Picture of new PAFOA servers

[andrewjs18]

nice.

I'm installing HP proliant blades in banks now. The bank uses a slighlty lower model than yours, but fuck is that puppy loud when it fires up...

3X 72gb SCSI drives.

[melD]

Originally Posted by andrewjs18

nice.

.. is that puppy loud when it fires up...

What...it's kind of a thrill to me. Sound like they might fly out of the rack at ya. They do that when they get angry also!

Get the Dell service agreement if you can. It pays off some times. We have got a few free machines out of ours.

Mel

[mojo]

Originally Posted by danp

Load balaning (for the short term) will just be round-robbin DNS. At some point I'd like to get us set up with a BigIP, but those cost some coin.

Do I hear a fund raising event coming up? I will always do whatever I can do to help someone play with a BigIP box.

[Kaos]

I remember the first time I fired up a rack of IBM blades...I thought someone turned on about 5 shop vacs all at once.

After spending a week working in a datacenter doing bicsi quality (to give you a frame of the amount of time for the necessary detail) work on cat5/cat6 and terminating fiber....earplugs are essential.

[andrewjs18]

we had 1 existing server in a 14u cabinet that was hella loud....the cabinet was a dust ball, so it probably needed a good blow out.

[P-11 shooter]

Originally Posted by danp

Yes, FreeBSD/apache for the main setup, things like images/stylesheets will most likely be run through lighttpd as the system currently does.

Load balaning (for the short term) will just be round-robbin DNS. At some point I'd like to get us set up with a BigIP, but those cost some coin.

Dan, I don't have a clue what you guys are talking about, but I'm happy as hell for ya.

That's some sexy hardware you have there.

[edstephan]

Very Nice Dan and Doug.
Good looking cluster
Ed

[625]

What about your switches, routers, and firewalls? Do you run a simple, single switch and single firewall, or do you have some type of fail over system that allows for component failure and easy maintenance? I realize this isn't some type of hospital system, I'm just curious. The rack of servers I am resonsible for has too many single points of failure on the sever side and the network side. It's driving me crazy.

[andrewjs18]

Originally Posted by 625

What about your switches, routers, and firewalls? Do you run a simple, single switch and single firewall, or do you have some type of fail over system that allows for component failure and easy maintenance? I realize this isn't some type of hospital system, I'm just curious. The rack of servers I am resonsible for has too many single points of failure on the sever side and the network side. It's driving me crazy.

what makes me very angry is when there is spaghetti wiring. =(

esp. when you have to reuse an existing port on the switch that isn't being used but there's still a cable plugged into the switch..

ugh, makes me mad.

[doug]

Originally Posted by 625

What about your switches, routers, and firewalls? Do you run a simple, single switch and single firewall, or do you have some type of fail over system that allows for component failure and easy maintenance? I realize this isn't some type of hospital system, I'm just curious. The rack of servers I am resonsible for has too many single points of failure on the sever side and the network side. It's driving me crazy.

Using a single cisco 2924 split into two vlans, that is fed directly into the 6500 aggregate switch.. the entire datacenter sits behind two 12000 series routers.

we're a gun forum, not a hospital, so no, we don't need crazy levels of redundancy...

the forums are now moved over to some of the pafoa hardware.