[Node-107] disconnected from Loadbalancer due to explicit disconnect call


(Ersoy Vatansever) #1

Using v1.7.4 on 1 cluster 3 data node, 1 master and 1 loadbalancer.
This issue fires from a week. I dont know what happend and start to fire this issue.
Can any one help me ?

LOG1 :

[2017-09-26 15:24:06,545][DEBUG][transport.netty ] [Node-107] disconnecting from [[Loadbalancer][aiZsZddTTG6qK_suRpZlyQ][PENTAv3ELSM16][inet[/xxx:9300]]{data=false, master=false}] due to explicit disconnect call
[2017-09-26 15:24:06,546][TRACE][transport.tracer ] [Node-107] [1115893][cluster:monitor/nodes/info[n]] received request
[2017-09-26 15:24:06,547][TRACE][transport.tracer ] [Node-107] [1115893][cluster:monitor/nodes/info[n]] sent response
[2017-09-26 15:24:06,547][TRACE][transport.netty ] [Node-107] disconnected from [[Loadbalancer][aiZsZddTTG6qK_suRpZlyQ][PENTAv3ELSM16][inet[/xxx:9300]]{data=false, master=false}] due to explicit disconnect call
[2017-09-26 15:24:06,549][TRACE][transport.tracer ] [Node-107] [254013][internal:discovery/zen/publish] sent response
[2017-09-26 15:24:06,549][TRACE][transport.tracer ] [Node-107] [61330242][indices:data/read/search[phase/query]] sent response
[2017-09-26 15:24:06,549][TRACE][transport.tracer ] [Node-107] [61330238][indices:data/read/search[phase/query]] sent response

LOG2 :

[2017-09-26 00:00:06,675][DEBUG][transport.netty ] [Loadbalancer] connected to node [[Master-105][c2tvdIkXQICHhRlIgS7_vQ][MasterElastic][inet[/xxx:9300]]{data=false, master=true}]
[2017-09-26 00:00:06,675][DEBUG][discovery.zen.publish ] [Loadbalancer] received cluster state version 63441
[2017-09-26 00:00:06,675][DEBUG][cluster.service ] [Loadbalancer] processing [zen-disco-receive(from master [[Master-105][c2tvdIkXQICHhRlIgS7_vQ][MasterElastic][inet[/xxx:9300]]{data=false, master=true}])]: execute
[2017-09-26 00:00:06,675][DEBUG][discovery.zen.fd ] [Loadbalancer] [master] restarting fault detection against master [[Master-105][c2tvdIkXQICHhRlIgS7_vQ][MasterElastic][inet[/xxx:9300]]{data=false, master=true}], reason [new cluster state received and we are monitoring the wrong master [null]]
[2017-09-26 00:00:06,675][DEBUG][discovery.zen ] [Loadbalancer] got first state from fresh master [c2tvdIkXQICHhRlIgS7_vQ]
[2017-09-26 00:00:06,675][DEBUG][cluster.service ] [Loadbalancer] cluster state updated, version [63441], source [zen-disco-receive(from master [[Master-105][c2tvdIkXQICHhRlIgS7_vQ][MasterElastic][inet[/xxx:9300]]{data=false, master=true}])]
[2017-09-26 00:00:06,675][INFO ][cluster.service ] [Loadbalancer] detected_master [Master-105][c2tvdIkXQICHhRlIgS7_vQ][MasterElastic][inet[/xxx:9300]]{data=false, master=true}, added {[Master-105][c2tvdIkXQICHhRlIgS7_vQ][MasterElastic][inet[/xxx:9300]]{data=false, master=true},}, reason: zen-disco-receive(from master [[Master-105][c2tvdIkXQICHhRlIgS7_vQ][MasterElastic][inet[/xxx:9300]]{data=false, master=true}])
[2017-09-26 00:00:06,675][DEBUG][cluster.service ] [Loadbalancer] set local cluster state to version 63441
[2017-09-26 00:00:06,675][DEBUG][cluster.service ] [Loadbalancer] processing [zen-disco-receive(from master [[Master-105][c2tvdIkXQICHhRlIgS7_vQ][MasterElastic][inet[/xxx:9300]]{data=false, master=true}])]: took 1ms done applying updated cluster_state (version: 63441)
[2017-09-26 00:00:06,690][DEBUG][cluster.service ] [Loadbalancer] processing [finalize_join ([Master-105][c2tvdIkXQICHhRlIgS7_vQ][MasterElastic][inet[/xxx:9300]]{data=false, master=true})]: execute
[2017-09-26 00:00:06,690][DEBUG][cluster.service ] [Loadbalancer] processing [finalize_join ([Master-105][c2tvdIkXQICHhRlIgS7_vQ][MasterElastic][inet[/xxx:9300]]{data=false, master=true})]: took 0s no change in cluster_state

Here is my loadbalancer's yml file.

##################### Elasticsearch Configuration Example #####################

cluster.name: penta-elasticsearch

node.name: "Node-166"
index.routing.allocation.disable_allocation: false

node.master: false
node.data: false

index.refresh_interval: 10s

index.number_of_shards: 5

index.number_of_replicas: 1

bootstrap.mlockall: true
network.host: xx.xx.x.xxx

transport.tcp.port: 9300

transport.tcp.compress: true
http.port: 9200

index.query.bool.max_clause_count: 1000000

(Ersoy Vatansever) #2

@Clinton_Gormley1 could you help me about this ?
Thanks you.


(David Pilato) #3

This version is not maintained anymore.
1.7.4 is not the last one. Was released in Dec 2015.
1.7.6 is the latest 1.7 one, released in Nov 2016.

At the very least you should upgrade to that one but for sure you need to upgrade to 2.x or better 5.x.


(Ersoy Vatansever) #4

thanks for your help. I think it's be better upgrate to 5.x


(Ersoy Vatansever) #5

@dadoonet Can i ask one question to you ?
Is it possible to use multiple loadbalancer for one cluster ?
Thank you.


(David Pilato) #6

You can use as many client nodes you wish in a cluster.


(system) #7

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.