# TCP: time wait bucket table overflow

**URL:** <https://www.ruby-forum.com/t/tcp-time-wait-bucket-table-overflow/207356>\
**Category:** NGINX\
**Created:** [May 24, 2011, 5:19pm UTC](https://www.ruby-forum.com/t/tcp-time-wait-bucket-table-overflow/207356 "2011-05-24T17:19:25Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![luislavena](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/l/0ea827/32.png) [@luislavena](https://www.ruby-forum.com/u/luislavena)\
**Post date:** [May 24, 2011, 5:19pm UTC](https://www.ruby-forum.com/t/tcp-time-wait-bucket-table-overflow/207356/1 "2011-05-24T17:19:25Z")

</div>

I am just going to pop-up this question. My nginx was doing about 10%  
cpu load and I was getting:

May 24 15:58:43 yasmin kernel: [983.444403] TCP: time wait bucket  
table overflow

I know I can fine-tune my /proc/sys/net/ipv4/tcp\_max\_tw\_buckets and some  
other stuff. But I am pretty sure somebody here has more experience with  
tcp tuning.

What are good Linux kernel settings for a proxy\_pass / load balancer  
nginx server?

Some URL’s to background articles would be nice.

In my configuration it passes everything to Apache servers as back-ends.

Thanks in advance,

Kind regards,

Jelle de Jong
