Yellow status

[Box293]

I have a Nagios Log Server running with one instance (2015R1.4).

When I go to Administration > Cluster Status it says Yellow.

What does Yellow mean? It doesn't clearly indicate what is causing this Yellow status.

Also, Underneath it says I have 986 Unassigned Shards. What does this mean?

Selection_081.png

[ssax]

It will stay yellow until you add a second cluster node, then it will turn green.

It is recommended to have more than one node, you can do it with one but you really should have two at a minimum just for failover capability.

[tmcdonald]

Also this:

http://chrissimpson.co.uk/elasticsearch ... ained.html

[Box293]

Yes that seems to have fixed it after adding another instance.

I did find in this document:

http://assets.nagios.com/downloads/nagi ... usters.pdf

Where it explains the log server health and colours:

Cluster Health Status,
This will be an indication of the current status of your cluster. It primarily indicates the current allocation status
of the shards that make up your cluster:
• Green:
Your Cluster is healthy and all shards have been allocated.
• Yellow:
Your Cluster has unassigned shards or had not complete allocating shards after a system change.
• Red:
Your Cluster timed out or isn't responding

I think it would be benefical if this appeared when the mouse hovered over the cluster status.

There is however, no clear indication that "Your Cluster has unassigned shards or had not complete allocating shards after a system change" is directly related to only having a one instance deployment ... this was only realised by coming here and asking the question in the forums.

[eloyd]

I've never liked the assumption that NLS was "supposed to be" run in a cluster environment. Our production NLS server runs on a VM that sits in the same virtual place as all our other VMs and if one fails, they pretty much all fail. A cluster makes no sense for us.

Makes sense to me that the "yellow" be green unless there's more than one node in the cluster.