Hi,Pablo, I have tried settings replicas to zero and putting it back to 1, it does not work as you say.
And finally, I found, when I turn cluster.routing.allocation.same_shard.host: true to false, the replicas was work well. But on other cluster, this setting does not affect use. And the official document is to describe this set: cluster.routing.allocation.same_shard.hostAllows to perform a check to prevent allocation of multiple instances of the same shard on a single host, based on host name and host address. Defaults to false, meaning that no check is performed by default. This setting only applies if multiple nodes are started on the same machine. Why is this so? (Sorry for my bad English : ) 在 2014年9月11日星期四UTC+8上午11时12分09秒，Pablo Musa写道： > > googled: elasticsearch java 1.6.0_45 shard unassigned > > https://github.com/elasticsearch/elasticsearch/issues/3145 (search for > 1.6) > > https://groups.google.com/forum/#!msg/elasticsearch/MSrKvfgKwy0/Tfk6nhlqYxYJ > > For all that I have researched it points to Java version problem. > You could try some things as settings replicas to 0 and putting it back to > 1 or forcing allocation (do not remember the exact command, but google for > unassigned chards and you will find it), but I do not think that they will > work. > > I really would try installing a new version of Java and running > Elasticsearch using it. > > Regards, > Pablo > > 2014-09-11 0:03 GMT-03:00 Sephen Xu <[email protected] <javascript:>>: > >> Thank you for your reply, the java version on each machine are same -- >> 1.6.0_45, and the elasticsearch version is 1.1.2. >> >> >> >> 在 2014年9月11日星期四UTC+8上午10时48分44秒，[email protected]写道： >> >>> I would check for the Java version on each machine. >>> I had the same problem on a running cluster when adding a node, and >>> unfortunately the last node had Java 1.7.0_65 instead of 1.7.0_55 >>> (recommended version and the version of my other machines). >>> >>> I did not have the time to create a post explaining the whole problem. >>> But, in summary, I ran a default install script using apt-get and, by >>> default, they use the "latest" Java version. >>> >>> One big problem for me is that they do not support versioned jdk >>> installation and I could not find a deb package for 1.7.0_55. Maybe someone >>> here can help with this. >>> >>> The "problematic" command: >>> apt-get install openjdk-7-jre-headless -y >>> >>> Regards, >>> Pablo Musa >>> >>> On Wednesday, September 10, 2014 10:31:24 PM UTC-3, Sephen Xu wrote: >>>> >>>> Hello, >>>> >>>> I startup 4 nodes on 2 machines, and when create index, all replicas >>>> are unassigned. >>>> >>>> { >>>> "cluster_name" : "elasticsearch_log", >>>> "status" : "yellow", >>>> "timed_out" : false, >>>> "number_of_nodes" : 4, >>>> "number_of_data_nodes" : 4, >>>> "active_primary_shards" : 22, >>>> "active_shards" : 22, >>>> "relocating_shards" : 0, >>>> "initializing_shards" : 0, >>>> "unassigned_shards" : 22 >>>> } >>>> >>>> How can I do? >>>> >>> -- >> You received this message because you are subscribed to a topic in the >> Google Groups "elasticsearch" group. >> To unsubscribe from this topic, visit >> https://groups.google.com/d/topic/elasticsearch/kn3UHQgQKJk/unsubscribe. >> To unsubscribe from this group and all its topics, send an email to >> [email protected] <javascript:>. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/elasticsearch/8a7a7fcc-9a93-4ae8-9b26-8cdd9715eb7b%40googlegroups.com >> >> <https://groups.google.com/d/msgid/elasticsearch/8a7a7fcc-9a93-4ae8-9b26-8cdd9715eb7b%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > > -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/elasticsearch/f20b0c5a-00c9-42cb-94b7-5ecb59f83904%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
