17 messages in ru.sysoev.nginxRe: Nginx + fair load balancer patch ...
FromSent OnAttachments
Alexander StauboFeb 29, 2008 1:39 am 
Igor SysoevFeb 29, 2008 2:45 am 
Joshua SierlesFeb 29, 2008 2:51 am 
Igor SysoevFeb 29, 2008 3:25 am 
Grzegorz NosekFeb 29, 2008 3:53 am 
Alexander StauboMar 13, 2008 4:29 pm 
Grzegorz NosekMar 14, 2008 5:50 am 
Alexander StauboMar 28, 2008 7:39 am 
Grzegorz NosekMar 28, 2008 8:11 am 
Alexander StauboMar 28, 2008 8:27 am 
Grzegorz NosekMar 28, 2008 8:39 am 
Alexander StauboMar 28, 2008 8:53 am 
Alexander StauboApr 16, 2008 2:16 pm 
Grzegorz NosekApr 17, 2008 7:16 am 
Grzegorz NosekApr 18, 2008 6:02 am 
Andy VerprauskusJun 5, 2008 11:33 am 
Grzegorz NosekJun 13, 2008 12:01 am 
Actions with this message:
Paste this link in email or IM:
Paste this link in email or IM:
Atom feed for this thread
Paste this URL into your reader:
Subject:Re: Nginx + fair load balancer patch loopingActions...
From:Igor Sysoev (is-G@public.gmane.org)
Date:Feb 29, 2008 3:25:53 am
List:ru.sysoev.nginx

On Fri, Feb 29, 2008 at 12:51:33PM +0200, Joshua Sierles wrote:

On Feb 29, 2008, at 11:39 AM, Alexander Staubo wrote:

An Nginx instance suddenly started spewing the following to its error log at a rate of about 1GB/minute, and using a bit more CPU than usual:

2008/02/29 10:33:47 [error] 16875#0: *126309461 upstream prematurely closed connection while reading response header from upstream [...]

Aside from the excessive logging, everything else seemed normal. Our next_upstream setting is:

proxy_next_upstream error invalid_header;

Restarting Nginx fixed the problem.

Could this be the fair load balancer going hairwire?

Alexander.

At Engine Yard we are seeing this happen as well, usually when a backend goes down or a 'clock skew' message is logged. It gets logged hundreds of thousands of times for each request.

Oh, I have missed "fair load balancer patch". No, I'm not able to debug the bug.

-- Igor Sysoev http://sysoev.ru/en/