Sphinx Search init script for Centos 5.1

Sphinx search is pretty new, and as a result I was unable to find a nice convenient package for it for CentOS 5.1. This is problematic since there is no init script included with the source tarball, and the issue of updating the index is the sysadmin and developer’s problem, and cannot be configured to simply update the index when the data changes.
Continue reading “Sphinx Search init script for Centos 5.1”

Making SELinux allow a nonstandard MySQL port number on Centos 5.1

SELinux is a recently added security system that’s installed by default with CentOS 5.1 (and Red Hat Enterprise Linux 5, and others). Since it’s newer than the classic “Discretionary Access Control” Unix security model, it’s not nearly as well documented, and unfamiliar to many. I had never even heard of it until this week.

After a lot of reading about it, and debating disabling it entirely, I figured out how to do some minor SELinux customization to fit my needs for a MySQL database server. Hopefully this will help folks who are in a similar situation.

Continue reading “Making SELinux allow a nonstandard MySQL port number on Centos 5.1”

CentOS 5.1 Minimal VPS Install Guide

I’m working on a project that is deploying on CentOS 5.1, and I found it not entirely obvious how to install a really stripped down server, as a starting point for a lean and mean, hardened production server. Since I’m doing work on this at home on VMWare, and it’s being deployed on a VPS initially (and probably will remain virtualized for ease of management as it scales up), this guide is specifically aimed at this kind of configuration.

Continue reading “CentOS 5.1 Minimal VPS Install Guide”

Document Databases – New Kids on an Old Block

There’s a new crop of databases that has appeared lately, under the rubric of “document databases”, and there’s quite a lot of enthusiasm for them given that they tend to be slow and very feature-poor compared to the SQL RDBMSs that are the typical persistence mechanism for web applications. What’s mainly appealing about them is that they are easy to use, and theoretically quite scalable, compared to the traditional “one big SQL database server” approach.
Continue reading “Document Databases – New Kids on an Old Block”