Wireless client ARP caching, or the worst kind of high-tech laziness

As I no longer work as the senior network engineer at Plantronics, I have been reconfiguring an old Cisco 8xxW series device for use as my local router/gateway.  Given that I no longer administer a WLC, I am configuring the access point in autonomous mode.  I have been suffering for the last couple days because everything was working terrifically, except that I no longer had connectivity between wireless devices.

More specifically, wireless clients could connect to wired clients, and vice versa.  Wireless clients could not interact with other wireless clients, unless the ARP request happened in the first few moments after one of the wireless clients authenticated (before the AP added the client’s IP address to it’s internal table).  After much gnashing of teeth, packet traces, and web searches related to things like PSPF, I found this little gem on Cisco’s website:

When the wireless device receives an ARP request for an IP address not in the cache, the wireless device drops the request and does not forward it. In its beacon, the wireless device includes an information element to alert client devices that they can safely ignore broadcast messages to increase battery life. (link)

As it turns out, all that is required is for my access point to actually respond to client ARP requests instead of sometimes forwarding the requests (dropped by the client) or dropping them out of hand.  I tested both of these, and they both worked great.

dot11 arp-cache
dot11 arp-cache optional

Now I have back the functionality I really wanted, which is to say, ARD from my laptop to the Mac Mini in the living room to add more time for children with homework and the burden of parental controls without actually having to get my lazy ass up and walk there.

Laziness rules.

Freedom and Belief

Freedom of religion is anatomically equivalent to freedom from religion; which is to say, there are no state-sponsored consequences resulting from your choice of religion or lack thereof. Whether that results in more or less ‘freedom’ for you really depends on you and not on the state.

Religion, on the other hand, is generally just acceptance of a standardized set of spiritual beliefs. Christians are people who accept that Jesus is their personal conduit to salvation. Catholics are people who, while believing in the same god, accept that the Church is their conduit to salvation. To be part of a religion is largely to accept a preconfigured set of beliefs and make them your own.

It is belief itself which destroys freedom. If you “believe” that a thing is valuable, then you no longer have the freedom to assist or ignore an attack on it.  If you believe that thunderstorms are troublesome for aircraft, then you no longer have the freedom to fly through hurricanes.  If you believe that Jesus is your main man, then you no longer have the freedom to sacrifice human beings to your idols.  All of these, of course, being freedoms that you give up in order to retain your beliefs and maintain an internally self-consistent mindset.

The question really is; if there exists truth out there somewhere, and you find it and believe it, and it constrains your actions in some manner… are you more free or less free?

Routing Loops

For amusement value, this is what happens when you have a rogue network engineer who decides to appropriate and advertise an IP network already in use in your enterprise.  IP addresses changed to protect confidentiality.  AddressA is located in China, Routers A-C are located at WHQ in California, and Routers D-H are located in Europe.

>tracert AddressA
Tracing route to AddressA over a maximum of 30 hops
1 1 ms <1 ms <1 ms routerA
2 <1 ms 1 ms <1 ms routerB
3 3 ms 3 ms 3 ms routerC
4 * * * Request timed out.
5 167 ms 162 ms 164 ms routerD
6 165 ms 171 ms 163 ms routerE
7 165 ms 165 ms 164 ms routerF
8 187 ms 189 ms 190 ms routerG
9 190 ms 190 ms 190 ms routerH
10 * * * Request timed out.
11 199 ms 199 ms 202 ms routerD
12 199 ms 211 ms 197 ms routerE
13 200 ms 199 ms 204 ms routerF
14 222 ms 225 ms 225 ms routerG
15 223 ms 224 ms 224 ms routerH
16 * * * Request timed out.
17 233 ms 234 ms 235 ms routerD
18 241 ms 237 ms 233 ms routerE
19 237 ms 234 ms 235 ms routerF
20 258 ms 259 ms 258 ms routerG
21 263 ms 264 ms 260 ms routerH
22 * * * Request timed out.
23 291 ms 321 ms 306 ms routerD
24 319 ms 327 ms 272 ms routerE
25 278 ms 275 ms 269 ms routerF
26 292 ms 295 ms 294 ms routerG
27 294 ms 295 ms 293 ms routerH
28 * * * Request timed out.
29 310 ms 304 ms 309 ms routerD
30 318 ms 304 ms 307 ms routerE
Trace complete.
>

grad school

Dear Colin …,

I am pleased to offer you admission into the Ph.D. program in … at the University of California, Santa Cruz beginning in the fall 2007 quarter. …

that is all.

2007 Jujo Jiang Invitational

I played in the tournament this weekend in San Francisco. I placed 4th in the 4k-8k division. I tied Karoline Burrall in points (we both had 3/5 wins), but she won the game that we played, so that was the tie-breaker. Prizes in this tournament went down to 4th place; I won $25, which was odd, since it cost me $35 to play. You’d think one would at least win enough to cover your entry fee. But no matter, it was $25 I didn’t have, and I enjoyed all my games except the third, which I lost because I played way too fast. The game against Karoline I lost because of a fighting error that was easily worth 70 points, I lost by 31. A simple extension instead of the clamp I played would have won me the game easily. Frustrated by the loss, I went on to destroy her younger sister, Julie, winning by 55.

In any case, I really enjoy tournament go, it’s great to play with a clock against new & different opponents who don’t play the way my usual opponents do. I’ve played in two tournaments now, and placed in both of them. I hope I can keep that up. ;]

smart-chucks

While driving today, I saw a bumper sticker that said Eschew Obfuscation. I silently agreed, but wondered what the point of the bumper sticker was. Then I realized…

Irony.

Nam Ch’i-hyeong

I went to the Sunnyvale Go Club tonight to see the visiting professional, Nam Ch’i-hyeong (another link), from Korea. She is actually a – professor – of Baduk in Korea (Baduk is the korean name for Go). She played a game against Lance Kemper (shygost) even, and won by resignation around move 100, best estimates are that she was already ahead by about thirty points. She played against another dan ranked amateur at 5 stones and won by 14 points. She had to leave, and those of us left sat down and played a few games. I played against someone ostensibly of my own level, and after making 4 or 5 grievous errors, I still won by 10 points. Go figure. (pun mercilessly intended)

I made an assertion not too long ago that what I needed at this stage was not more study, or problems, or books, but just to play many many games to get more experience under my belt. The inconsistency of my play tonight was easy proof of that. Back to KGS.

I’m looking forward to a 4 day workshop starting a week from Wednesday with Yilun Yang (link, link). This is going to be a solid 4 days of go training. I’m very excited.