Skip to content

Instantly share code, notes, and snippets.

@sgnl

sgnl/README.md

Last active April 9, 2020 09:48
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save sgnl/423e02c13a191351d90690314c2d6464 to your computer and use it in GitHub Desktop.
Save sgnl/423e02c13a191351d90690314c2d6464 to your computer and use it in GitHub Desktop.
Download ZIP
Jenkins Dashboard Process Notes (WIP)
Raw
README.md

Jenkins Dashboard

Set-up and Configuration

I had this guide open in a tab just in case: https://www.influxdata.com/blog/monitoring-jenkins-with-influxdb/

Create Influxdata Cloud Account

After Set-up we need to create/gather these piece of information from our account:

  • token api token
  • organization Free Tier accounts, this value is your email
  • bucket

You can create buckets, find tokens, and set the permissions in the Load Data tab at https://us-west-2-1.aws.cloud2.influxdata.com.

Jenkins Instance

There are three keys

  • url, URL of your the jenkins service you want to monitor
  • username, login credentials
  • password, login credentials

Create Telegraf Configuration File

[[outputs.influxdb_v2]]
## The URLs of the InfluxDB cluster nodes.
##
## Multiple URLs can be specified for a single cluster, only ONE of the
## urls will be written to each interval.
## urls exp: http://127.0.0.1:9999
urls = ["https://us-west-2-1.aws.cloud2.influxdata.com"]

## Token for authentication.
token = "your-api-key-here"

## Organization is the name of the organization you wish to write to; must exist.
organization = "your-organization-on-influx-cloud-account"

## Destination bucket to write into.
bucket = "a-bucket-associated-to-your-influx-cloud-account"

Install Telegraf

Followed this Guide for OS X

(note: Installation was a breeze)

Creating a Dashboard and cells

References

The difficulty at first was trying to build the cells without looking at the data "view raw data" toggle feature. It was literally like building something blind. I took a step back and explored how the data in the raw view looked like.

The documentation for functions, aggregators, etc in the "script editor" was tremendously helpful. I feel comfortable with the "Script Editor" after some tinkering!

🚫 ✅ Figuring out how to add variables was a challenge at first because I tried to do a "Query Variable" with the hopes that if I return a table with unique "_value" per row that I would be a variable that has a "drop-down" selection like a "CSV variable". I found out after that the query variable does not work that way. So I switch it to a CSV variable and all worked great.

✅ I like the power of the variables and how it's used on the dashboard.

🚫 I wish variables could be used as values in the "customize" feature of cells. I'd use a query variable that would query for a count of unique values in a list (job names), and use this value and the upper range for a chart like the "Gauge"'s maximum threshold.

🤔 I just wrote a script with groups twice... I think I should use join() instead of the last group()? I have removed the script but I'm glad to have learn ahout join()

🚫 "Need Help?" I often trigger this feature due to invisible box-sizing

🚫 🤔 I created a histogram but I think i should use the histrogram() function in it

😲 I didn't realize that the autorefresh value at the dashboard level was a toggle for the value inside of a cell! After learning this I renamed all my cells by removing any units of time (e.g. "failures in past hour" was renamed to "failures").

variables defined

  • jenkinsHostname type: CSV variable
  • jenkinsJobNames type: CSV variable
  • jenkinsBucket type: CSV variable

Exporting the dashboard

After a few hiccups the exporting was simple. I was able to import the dashboard into my cloud account and everything worked fine once I updated the variables. Below are some notes for the exporting process.

At first I used the export feature through the UI. I noticed that it only exported as a JSON. I noticed that the examples in GitHub were in .yml so I dived into the documentation.

Installed Influxdb locally.

Imported the dashboard from the cloud onto my local computer via JSON.

🚫 encountered a bug where I was unable to export the dashboard due to the charts using a "smoothing" alogorithmn would cause an error. Changing these charts to "linear" got me around this blocker

🚫 Encountered environment variable issue with CLI tool influx and setting a file at ~/.influxdbv2/credentials with export INFLUX_TOKEN=TOKEN_HERE. The file is removed (not specified in documentation that thing would happen) and the influxdb log shows an error: 2020-04-09T02:12:09.840928Z info Unauthorized {"log_id": "0M37Sdzl000", "error": "authorization not found"} . Generated a couple tokens before I stopped and tried the flag.

✅Running the CLI and setting the token with -t flag works as expected

Documentation for INFLUX_TOKEN difference between these two pages, just something I noticed.

  1. https://v2.docs.influxdata.com/v2.0/reference/cli/influx/
  2. https://v2.docs.influxdata.com/v2.0/get-started/

I was able to export and use the -f flag to create a .yml file

Updated the file with [[inputs.jenkins]] configs

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment