#Configuration of Grafana with netapp harvest

1 messages · Page 1 of 1 (latest)

rotund osprey
#

Hello all, I am looking for some information on how to conufigure netapp harvest to send the metrics towards grafana.
At this point in time I am at the step of all the pollers are working .
I followed the steps described in https://netapp.github.io/harvest/22.11/

prime python
rotund osprey
#

Thanks, as far as i can see prometheus is configured

#

root@lx10970> /opt/harvest/bin/harvest status
Datacenter | Poller | PID | PromPort | Status
-------------+-----------+---------+----------+--------------
local | unix | 2702065 | 12990 | running
remote | a220mrs01 | 2702066 | 13001 | running
remote | a220ext01 | 2702081 | 13002 | running
remote | a220vil01 | 2702077 | 13003 | running
remote | a250ext01 | 2702091 | 13004 | running
remote | a250vil01 | 2702100 | 13005 | running
remote | a250mrs01 | 2702105 | 13006 | running
remote | a250mrs02 | 2702128 | 13007 | running
remote | a300vil01 | | | not running
remote | a300ext01 | | | not running
remote | a300mrs01 | | | not running
remote | nasvit22 | | | not running
remote | nasvit02 | | | not running

#

Exporters:
prometheus:
exporter: Prometheus
local_http_addr: x.x.x.x
port: 12990

prometheus1:
exporter: Prometheus
port_range: 13000-14000

unix:
datacenter: local
addr: localhost
collectors:
- Unix
exporters:
- prometheus

xxxxxx:
datacenter: remote
addr: x.x.x.x
auth_style: basic_auth
use_insecure_tls: true
username: xxxxxx
password: xxxxxxxx
exporters:
- prometheus1

prime python
#

Hi @rotund osprey , This is the Harvest metrics which are exposed at different ports where Prometheus can pull data. For example, http://{{HARVEST_IP}}:13001/metrics would gives raw metrics of a220mrs01 poller. Prometheus would be running on 9090 port where it's deployed. You could get more detail in their documentation hyperlink in the screenshot of how Prometheus is installed in https://netapp.github.io/harvest/22.11/prometheus-exporter/

image.png
rotund osprey
#

thanks I ill have a read through . from my understanding: overall prometheus is configured but I do not know how to integrate the whole solution with Grafana

normal sky
#

@rotund osprey we can double check Prometheus to make sure it's collecting data from the Harvest pollers you pasted above. If you go to your Prometheus instance and click Status then Targets you should see each of your Harvest endpoints there

image.png
rotund osprey
#

@normal sky apologies for the late reply . I checked the metrics by running http://x.x.x.x:12990/metrics and I do see some data. Unfortunately I do not have a user interface for the prometheus DB

normal sky
#

:12990 is the port Harvest publishes metrics on - if you're seeing data there it means the Harvest poller is up and running. Prometheus then needs to be told to scrape that target. Do you have Prometheus installed? By default, Prometheus will listen on port 9090

rotund osprey
#

@normal sky thanks all for your comments (it helps me to better understand the product) . yes I have prometheus installed on the same server as netapp harvest and it is scrapping its own local server. when I am trying to add prometheus as a datasource in grafana (different server) I get two types of error : 1. when selecting access type "browser" I get the error "Unknown error during query transaction. Please check JS console logs." 2. when i select access "server 'default'" I get the error "HTTP Error Bad Gateway". in both case the url I use it "http://x.x.x.x:12990"

normal sky
#

from your Grafana machine, are you able to curl the Prometheus endpoint? Maybe there is a firewall in the way? Something like this should work from the Grafana machine http://x.x.x.x:12990/api/v1/status/config

solemn smelt
normal sky
solemn smelt
#

+1