SolrCloud election leader

Question

I would like to use SolrCloud (Solr4.10.3 ).. I follow this wiki(Example B: https://wiki.apache.org/solr/SolrCloud#Example_B:_Simple_two_shard_cluster_with_shard_replicas )

SolrCloud uses leaders and an overseer as an implementation detail. This means that some nodes/replicas will play special roles. You don't need to worry if the instance you kill is a leader or the cluster overseer - if you happen to kill one of these, automatic fail over will choose new leaders or a new overseer transparently to the user and they will seamlessly takeover their respective jobs. Any Solr instance can be promoted to one of these roles.

When I stop the first instance where i run the zookeeper, i can't get another leader.

I get this Info in the same node(Example2 log file) :

INFO  - 2015-03-03 11:35:39.508; org.apache.solr.cloud.OverseerElectionContext; I am going to be the leader 192.168.1.38:7574_solr
INFO  - 2015-03-03 11:35:39.548; org.apache.solr.common.cloud.SolrZkClient; makePath: /overseer_elect/leader
INFO  - 2015-03-03 11:35:39.576; org.apache.solr.cloud.Overseer; Overseer (id=93413623381032961-192.168.1.38:7574_solr-n_0000000001) starting
WARN  - 2015-03-03 11:35:39.957; org.apache.zookeeper.ClientCnxn$SendThread;    
Session 0x14bdf320ecf0001 for server 0:0:0:0:0:0:0:1/0:0:0:0:0:0:0:1:9983,    
unexpected error, closing socket connection and attempting reconnect
java.io.IOException: An existing connection was forcibly closed by the remote host
at sun.nio.ch.SocketDispatcher.read0(Native Method)
at sun.nio.ch.SocketDispatcher.read(SocketDispatcher.java:43)
at sun.nio.ch.IOUtil.readIntoNativeBuffer(IOUtil.java:223)
at sun.nio.ch.IOUtil.read(IOUtil.java:192)
at sun.nio.ch.SocketChannelImpl.read(SocketChannelImpl.java:379)
at org.apache.zookeeper.ClientCnxnSocketNIO.doIO(ClientCnxnSocketNIO.java:68)
at org.apache.zookeeper.ClientCnxnSocketNIO.doTransport(ClientCnxnSocketNIO.java:366)
at org.apache.zookeeper.ClientCnxn$SendThread.run(ClientCnxn.java:1081)
INFO  - 2015-03-03 11:35:40.064; org.apache.solr.common.cloud.ConnectionManager; Watcher org.apache.solr.common.cloud.ConnectionManager@1537a92 name:ZooKeeperConnection Watcher:localhost:9983 got event WatchedEvent state:Disconnected type:None path:null path:null type:None
INFO  - 2015-03-03 11:35:40.064; org.apache.solr.common.cloud.ConnectionManager; zkClient has disconnected

In other Node (Example2B) i get that in the log file

INFO  - 2015-03-03 11:35:39.507; org.apache.solr.cloud.ShardLeaderElectionContext; Running the leader process for shard shard1
INFO  - 2015-03-03 11:35:39.553; org.apache.solr.cloud.ShardLeaderElectionContext; Checking if I (core=collection1,coreNodeName=core_node3) should try and be the leader.
INFO  - 2015-03-03 11:35:39.574; org.apache.solr.cloud.ShardLeaderElectionContext; My last published State was Active, it's okay to be the leader.
INFO  - 2015-03-03 11:35:39.578; org.apache.solr.cloud.ShardLeaderElectionContext; I may be the new leader - try and sync
WARN  - 2015-03-03 11:35:39.956; org.apache.zookeeper.ClientCnxn$SendThread; Session 0x14bdf320ecf0002 for server 0:0:0:0:0:0:0:1/0:0:0:0:0:0:0:1:9983, unexpected error, closing socket connection and attempting reconnect
java.io.IOException: An existing connection was forcibly closed by the remote host
at sun.nio.ch.SocketDispatcher.read0(Native Method)
at sun.nio.ch.SocketDispatcher.read(SocketDispatcher.java:43)
at sun.nio.ch.IOUtil.readIntoNativeBuffer(IOUtil.java:223)
at sun.nio.ch.IOUtil.read(IOUtil.java:192)
at sun.nio.ch.SocketChannelImpl.read(SocketChannelImpl.java:379)
at org.apache.zookeeper.ClientCnxnSocketNIO.doIO(ClientCnxnSocketNIO.java:68)
at org.apache.zookeeper.ClientCnxnSocketNIO.doTransport(ClientCnxnSocketNIO.java:366)
at org.apache.zookeeper.ClientCnxn$SendThread.run(ClientCnxn.java:1081)

We can see in the the same instance, there is a call for election leader process but it failed.. i don't see the probleme.

How to get automatically another leader?

Is there a special config in my solr server?

How to force the election for another leader?

Thx!

i get this Warn I am going to be the leader 192.168.1.38:7574_solr ..INFO - ; org.apache.solr.common.cloud.SolrZkClient; makePath: /overseer_elect/leader ..WARN - ;unexpected error, closing socket connection and attempting reconnect ..java.io.IOException: An existing connection was forcibly closed by the remote host — user4054093
after few search, i'm in the point to use solr with external zookeeper....Although Solr comes bundled with Apache ZooKeeper, you should consider yourself discouraged from using this internal ZooKeeper in production, because shutting down a redundant Solr instance will also shut down its ZooKeeper server, which might not be quite so redundant. Because a ZooKeeper ensemble must have a quorum of more than half its servers running at any given time, this can be a problem. — user4054093
my objective is to run solr with external zookeeper into 2 host.. I can't terminate this task.. i use this link : SolrCloud HOWTO, Configure-Solr-Cloud-with-External-Zookeeper — user4054093
It's not recomended to use embeded zookeeper for our production.. when we stop a solr instance where run a zookeeper, there is no leader election..we should use external zookeeper! — user4054093

Unknown Unknown · Accepted Answer · 2015-03-05T16:08:49

config zookeeper(host : xxx.xxx.xxx.1):

tickTime=2000
initLimit=10
syncLimit=5
clientPort=2181
server.1=127.0.0.1:2888:3888

Start Solr1 (host : xxx.xxx.xxx.2)

java -Dbootstrap_confdir=./solr/collection1/conf 
                 -Dcollection.configName=solr1 
                 -Dsolr.allow.unsafe.resourceloading=true 
                 -DzkHost=xxx.xxx.xxx.1:2181 
                 -DnumShards=2 
                 -jar start.jar

Start Solr2 (host : xxx.xxx.xxx.3)

java -Djetty.port=8900 -DzkHost=xxx.xxx.xxx.1:2181 -jar start.jar

Close all firewall !

SolrCloud election leader

1 Answers