Elasticsearch doesn't recover data from NAS storage volume

[tmporary]

Hi,
Our elasticsearch instance starts up fine and in the logs I can see that it found the mounted storage but it doesn't recover any indices info into the cluster.

[2020-05-28T06:51:41,853][INFO ][o.e.e.NodeEnvironment] [es_master] using [1] data paths, mounts [[/media/storage (NAME_OF_NAS)]], net usable_space [10.3gb], net total_space [10.4gb], types [nfs]

[2020-05-28T06:51:54,950][INFO ][o.e.g.GatewayService ] [es_master] recovered [0] indices into cluster_state

We have almost everything set to defaults when it comes to config the only config properties we set outservles are:

cluster.name: cluster_name

node.name: esmaster
node.master: true
node.data: false

discovery.seed_hosts: address to master

cluster.initial_master_nodes: esmaster

network.host: 0.0.0.0

path.data: /media/storage/data

node.max_local_storage_nodes: 3

The same properties are set for our data node with the difference that it has assigned data role instead of master

Am I missing something or doing something wrong? Maybe there's some specific way how to set up Elastic in order for it to be able to retrieve data from network mounted nfs shared drive

[Christian_Dahlqvist]

I would recommend you have a look at this post.

[DavidTurner]

I suspect this may be your problem:

@tmporary

node.max_local_storage_nodes: 3

Don't use this, it's a little dangerous and it's deprecated. From the docs:

WARNING: Never run different node types (i.e. master, data) from the same data directory. This can lead to unexpected data loss.

Instead, give each node its own specific data.path underneath /media/storage/data

Also as per the post Christian linked, NFS can theoretically be configured to work correctly but it's easy to get it wrong without realising, and often doesn't perform too well.