[
https://issues.apache.org/jira/browse/HDDS-6560?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Xu Shao Hong updated HDDS-6560:
-------------------------------
Description:
Currently, the existing Grafana dashboards have only a little information about
the cluster. One shows the RPC metrics and the other exhibits some namespace
info of the object store. These cannot support HA situation.
I think we could have a more general and detailed metrics dashboard to support
every environment(docker,k8s, bare metal) and every situation(HA, and non-HA).
Here I collected some useful info and made a dashboard template. (see the
attachment)
What's included:
1. Support more envs. Need to have the *component* label.
2. More dimensions of metrics.
3. Support HA situation.
The drawback.
RPC metrics on the data node could be blank. It is reasonable when selecting
``all``` in the component label option because the panels are repeated by the
component and there is no such RPC metrics on the data node side.
was:
Currently, the existing Grafana dashboards have only a little information about
the cluster. One shows the RPC metrics and the other exhibits some namespace
info of the object store. These cannot support HA situation.
I think we could have a more general and detailed metrics dashboard to support
every environment(docker,k8s, bare metal) and every situation(HA, and non-HA).
Here I collected some useful info and made a dashboard template. (see the
attachment)
What's included:
1. Support more envs. Need to have the *component* label.
2. More dimensions of metrics.
3. Support HA situation.
> add general HA support Grafana dashboard
> ----------------------------------------
>
> Key: HDDS-6560
> URL: https://issues.apache.org/jira/browse/HDDS-6560
> Project: Apache Ozone
> Issue Type: Improvement
> Reporter: Xu Shao Hong
> Priority: Major
> Labels: pull-request-available
> Attachments:
> screencapture-9-186-21-243-d-7noCz-s7z-ozone-overall-metrics-2022-04-07-14_49_21.png
>
>
> Currently, the existing Grafana dashboards have only a little information
> about the cluster. One shows the RPC metrics and the other exhibits some
> namespace info of the object store. These cannot support HA situation.
> I think we could have a more general and detailed metrics dashboard to
> support every environment(docker,k8s, bare metal) and every situation(HA, and
> non-HA).
> Here I collected some useful info and made a dashboard template. (see the
> attachment)
>
> What's included:
> 1. Support more envs. Need to have the *component* label.
> 2. More dimensions of metrics.
> 3. Support HA situation.
>
> The drawback.
> RPC metrics on the data node could be blank. It is reasonable when selecting
> ``all``` in the component label option because the panels are repeated by the
> component and there is no such RPC metrics on the data node side.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
