# How could such a large open source project manage the Pull Requests?

**URL:** <https://discuss.kubernetes.io/t/how-could-such-a-large-open-source-project-manage-the-pull-requests/17571>\
**Category:** General Discussions\
**Tags:** development\
**Created:** [September 29, 2021, 1:55am UTC](https://discuss.kubernetes.io/t/how-could-such-a-large-open-source-project-manage-the-pull-requests/17571 "2021-09-29T01:55:28Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![markdocker07](https://avatars.discourse-cdn.com/v4/letter/m/a88e4f/32.png) [@markdocker07](https://discuss.kubernetes.io/u/markdocker07)\
**Post date:** [September 29, 2021, 1:55am UTC](https://discuss.kubernetes.io/t/how-could-such-a-large-open-source-project-manage-the-pull-requests/17571/1 "2021-09-29T01:55:28Z")

</div>

I was thinking about a very complicated GitHub open source project, how to manage Pull Requests.

check this:

[kubernetes Pull Request](https://github.com/kubernetes/kubernetes/pulls?page=39&q=is%3Apr+is%3Aopen)

there are close to a thousand pull requests there, how could the k8s administrators could validate the function correctness of those Pull Requests?

and how could them be efficiently to verify and merge to master?

How they can control so many updates?

---

<div class="post-metadata">

**Author:** ![mrbobbytables](https://sea2.discourse-cdn.com/flex016/user_avatar/discuss.kubernetes.io/mrbobbytables/32/7_2.png) [@mrbobbytables](https://discuss.kubernetes.io/u/mrbobbytables)\
**Post date:** [September 29, 2021, 1:30pm UTC](https://discuss.kubernetes.io/t/how-could-such-a-large-open-source-project-manage-the-pull-requests/17571/2 "2021-09-29T13:30:36Z")

</div>

TL;DR - lots of automation and delegation 🙂

Responsibilities over areas of the codebase are delegated to Special Interest Groups (SIGs) that oversee a certain areas such as storage, networking etc. You can see various issues and PRs labeled with things like `sig/storage` or `sig/api-machinery`. These are triaged independently by each one of those groups.

Also when it comes to assigning people to review PRs, the project does not use GitHub’s native selection method and instead uses [OWNER files](https://www.kubernetes.dev/docs/guide/owners/) which our CI system ([Prow](https://elder.dev/posts/prow/)) understands and assigns the people according to our own set of rules.

---

<div class="post-metadata">

**Author:** ![thockin](https://sea2.discourse-cdn.com/flex016/user_avatar/discuss.kubernetes.io/thockin/32/2457_2.png) [@thockin](https://discuss.kubernetes.io/u/thockin)\
**Post date:** [September 29, 2021, 7:56pm UTC](https://discuss.kubernetes.io/t/how-could-such-a-large-open-source-project-manage-the-pull-requests/17571/3 "2021-09-29T19:56:23Z")

</div>

We also spend an incredible amount of money and energy on CI and testing. It’s not perfect, but it does pretty well.