Umm, Why Is Hoosier Access Taking Forever to Load?
Very astute of you fair reader. I too was wondering why Hoosier Access wasn’t loading. The unfortunate thing is, our webmaster is in China. Taking a chance, I emailed him considering that it’s evening over there. I got an immediate response! (Yes, he’s that good!) Turns out some files or what not are being moved around on our server. Here’s what he sent me.
(Read more below the fold)
(From the Server Update Blog on Sunday)
Intrepid RAID Migration
Sunday, March 23rd, 2008Your server, intrepid.acenet-inc.net (’Intrepid’), will be swapped out with a new server. The new server will be using RAID-1O instead of RAID-6. You can look forward to a much faster website… two enterprise web hosting servers have already went through this and the feedback has been breathtaking. Server performance metrics are just as impressive, too. This isn’t hype — you can expect your website to be a lot faster in 2-3 days.
Right now all the data on your server is being copied to the new server. Be warned: service could be a little sluggish during this massive file migration. But… This short term grief will provide long term relief (I stole that slogan from a highway contruction sign). Due to the amount of data on this server the file migration could take a couple of days.
Stay Tuned… this blog will be updated frequently (at least daily).
(And then from Tuesday)
Intrepid files are still being moved
Tuesday, March 25th, 2008There really is no status change…All the files are still copying, and it looks like this will take several more days, running 24/7.
I’ll update this blog when the file copy gets close to completion. And then I’ll update this blog frequently during the migration.
For you techies, I hope you understand that. I just hope things are up and running soon.
March 26th, 2008 at 2:01 pm
Thanks for whatever that means.
March 26th, 2008 at 2:37 pm
clear as mud…
March 26th, 2008 at 2:50 pm
Yep, exactly what I thought.
March 26th, 2008 at 9:16 pm
The site seems to be loading pretty quickly right now. It will be interesting to see it when it’s up and running on the new servers!
March 27th, 2008 at 10:00 am
The quickness leads me to believe that the transfer is either completed or near completion. Here’s hoping so.
March 27th, 2008 at 12:34 pm
RAID 6 stripes data across multiple (mirrored) disks, and calculates parity for added redundancy and stores parity information across one set of stripes. RAID 6 is highly fault-tolerant as a result; if a disk fails, you pull it, put in a new one, and the array is rebuilt without losing data. What makes RAID 6 slow is the need to calculate parity on write operations.
RAID 10 (aka RAID 1+0) also stripes your data across multiple (mirrored) disks, but without parity. Without parity, there is an “acceptable” level of fault-tolerance as long as failed disks are replaced promptly, and since there’s no parity calculation, the configuration is quite fast.
Under RAID 10, if you lose all of the disks in a single mirror array, you lose all of your data across the entire stripe set. With RAID 6, even if you lose all the disks in the mirror, it can still be rebuilt from the remaining disks by using the parity bits.
In other words, you’re been moved to a faster but somewhat less fault-tolerant configuration. I’d call this a reasonable tradeoff.
But why in the world is your webmaster in China?
March 27th, 2008 at 12:37 pm
Thanks for laymans interpretation at the end of your explanation. I was lost a bit.
As for our webmaster, he’s there on business for the company he works for which is based here in Indianapolis. He’ll be returning this weekend.
Don’t worry, we’re not going to export that job!
March 27th, 2008 at 1:52 pm
Mud? What mud–I feel like I’m the high-school jock lost in a dungeons and dragons convention…
March 27th, 2008 at 4:17 pm
….and I just happen to be a great big dork….