re-establish. The following is the zkclient
log, "zk state changed" and "re-reginstering broker info in zk" was not printed after syncconnected.
   So I want to check the node periodically and call the function register()(in KafkaHealthcheck)
when the node is not exist.  What's about this solution?


View raw message