I have a zk cluster set up , however, I ran into errors:
in zookeeper.properties, I have jute.maxbuffer=1073741824 But in zk logs during start up, I've got 2024-07-09 14:12:58,105 [myid:] - INFO [QuorumPeermyid=1(secure=disabled):o.a.z.s.ZooKeeperServer@1633] - The max bytes for all large requests are set to 104857600
And eventually, I've got IO len error: zookeeper-0 java.io.IOException: Len error. A message from /10.42.18.0:45980 with advertised length of 973078528 is either a malformed message or too large to process (length is greater than jute.maxbuffer=1048575)
my question is why the jute.maxbuffer value in zookeeper.properties is not taken
If the value of jute.maxbuffer is too high, ZK seems to ignore it
See this link https://user.zookeeper.apache.narkive.com/bveQQuI1/configuring-jute-maxbuffer-on-zookeeper-nodes