Recent
Articles |
Area 51: Oracle Readies 51 Fixes For Database
Oracle will ship 51 fixes for its products in their October release of quarterly patches for their products. Patch Tuesday arrived for Oracle administrators today, and it's a doozy. Oracle releases patches on a quarterly...
Linux to Save Power
Via Stephen Shankland at CNet - Very cool news about an effort to run Linux more efficiently on servers, laptops & mobile devices. The effort was kicked off by Intel, but is open to all. According to Stephen's report...
Embedded Virtualization
VMware and XenSource are moving to hardware near you: Virtualization: A feature of the hardware, not the OS?
It's not clear to me how much cost this will add...
Questions For Interviewing Your Next Security Engineer
This is going to be completely biased in favor of a security engineer that is not tightly wound around military grade information security for everyone at all times. Most interview questions are really about...
Truthy Statements About Access To Source Code
It was a long weekend here in the Toronto and I met up with two friends from university. They told me of a "guaranteed" system for winning at Blackjack that...
Looking At The State Of Web 2.0 Security
Some of the commentary back on this blog is that the state of Web 2.0 security is in relatively poor shape, and as we have gone through a lot of the code available from larger companies, everyone approaching web...
|
|
|
10.30.07
Halloween IT Horror Stories: Hosting Providers
 By
Dan Morrill
Sometimes you just want to sit down and scream, there you are with your web site gathering an audience like you have never seen, you have arrived, you have a page rank, decent connections via all the social media, you did all the right things, life is good, then your hosting provider for some reason changes something, and there you are, DOA, dead, swimming in the dead pool.
Everyone has web hosting provider horror stories, from companies to developers, from IIS and PHP to dynamic systems where the MySQL database keeps on getting corrupted for one reason or another, just to find out that the RSS feed you were pulling keeps on sending corrupted data. Or that the hosting site takes it upon themselves to change a section of code in the web site, while well meaning...
My email was sent to the suggested "urgent" address and responded to 12 hours later. Overall down time for the site is 16-18 hours. What the hosting company decided to do was disable access to the directory hosting the RSS button tool as well as rename the .htaccess file. In their email they only stated "disabling" a script. They never mentioned anything about making changes to the htaccess file. They did this without notifying us first that there was any issue. Source: Top Rank Blog.com
If they are available, or even if they are not available, the hosting provider is going to do something at some critical moment in time that brings your whole operation to a stand still. They can't help it, this is a function of Murphy's law, a little know corollary, that the minute you really need something to work, it will go down, down hard, and essentially no matter what you do, no matter what you are in the middle of, no matter how good the day is going, the hosting provider will change something, tweak something, or otherwise ruin what was ending up a good day.
While the immediate knee jerk is to pull your account and go someplace else, think of the things you need to do, move all the data, databases, e-mail, DNS, from one host to another host. While there are programs like EZ-Move to help you out, the DNS issue will be at least 48 hours to globally change your DNS resolution. So while you suffer through the down time, the screaming customers, the screaming management, and everyone else who has raised the level of noise close to a 747 Jet Engine, there you are, waiting and praying that the hosting company will get it together.
Then magically, without admitting any form of guilt, or let alone letting you know that there was anything wrong, or that there was some other issue going on, it all works, the pressure reduces, and you forget about moving your web site because "It will be so painful" to do so.
But here is a parting word from a hosting company that I used some 12 hours into an outage.
Looks like when the server was restarted the PHP extensions remapped to the CGI version. We've updated your site to run FastCGI.
We will hang around, but given 12 hours of outage here, and 8 hours of outage there, this hosting horror story is going to be worth the 48 hour outage if I can find someone more stable.
Comments
About the Author:
Dan Morrill has been in the information security field for 18 years, both
civilian and military, and is currently working on his Doctor of Management.
Dan shares his insights on the important security issues of today through
his blog, Managing
Intellectual Property & IT Security, and is an active participant in the
ITtoolbox blogging community.
|
