Polling interval related

[madboss89]

I’ve configured NetApp Harvest and installed the Alloy Agent to scrape and push metrics to our internal database server. After adjusting the polling intervals changing the performance related metrics polling from 1 minute to 6 minutes, and the REST/ZAPI collection from 3 minutes to 15 minutes, and the Alloy Agent polling interval to 3 minutes,
some data is missing from the performance related dashboards. Other dashboards (such as capacity, inventory, and health) that I modified to a 15-minute interval are working as expected, but the performance dashboards are still missing certain data.

[madboss89]

If I set restperf/zapiperf collector to 7mins and rest/zapi to 5mins and alloy agent to 5mins, everything works good without any modification on the grafana dashboards,

[rahulguptajss]

@madboss89 You should also increase cache_max_keep in exporter setting. It should be higher than your polling interval of collectors.
https://netapp.github.io/harvest/latest/prometheus-exporter/#parameters

[rahulguptajss]

You can see here related to jitter for custom.yaml
#2856

[madboss89]

Thank you, I mean, I can copy the current yaml files which are using by default.yaml and rename them with custom and update the custom,yaml right.

[rahulguptajss]

custom.yaml is used to customize default.yaml. If your use case is to disable most objects and enable only a few, you can follow the limited.yaml approach described here: https://netapp.github.io/harvest/latest/configure-templates/#collector-templates

More details on custom.yaml can be found here: https://netapp.github.io/harvest/latest/configure-templates/#templates-customization

[rahulguptajss]

You can also use conf_path approach for customization needs. https://netapp.github.io/harvest/latest/configure-templates/#conf-path

[madboss89]

Thank you so much ! have a great day.