gl-pipeline-analyzer

📈 A CLI tool to analyze GitLab Pipelines

Highlights 🏆

All right from the command line you can

• View the slowest jobs in your pipelines
• View the jobs that spent the longest time in pending
• View stats from a certain pipeline

More features coming soon...

Installation

npm install gl-pipeline-analyzer

Usage

View all commands

gl --help

Register GitLab personal access token

You will need to run this command and register-project before using the API

This is stored locally on your machine with the package conf

gl register-token <access-token>

Register project

You will need to run this command and register-token before using the API

Use this command to persist a certain project

gl regsiter-project <your-project-path>

Return the latest pipelines iid

So you don't have to search for the latest pipeline iid

gl latest-iid

View slow jobs

Returns the 10 slowest jobs in the pipeline

gl slow-jobs <pipeline-iid>

Example output

⏳ Job slow_job_one took 195 seconds to complete
⏳ Job slow_job_three took 166 seconds to complete
⏳ Job slow_job_two took 133 seconds to complete
⏳ Job fast_job_three took 57 seconds to complete
⏳ Job fast_job_two took 33 seconds to complete
⏳ Job fast_job_one took 21 seconds to complete
⏳ Job allow_failure_test_job took 14 seconds to complete
⏳ Job coverage_job took 12 seconds to complete

View queued jobs

Returns 10 jobs that spent the longest time queued

gl queued-jobs <pipeline-iid>

Example output

⏳ Job slow_job_two waited 0.4744 seconds in the pending state
⏳ Job fast_job_one waited 0.467092 seconds in the pending state
⏳ Job slow_job_three waited 0.455937 seconds in the pending state
⏳ Job fast_job_three waited 0.396569 seconds in the pending state
⏳ Job fast_job_two waited 0.388403 seconds in the pending state
⏳ Job coverage_job waited 0.359404 seconds in the pending state
⏳ Job slow_job_one waited 0.139841 seconds in the pending state
⏳ Job allow_failure_test_job waited 0.136156 seconds in the pending state

View usueful pipeline stats

gl pipeline-details <pipeline-iid>

Example output

Pipeline iid: 1
Ran for 209 seconds
Was queued for 1 seconds before starting
Has a status of SUCCESS