No monitoring data found

[Chris_Gatihi]

Hi,

I'm running on version 7.11.1 and I'm getting a "No monitoring data found" when I navigate to stack monitoring in Kibana. Kibana can access my Elasticsearch cluster just fine when going through the dev tools console or looking at the indices via the stack management UI.

But for some reason the monitoring fails and if you look at the console in my screenshot it's failing to get anything back when it's making an AJAX call to for cluster and node info.

I also noticed that there are no monitoring indices being created (GET /_cat/indices/.monitoring*?v returns nothing).

Any ideas what might be happening and how I might address this?

Thanks,
Chris

[warkolm (Mark Walkom)]

Did you make any configuration changes for Monitoring?

[Chris_Gatihi]

Do you mean changes to kibana.yml?

[warkolm (Mark Walkom)]

Or to the Elasticsearch config, yep.

[stephenb (Stephen Brown)]

Hi @Chris_Gatihi you need to explicitly enable monitoring of an elasticsearch cluster see the overview Here and The specifics Here

[Chris_Gatihi]

No, no changes to either elasticsearch.yml or kibana.yml other than what was necessary to upgrade due to breaking changes.

What's even more strange is that I have the same configuration in multiple regions monitoring works in the other regions but not in this one for some reason. And it used to work in this particular reason but for some reason stopped working recently.

[Chris_Gatihi]

Monitoring has already been explicitly enabled for a while. I upgraded to the latest version recently and it wasn't working for a while on 6.8, neither is it now working on 7.11. I'm pretty sure it worked when I first got on 6.8. And the strangest part is that I have other regions and this same configuration works in other regions, just not this one.

[stephenb (Stephen Brown)]

The old / legacy self monitoring (click and enable) has been deprecated that may be why it is no longer working for you in 7.11.

I suggest you look the docs that were referenced above.

In the end the new way to do it is with metricbeat

The legacy method should still work but I suspect the settings have changed since 6.8

[stephenb (Stephen Brown)]

I just set up a fresh 7.11.1 Cluster with Elasticsearch
No Changes to the .ymls

I setup the legacy default monitoring and it still works.

GET _cat/indices/.mon*/?v

health status index                           uuid                   pri rep docs.count docs.deleted store.size pri.store.size
green  open   .monitoring-kibana-7-2021.04.27 jNEqquJGQKu69G9LNPf9Sg   1   0         20            0    159.6kb        159.6kb
green  open   .monitoring-es-7-2021.04.27     LvxAIGA7SvOn8nEoXyeiLQ   1   0        218          160    238.3kb        238.3kb

Check your cluster settings to see

GET _cluster/settings

And take look at

xpack.monitoring.collection.enabled

If doing legacy on localhost should look something like

{
  "persistent" : {
    "xpack" : {
      "monitoring" : {
        "collection" : {
          "enabled" : "true"
        }
      }
    }
  },
  "transient" : { }
}