From: "M. Edward (Ed) Borasky" <znmeb@...>
Date: 2007-11-12T09:11:12+09:00
Subject: Re: Ruby/Fastcgi going into uninterruptible after random periods                     of      time

nate wrote:
> nate wrote:
>> M. Edward (Ed) Borasky wrote:
>>
>>
>>> Since you're not swapping, any free memory you have above the minimum
>>> set by the kernel *should* be going into "buffers" and "cached" to
>>> reduce the amount of I/O you need to do.
>> Yep I understand all that, to illustrate, one of the server's memory
>> usage -
>>
> 
> Rolled 3 of the 4 servers back to the older OS last night, this
> morning the 4th system freaked out again and it wasn't caught straight
> away, when I logged in I noticed another ruby process was in "D",
> and this one wasn't being served by apache, but rather it was a
> ruby cron job, so there must be some bug in ruby or interaction between
> this version of ruby and the OS. Going to upgrade to a newer rev
> of ruby 1.8.5(-114) in test environment today to hopefully roll to the
> prod system tomorrow to see if that helps.
> 
> nate
> 
> 
> 

Interestingly enough, "older" is a misnomer, since the Fedora OS was 
something like 2.6.17-xx and the CentOS/RHEL 4.5 was 2.6.9-xx. :) But 
Red Hat tweaks their kernels mercilessly, and considering some of the 
talented folks they have and the close ties they have with the server 
hardware vendors, an RHEL kernel or its CentOS clone is usually better 
than one from Fedora.

Does anyone here know what version of Ruby and what OS/DBMS/webserver 
the BaseCamp people run in production? BaseCamp, after all, is the 
mother of all Rails applications. :)