My Hardware Solution . . .
-
@cmb:
Nice setup!
The patch cables are shorter than the spec technically permits, but that shouldn't be an issue (maybe if you need collision detection to work properly, but that's not necessary with switched networks). I've seen custom made 1" patch cables working in production for years with no issue.
I think short patches to a patch panel is ok, where the length behind the patch panel gets you back above the minimum spec length. But, again, there has been much debate about real minimum lengths other than anecdotal stories where a too short cable was blamed for a problem or a short cable causing other issues (such as bend radius, plug/boot strain, equipment proximity, etc.)
-
I have heard of the cable "minimum length" requirement, some say a foot, others a meter, others 8 feet, and others 17 feet–between active components (i.e. a router to a switch). Mostly, I have heard this requirement in reference to cable testing equipment. I have never seen an informed explanation as to why there should be a minimum length--it does not appear to be a gain (required loss) issue--some say too short a cable will lead to reflections (end to end resonances at undesirable frequencies) that can cause data loss. If that were true, there would likely be a minimum length for coaxial (rf) cable also (i.e., between a transmitter and an antenna)--I have never read a discussion about such a theory. By the way, behind my patch panel I use two foot cables to connect to my active components--so my minimum length from active device to active device is 5 feet (but the connectors on the patch panel themselves could cause reflections . . .). "If it ain't broke, don't fix it."
-
LackRack is a hoot, less then $10 for the US. If only I was still a bachelor :P
Nonsense, is the exhaust out the back or sides of rack?
-
I personally have one of the 'Enterprise' Lack Racks and the shelf is great for your larger server. Also good for supporting coffee mugs. ;)
Steve (still a bachelor)
-
I have heard of the cable "minimum length" requirement, some say a foot, others a meter, others 8 feet, and others 17 feet–between active components (i.e. a router to a switch). Mostly, I have heard this requirement in reference to cable testing equipment. I have never seen an informed explanation as to why there should be a minimum length--it does not appear to be a gain (required loss) issue--some say too short a cable will lead to reflections (end to end resonances at undesirable frequencies) that can cause data loss. If that were true, there would likely be a minimum length for coaxial (rf) cable also (i.e., between a transmitter and an antenna)--I have never read a discussion about such a theory. By the way, behind my patch panel I use two foot cables to connect to my active components--so my minimum length from active device to active device is 5 feet (but the connectors on the patch panel themselves could cause reflections . . .). "If it ain't broke, don't fix it."
You're definitely good then, I wasn't thinking patch panel when I posted. The only explanation I've seen for minimum cable length on CAT* Ethernet is if collision detection is required to function, which generally doesn't matter anymore given there aren't many hubs around.
-
I personally have one of the 'Enterprise' Lack Racks and the shelf is great for your larger server. Also good for supporting coffee mugs. ;)
Steve (still a bachelor)
I've got an Enterprise model too, but I had to add structural support for heavy servers.
Matt (recently a bachelor again)
-
What's that WLAN gear there?
-
Gluon99–through the back and sides--stays very cool as Atom does not generate a whole lot of heat.
Gridrun--I don't follow your question--see my opening post.
-
Is it wrong that I now want a red LackRack to match my Watchguard Firebox x700? :)
-
What could be wrong about that? ;D
Why didn't I think of this at Ikea?Hmm, perhaps I could add a beech veneer to my firebox….
Steve
