Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/imrajdas/google-summer-of-code

GSoC CNCF Summarization
https://github.com/imrajdas/google-summer-of-code

cncf google-summer-of-code gsoc prometheus

Last synced: about 2 months ago
JSON representation

GSoC CNCF Summarization

Awesome Lists containing this project

README 

        
	

## GSoC'20 Report- CNCF(Prometheus)



### Introduction

Hello everyone, I am here to present my GSoC 2020 report under CNCF. My contributions/work can be viewed at  prometheus/test-infra and prombench.prometheus.io. Also I will be continuing my contributions after the GSoC period

	

> I would like to give a huge thanks to my mentor, who continuously helped me to complete this awesome journey. 



### Links:

* My Evaluations: 

   - First Evaluation

   - Second Evaluation

   - Final Evaluation

   

* GSoC Project URL 


  https://summerofcode.withgoogle.com/archive/2020/projects/4546170073907200/



* CNCF Blog Post 


  https://www.cncf.io/blog/2020/09/17/16-cncf-interns-graduate-from-summer-of-code-gsoc-2020/



* Certificate 


  Completion_2020_0234.pdf



* Completion Letter 


   GSoC 2020 Certificate GSoC20200234.pdf

 

### What did I solve?

  1. Develop prometheus benchmark infrastructure with KIND 

  

	 Description: Currently, prombench is deployed on GKE which takes a lot of time and resources. To overcome this we got a solution to use KIND for deployment of prombench on local/virtual machine, which is very fast as compared to GKE and useful to run it locally.

	    * Issue(s): prometheus/test-infra/issue#333 

        * PR(s): prometheus/test-infra/pull#390



  2. Refactor prombench infrastructure and manifest deployable for KIND and EKS.

        

     Description: Due to the tightly coupled of prombench with GKE, we have to refactor both infra and manifests to make it more generic and deployable for both EKS and KIND.

	    * Issue(s): prometheus/test-infra/issue#371

	    * PR(s): 

	      * prometheus/test-infra/pull#412

	      * prometheus/test-infra/pull#406

	      * prometheus/test-infra/pull#391

	      * prometheus/test-infra/pull#371

	      * prometheus/test-infra/pull#369

	



  3. Fix Node Exporter to show the metrics related to SSD

  

     Description: Node Exporter is not showing metrics related to SSD. To fix this we need to upgrade node exporter and add a new flag called “--path.rootfs”

     * Issue(s): 

        * prometheus/test-infra/issue#159

        * prometheus/test-infra/issue#328



	 * PR(s): prometheus/test-infra/pull#408



  4. Add descriptions to the Grafana dashboard regarding prombench

  

	 Description: Current Grafana dashboard is not having the descriptions of prombench, which will be difficult for a new user to understand. 

	 * Issue(s): 

	    * prometheus/test-infra/issue#328

	    * prometheus/test-infra/issue#305

	 * Doc(s): Descriptions for Prombench Graphana Dashboard

	 * PR(s): prometheus/test-infra/pull#428

	

  5. Increase Retention period of loki to 90 days



     Note: This task is not a priority, so I will complete it after the gsoc period. BTW, a PR has been raised to solve this issue. 

     * Issue(s): 

        * prometheus/test-infra/issue#322

        * prometheus/test-infra/issue#328

     * PR(s): prometheus/test-infra/pull#423



## Extra fixes

1. PR: prometheus/test-infra/pull#432

2. PR: prometheus/test-infra/pull#433



### What did I learn?

1. First and most important- Architecture of Prometheus and infrastructure of prombench

2. Good maintenance of code repository

3. KIND(Kubernetes In Docker) architecture and usage of its go pkg.

4. Have patience for PR reviews :smiley:



---