# Post mortem of a HackerNews Launch & references to mistakes made with Nginx config

**URL:** <https://www.ruby-forum.com/t/post-mortem-of-a-hackernews-launch-references-to-mistakes-made-with-nginx-config/224901>\
**Category:** NGINX\
**Created:** [November 30, 2012, 4:32pm UTC](https://www.ruby-forum.com/t/post-mortem-of-a-hackernews-launch-references-to-mistakes-made-with-nginx-config/224901 "2012-11-30T16:32:41Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![aris](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/a/258eb7/32.png) [@aris](https://www.ruby-forum.com/u/aris)\
**Post date:** [November 30, 2012, 4:32pm UTC](https://www.ruby-forum.com/t/post-mortem-of-a-hackernews-launch-references-to-mistakes-made-with-nginx-config/224901/1 "2012-11-30T16:32:41Z")

</div>

Hello family,

I wrote about some of the mistakes I made while doing a HN launch for my  
new startup Cucumbertown [http://www.cucumbertown.com/](http://www.cucumbertown.com/)

[http://www.gigpeppers.com/post-mortem-of-a-failed-hackernews-launch/](http://www.gigpeppers.com/post-mortem-of-a-failed-hackernews-launch/)

It has references to Nginx and how making a simple mistake in config  
cost  
me a server crash that lasted for 20-30 minutes.

\*Hope this will benefit some of the new Nginx users who are trying to  
scale.  
\*

On the positive side, if not for Nginx we wouldn’t have lasted 10  
minutes  
of peak traffic at 1000 users/sec+. Besides that, this blog  
[gigpeppers.com](http://gigpeppers.com) is  
hosted on a 128MB single core [prgrmr.com](http://prgrmr.com) machine and has been serving  
peak HackerNews  
front page traffic for almost 17 hours now  
[http://hnrankings.info/4847665/](http://hnrankings.info/4847665/),  
thanks to Nginx.

> **[Post mortem of a failed HackerNews launch](https://news.ycombinator.com/item?id=4847665)**
>
> 150 points —
> 101 comments —
> gingerjoos —
> 1:56 PM - 29 Nov 2012

Thank you Igor, Maxim, Antonio, Velentin and all other for this software  
marvel.

- Cherian

---

<div class="post-metadata">

**Author:** ![Cherian\_Thomas](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/c/dec6dc/32.png) [@Cherian\_Thomas](https://www.ruby-forum.com/u/Cherian_Thomas)\
**Post date:** [December 1, 2012, 9:06am UTC](https://www.ruby-forum.com/t/post-mortem-of-a-hackernews-launch-references-to-mistakes-made-with-nginx-config/224901/2 "2012-12-01T09:06:58Z")

</div>

Hi,

On Fri, Nov 30, 2012 at 11:32 PM, Cherian Thomas  
[removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)wrote:

> \*Hope this will benefit some of the new Nginx users who are trying to  
> Thank you Igor, Maxim, Antonio, Velentin and all other for this software  
> marvel.

Nice post!

I think you can use the sysguard module (  
[ngx\_http\_sysguard\_module - The Tengine Web Server](http://tengine.taobao.org/document/http_sysguard.html) ) comes from  
Tengine.  
It can protect your server when load or memory use goes too high. We  
have  
taken it from Tengine as a separate Nginx module (  
[https://github.com/taobao/nginx-http-sysguard](https://github.com/taobao/nginx-http-sysguard) ). Hope it helps.

BTW, the sysguard module played a very important role in the 24-hour  
rush  
when Taobao conducted $3.06 billion USD of business (  
[http://westiseast.co.uk/blog/taobao-sales-19-billion-bonanza/](http://westiseast.co.uk/blog/taobao-sales-19-billion-bonanza/) ).

Regards,