How to fix cluster health yellow with Elastic Search

21
votes

I have setup on server, with MongoDb and ElasticSearch. Using https://github.com/richardwilly98/elasticsearch-river-mongodb I have connected ElasticSearch and MongoDb together.

I create a new index using:

curl -XPUT 'http://127.0.0.1:9200/_river/mongodb/_meta' -d '{ 
        "type": "mongodb", 
        "mongodb": { 
        "db": "my_database", 
        "collection": "my_collection"
    }, 
        "index": {
        "name": "mainindex", 
        "type": "string",
        "bulk": {
            "concurrent_requests": 1
        }
    }
}'

Once the command is executed and I go to http://x.x.x.x:9200/_plugin/head/ I see the message: cluster health: yellow (1, 6).

cluster health: yellow (1, 6)

enter image description here

3
elasticsearch

3 Answers

22
votes

A cluster is being configured by having the same cluster name inside the elastic search config

The default elasticsearch.yml you are probably using has these settings in the Beginning like this:

################################### Cluster ###################################

# Cluster name identifies your cluster for auto-discovery. If you're running
# multiple clusters on the same network, make sure you're using unique names.
#
# cluster.name: elasticsearch


#################################### Node #####################################

# Node names are generated dynamically on startup, so you're relieved
# from configuring them manually. You can tie this node to a specific name:
#
# node.name: "Franz Kafka"

here you would need to configure a unique

cluster.name: "MainCluster"

and for each machine and/or instance a different unique

node.name: "LocalMachine1"

you now need to copy this elasticsearch.yml to another machine (in the same Network), or to the same place as e.g.elasticsearch_2.yml edit it for:

node.name: "LocalMachine2"

and your cluster is ready to go

if not configured elastiscsearch will use a random Marvel Character (of 3000 according the the documentation), so not to change the node.name should be ok also

For having two nodes running on the same machine, you must make a configuration e.g. elasticsearch_2.yml copy, with above changes. Also you must have copies of the data and log path e.g. (homebrew specific paths:)

cp -r /usr/local/var/elasticsearch /usr/local/var/elasticsearch_2
cp -r /usr/local/var/log/elasticsearch /usr/local/var/log/elasticsearch_2

might look like

#################################### Paths ####################################

# Path to directory containing configuration (this file and logging.yml):
#
# path.conf: /path/to/conf

# Path to directory where to store index data allocated for this node.
#
path.data: /usr/local/var/elasticsearch_2/
#
# Can optionally include more than one location, causing data to be striped across
# the locations (a la RAID 0) on a file level, favouring locations with most free
# space on creation. For example:
#
# path.data: /path/to/data1,/path/to/data2

# Path to temporary files:
#
# path.work: /path/to/work

# Path to log files:
#
path.logs: /usr/local/var/log/elasticsearch_2/

make sure you do not have running elasicsearch on localhost loopback device

127.0.0.1

just comment it out in case it is not (homebrew does patch ist this way)

############################## Network And HTTP ###############################

# Elasticsearch, by default, binds itself to the 0.0.0.0 address, and listens
# on port [9200-9300] for HTTP traffic and on port [9300-9400] for node-to-node
# communication. (the range means that if the port is busy, it will automatically
# try the next port).

# Set the bind address specifically (IPv4 or IPv6):
#
# network.bind_host: 192.168.0.1

# 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: 192.168.0.1

# Set both 'bind_host' and 'publish_host':
#
# network.host: 127.0.0.1

now you can start elastic search like this:

bin/elasticsearch -D es.config=/usr/local/Cellar/elasticsearch/1.0.0.RC1/config/elasticsearch.yml

for the first node and master (because started first)

and then 

bin/elasticsearch -D es.config=/usr/local/Cellar/elasticsearch/1.0.0.RC1/config/elasticsearch_2.yml

Now you should have got 2 Nodes running

12
votes

Looks like you don't have a node for replica shards to go to. You can lower your replica count to 0 or add a second node to your cluster so that primary and replica shards can be safely placed on different nodes.

The default configuration in elasticsearch.yml is probably this:

index.number_of_shards: 5  
index.number_of_replicas: 1

The idea being that if your node crashes, another node in your cluster will have a copy of the shard. Since you only have one node, the cluster has no idea where to put the replicas and thus is in a yellow state.

4
votes

I am posting an answer since I tumbled upon this question while looking for solution of similar issue.

I have a dev environment where I need only 1 Elasticsearch node so there is no need for replicas for all my indices.

To solve cluster health: yellow issue, I just made index.number_of_replicas: 1

PUT /_settings
{
  "index" : {
        "number_of_replicas" : 0
    }
}