While each vendor has category-compliant products, they each will present a different profile in sweep tests used to verify certification. For example, characteristic impedence is permitted a plus or minus ten percent variance for the swept frequency range. Vendor A may be below the line at the lower end of the sweep, but still be within the permissable variance/tolerence while Vendor B is above the line (when I say "line" i'm talking about the target, the goal, the specified median) and Vendor C can vary back and forth across the for the full range. When Vendor A creates panels and jacks for its cable, it can adjust (through the design & manufacturing process) the panels and jacks to compensate for the below-the-line values and create a balanced, overall improvement to the cabling SYSTEM performance. The same for Vendor B and C ... because they know the absolute performance characteristics for their cable, they can compensate to improve performance of thier system in the other components. This is generally why it's a good idea to stay with one vendor's product, end-to-end. i.e., putting Vendor A's jacks and / or panels on vendor B's cabling may result in an amplification of Vendor B's "above the line" stats and actually push the system outside the Category-rated envelope. I used characteristic impedence as an example, but the concept applies to any of the basketfull of cabling component specifications. Also remember that "structured cabling" was designed for more than just Ethernet networking. Once vendor's cabling system may offer better specs for applications other then Ethernet networking (like video, RF, or or serial / terminal operations). Many cabling component providers will warranty the performance of their cabling system, as long as all primary components are part of that vendor's current cabling system catalog. AT&T used to (maybe still does) warranty their professionally installed and certified cable plants for ten years (I think it was quite a long time, for cabling). Any performance failures shown to be attributable to the (professionally installed and certified) AT&T cable plant was fixed at no additional charge. If you've ever chased ugly cable bugs, you can understand what kind of cost savings this can represent. Staying within one vendor's component system contribute to the likelihood that the overall cable plant will operate at full, optimized performance. ... View more

I don't know the maker or model number, but it's almost certainly an omni, probably 5db or less; probably ~2-3dbi unless there's a hidden power wire. It's just in a decorative radome (the cone-shaped shell around the antenna element). Google on "wifi antenna" and you'll get a fairly long list of suppliers. Good Luck ... View more

Tftp is a native application for those distributions. You should be able to enable them in the services admin tab. There is also likely a ftpd.conf file in /etc or a subdirectory below /etc, where you configure the target directory, access, and the rest of the details. ... View more

I like the idea. It's a common feature in many BBS applications. It might be interesting to see how many hits are collected over time (as a metric to how many visitors are using search), or as an indicator for archive (no increment in six months gets sent to archive), or even just to get a general interest level, i.e., a post regarding some ew feature gets a lot of hits, maybe the Cisco writers want to do an article to clarify the point-in-question. IMO, it'd be a Good Thing and should be added, if possible. ... View more

As general information to cover a couple of your points, keep in mind that antennas and environment will be the primary factors in how far a signal goes at a specific power level. Formulas are nice for ballpark, but you'd be surprised how many things can screw up a "straight forward" installation. Any objects intermediate to the AP and client can have a very broad spectrum of effects. A tall shelf can cause multipath, a human can suck up a big chunk of signal, leaded glass decorative partitions can create a shadow such that a client won't connect from 20 feet away. While it is a Good Thing to have paper plans and calculations, they should be confirmed with a comprehensive site survey. Once you have a good idea of how reactive (or not) the environment is, you can match up the desired operational areas with antennas that can create the pattern and the power necessary to meet the goals. You may also be fighting noise / interference and some adjustments to your preliminary power levels maybe needed. The best way to reliably judge is to perform a survey; initially for design, then post-installation to pick up nulls, dead spots, hot spots, and places where the signal goes beyond the intended area. Corrections can be made by using different antennas, changing power levels, moving the AP/Antenna(s), moving objects in the room for better signal path, etc. A final survey for signoff should (IMO) be done, so the customer has an opportunity to see that the goals are met, everything works as agreed, all issues have been resolved, AND THE DOCUMENTATION MATCHES THE "AS BUILT" configuration (so payment will be forthcoming and unhindered). "No job is complete until the paperwork is done" ... View more

Looks good! Congrats on another great feature! ... View more

The back-to-back cable has a DTE end and a DCE end ... check to make sure you have them connected to the right ports (i.e., check the labels, and / or swap ends on the cable). ... View more

Flip to the main Cisco site and search on "Terminal Server" and / or "Reverse Telnet" (with the right choice of IOS, the same general principles and processes apply between SSH and Telnet). ... View more

The best FREE tool I've seen is Qcheck. It's an ultra-small version of Chariot, an awesome (but expensive) software test suite. It'll give you throughput, latency, and packet drop (used to give you IP and IPX); latency might be a trick, but clever use of WireShark (also free) can provide that number as a delta response time. QCheck http://www.ixchariot.com/downloads/qcheck.html WireShark http://www.wireshark.org/ Good Luck ... View more

I put an answer in, whether I feel the OP will like it or not. While the question comes from the single source that was motivated enough to ask, the answer will likely serve many more times the people who just search to see if their questions has already been asked. Giving only sunshine & flowers answers doesn't serve the greater community; you gotta call a stupid idea a stupid idea ... it might save someone from walking down a seriously bad path. You can sense, in many cases, when an OP is looking for validation of what they feel is a good idea (or at least acceptable idea - but it's really not, in any sense) ... in those cases, IMO, you have to lay it down hard to let them know how far outside the "best practices" envelope they are heading. I don't chase ratings, I've never asked for a score; I answer when I believe I have helpful input (and will have time to follow-up if necessary). It's not about the score, though there is an undeniable anticipation ... ... View more

As described, yes, it will cause major problems. To use concurrent parallel/redundant links, other protocols come into play, or you will have created a bridge loop ... which has the potential (and usually does) swamp the switch with broadcasts and take everything down (occasionally requires a power-cycle to recover). If you want to do "teaming" you need to set up the client and switch to use the same teaming protocols. Teaming is usually implemented for two general functions; redundant/backup links or/and bandwidth increases. What are / were you trying to accomplish? As far as troubleshooting goes, if you don't have labeled cable runs, it can be ugly. Ultimately the quickest easiest to way is to unplug everything, and then start plugging things back in one link at a time, wait for the link to come up and watch the utilization indicators on the switch ... it it goes from blink...blink ... blink to solid all lights on within a few seconds, you have a redundant link, leave it out. There are other ways, this is the fastest. Your best bet is to hire a professional network person / consultant / VAR out to scope you network and make some educated recommendations ... then hire a Hitman to smoke the person that set up the original network and the person that hired 'em. This is what "getting off cheap" does for you ... a professional will walk away rather than set it up poorly (like this) because of unrealistic budgetary constraints. ... View more

This used to happen a lot, especially with Symbol access points. The issue is when a client roams from AP1's area to AP2's area, the APs are not handing off the client cleanly, such that the switch sees the same client in two domains. You might be able to configure the APs for a cleaner handoff, or the clients for a snappier transition threshold. Or maybe there is a firmware update available for the switch or APs (or clients) or all of the above. Good Luck ... View more

Please post a SH INT for the affected interface(s); a SHOW BUFFERS might be useful as well. Packet drop / loss is usually either congestion, as seen by the transmitter, {or | which causes} queuing timeouts. The congestion can be the receiver's inability to catch the traffic being pushed to it. This happens a lot when you have asymmetrical bandwidth (one end is a T1, the other is a 56K DDS or Frac T1, for example). ... View more