Elasticsearch使用java讀取數據報錯NoNodeAvailableException: None of the configured nodes are available: [127.0.0.1:9300]

對於這個問題，大部分人出現在這個地方：

Client client = new TransportClient(settings).addTransportAddress(new InetSocketTransportAddress("172.16.2.13", 9300));​

問題在於前面初始化settings時給cluster設置了個新的名字，如:Settings settings = ImmutableSettings.settingsBuilder().put("cluster.name", "tonsonmiao").build();

因為如果設置clustername后，容器會在添加transportaddress時，從集群名為tonsonmiao里查找是否有要設置的這個IP和端口，此時肯定找不到，所以會報這個錯。

但是我今天又遇到這個問題，而在此之前一切正常，也就是說並不會因為代碼的錯誤導致這個問題出現，這幾天修改了下網關，可以訪問這台服務器，於是我覺得可能是網絡的變動，導致es出了問題，進入服務器查看：

[root@es elasticsearch-1.4.2]# ps -eaf | grep java
root      8782  8752  0 05:53 pts/1    00:00:00 grep java
root     27251     1  1  2015 ?        1-10:15:49 /usr/bin/java -Xms256m -Xmx1g -Xss256k -Djava.awt.headless=true -XX:+UseParNewGC -XX:+UseConcMarkSweepGC -XX:CMSInitiatingOccupancyFraction=75 -XX:+UseCMSInitiatingOccupancyOnly -XX:+HeapDumpOnOutOfMemoryError -XX:+DisableExplicitGC -Dfile.encoding=UTF-8 -Delasticsearch -Des.path.home=/opt/elasticsearch-1.4.2 -cp :/opt/elasticsearch-1.4.2/lib/elasticsearch-1.4.2.jar:/opt/elasticsearch-1.4.2/lib/*:/opt/elasticsearch-1.4.2/lib/sigar/* org.elasticsearch.bootstrap.Elasticsearch


[root@es elasticsearch-1.4.2]# netstat -anp | grep 9300
tcp        0      0 10.18.7.97:9300             10.17.3.96:51633            SYN_RECV    -                   
tcp        0      0 10.18.7.97:9300             10.17.3.96:51635            SYN_RECV    -                   
tcp        0      0 :::9300                     :::*                        LISTEN      27251/java          
tcp        0      0 ::ffff:10.18.7.97:41035     ::ffff:10.18.7.97:9300      ESTABLISHED 27251/java          
tcp        0      0 ::ffff:10.18.7.97:9300      ::ffff:10.18.7.97:41030     ESTABLISHED 27251/java          
tcp        0      0 ::ffff:10.18.7.97:41033     ::ffff:10.18.7.97:9300      ESTABLISHED 27251/java          
tcp        0      0 ::ffff:10.18.7.97:41037     ::ffff:10.18.7.97:9300      ESTABLISHED 27251/java          
tcp        0      0 ::ffff:10.18.7.97:41036     ::ffff:10.18.7.97:9300      ESTABLISHED 27251/java          
tcp        0      0 ::ffff:10.18.7.97:41026     ::ffff:10.18.7.97:9300      ESTABLISHED 27251/java          
tcp        0      0 ::ffff:10.18.7.97:9300      ::ffff:10.18.7.97:41025     ESTABLISHED 27251/java          

 

可以看到es監聽的ip地址發生了變化，對於多網卡的情況下，es默認會綁定其中任意一張網卡，如果es中間出現問題自動修復，那么會隨機修改綁定網卡，導致節點被t出集群，曾經我在某銀行系統，就遇到這個問題，某幾個節點不定時的被t出集群。

對於這個問題的解決，需要修改es配置：

# Set the bind address specifically (IPv4 or IPv6):
#
network.bind_host: 10.18.7.97

# Set the address other nodes will use to communicate with this node. If not
# set, it is automatically derived. It must point to an actual IP address.
#
network.publish_host: 10.18.7.97

# Set both 'bind_host' and 'publish_host':
#
network.host: 10.18.7.97

強制指明ip地址即可