New to the forum, but not new to FA, and certainly not new to hardware failures. I'm not trying to be condescending in the least bit, nor am I trying to make myself seem better than anyone else. I'm not here attempting to troll. I'm simply asking for clarification, and maybe a clue or two as to why the long delay.
I work for the government. I'm used to lowest-bidder crap for hardware. We've had (very) critical hardware failures at least six times since the beginning of my employment, as a result, corrupting and destroying terabytes of data. And the equipment, I can guarantee, is a bit more advanced than a simple web-server. Over 128 terabytes of arial photographs, photographs of title and deed documents, mortgage documents, among other heavily used information that is maintained on a daily basis. And despite several NIC, motherboard, hard drive, RAID controller, CPU, and RAM failures, it's never taken us longer than a weekend to revive our systems. This includes a full redeployment of Windows Server 2003, and several UNIX and Sun virtual systems, a full restore of an active directory environment for over 7,000 computers, and 6,000 employee accounts, completed by two people, including a full security lockdown/test setup. Myself, and my supervisor were said two people. This is minus order/shipping times, which took five business days. So my question is, if two barely mediocre hacks such as myself and (less so, he's pretty good) my supervisor can take care of these issues in a short amount of time, why is FA going to be down for so long?
On another account, we had a full RIS server failure in the middle of a deployment of sixty new public-access terminals. It took approximately six working hours after the "NTLDR.EXE is missing" message to repair, including a full upgrade to WDS over RIS, and several added images. The issue was a dying on-board RAID controller. Putting more than thirty computers on the connection at once over-taxed the system and finally got it to drop. Now the system is back up and running, and can handle over three hundred simultaneous imaging connections without even so much as a double-take.
On a third account, the Clerk's Office home page has gone down due to hardware failures twice in its existence, exists on old Gateway (I didn't make the decision, politicians with tight pockets did) servers, and has a total down-time of 15 hours since its creation in 2002. Average hits per day? 1,265,000. The site has obviously undergone several changes since 2002, however has remained stable despite being hosted on a $1300 server with a single T3 connection. (The Hamptons and the real estate being what it has been with foreclosures and the like have increased the number of hits significantly, and the sever still rarely sees above 80% load.)
I fail to see how it could possibly take a group of people two weeks to install, deploy, and test a single group of web-servers, some of the most basic kinds of servers to set up.
Maybe I'm just short-sighted. Maybe my standards are a little high/stringent. Maybe working with private sector businesses, then subsequently the government has me a little tight-arsed with what I expect out of a site. But all that aside, why over a week from hardware install to final deployment? Even working two hours a day, that should only take three days. And that's assuming you're using a *nix-based web server, and throwing IIS into the trash where it belongs. (SQL injection attacks FTL. See also: Comcast.)
If you'd like, I'll gladly take over, complete the project. What's the donation amount up to now? $10k? With 6 hours and a $10k budget, I'd build a server cluster that could handle the load FA recieves on a daily basis, and still have about 60% resources free. Then again - that's also assuming the inflated government prices companies usually throw at us. Seeing as it's technically private sector, I'm sure I could also complete the project under-budget too, given the chance to shop around some. Hell - have the server built by Dell, and get yourself some 4-hour support, and guaranteed compatibility. Critical hardware failure? Four hours to repair and be back up and running, or your money back. Our section of the government switched to Dell two years ago for most of its servers, and down-time has been no longer than four hours per incident since. Total cost more than custom-built? $1150.00 Not too shabby, considering we've gotten over $8,000 worth of free parts out of it.
Sorry if my first post is a little confrontational. It's just a touchy subject with me when I know it can be done quicker, with the same amount of confidence, using different resources. I'm only trying for some answers, not to ruffle any feathers or to rub any fur the wrong way. I suck at art. I'm a barely mediocre musician. I'm an okay author. I only do two things exceptionally well. Cars and Computers. Cars aren't very useful to the furry community. Computers can be. I'm more or less just offering my services to the community.