Remove Node

Hi,

how can I safely remove a node?

I started with 3 ES instances then the shards are distributed to those
node.

Now I would like to run ES with only 1 node but while indexing I get
an error * indicating that the shards are missing of the node which I
shutdown via the API. Also when querying a lot of docs are missing and
the facet count is wrong. When starting up again all nodes all is
fine.

How can I handle removing nodes, so that all shards are available to
the remaining nodes?

Regards,
Peter.

[168]: index [newtwindex], type [tweet], id [39839915390799873],
message [UnavailableShardsException[[newtwindex][0] [2] shardIt, [0]
active : Timeout waiting for [1m], request:
org.elasticsearch.action.bulk.BulkShardRequest@16ff83ca]]

Hi

Now I would like to run ES with only 1 node but while indexing I get
an error * indicating that the shards are missing of the node which I
shutdown via the API. Also when querying a lot of docs are missing and
the facet count is wrong. When starting up again all nodes all is
fine.

That should work. Was your cluster health green before you shut down?

Could you gist a recreation?

clint

Could you gist a recreation?

What do you mean should I gist? The logs?

Hi,

I think Clinton refers to Elasticsearch Platform — Find real-time answers at scale | Elastic

http://www.elasticsearch.org/help/If there is a simple way how the issues
can be recreated with curl command then that would be helpful.

Regards,
Lukas

On Tue, Feb 22, 2011 at 2:26 PM, Karussell tableyourtime@googlemail.comwrote:

Could you gist a recreation?

What do you mean should I gist? The logs?

On Tue, Feb 22, 2011 at 10:26 AM, Karussell tableyourtime@googlemail.comwrote:

Could you gist a recreation?

What do you mean should I gist? The logs?

He meant that: Elasticsearch Platform — Find real-time answers at scale | Elastic

Was your cluster health green before you shut down?

Ok, this was the problem. Thanks!