I’ve experienced these issues on iSCSI storage, Fiber channel, and high speed 15k RPM SAS drives (albeit I haven’t tried solid state).  The remedy I’ve found (and it rather cheap), is to using a caching DNS relay.

My particular setup uses CentOS and Unbound.  It also uses paravirtualization mode and vmxnet3 adapters on a VMware ESX cluster for enhanced performance.  If you read further into the Unbound Optimization Documentation, you’ll get some really good tips on how to optimize the setup for your particular environment.

You’ll have to upgrade libevent as of CentOS 5.4 (version 1.4.8 seems to work as they indicate), but this is rather simple.  Just do a simple “./configure && make && sudo make install” job on that library, then when you’re running your ./configure on libevent, be sure to add –with-libevent=/usr/local as an argument.

My particular DNS is only allocated 1 CPU (which works rather well, DNS’s usually aren’t too processor intensive, just I/O intensive), so I chose to enable forked operation as well for an extra little boost (10-20% since no thread locking needs to take place).

Finally, once all you have everything set up, be SURE to add the following lines to your config

(under Forward Zones section)

name: .
forward-addr: DNS-Server-1-IP-Address
forward-addr: DNS-Server-2-IP-Address

and so on.  You can use as many or as few “forward-addr:” lines as you see fit.  The “name: .” basically tells it to forward all requests to your Windows DNS servers.  You can set it up where it will only look up your AD domain names on your Windows DNS’s then use another DNS server for everything else as well.

You also might change a few other variables like “cache-min-ttl” and “cache-max-ttl” to something like 30 and 90 respectively, to ensure that the addresses are cached for a reasonable amount of time to offload the Windows DNS servers, but also don’t get in the way of maintenance purposes.  Other variables like “msg-cache-size” and “so-rcvbuf” might need to be bumped up as well (with so-rcvbuf you might have to change a kernel variable to allow you to have larger I/O buffer sizes so DNS requests don’t get inadvertently dropped during spikes).

More information can be found in the unbound.conf documentation.

You should see a drastic cutdown in the response time and the amount of I/O load on your DNS servers.

So you’ve installed Ubuntu on your Mac running a dual boot with OSX, but you messed up and accidentally installed the bootloader to the MBR.  Well this will render your system useless as Mac’s use EFI and GPTs, rendering MBRs rather useless, outside of its MBR emulation mode that is.  However in order for it to function properly, it needs to be on your first Ubuntu partition, not in the whole disk’s MBR segment.

You could just reinstall, but that’ll will leave another icon of Tux just sitting there in your rEFIt menu.  Its a little annoying in my opinion so lets get rid of it.

Within OSX, pop open your Terminal (yeah, that thing that none of you Mac-heads ever use but us UNIX guys can’t live without).  If you’re running a Mac with only 1 hard drive then its pretty simple.  Just type in sudo fdisk -u /dev/disk0^{See note} and just type in your password and go.  This will write a new MBR to the proper segment and remove the extra logo from rEFIt.

If you’re running multiple hard drives with ZFS concatenation, or RAID, or whatever the heck you may be doing, you’re on your own, sorry

Note: Be careful in performing this action, because it does mess with your MBR. I don’t think your Mac is going to be that picky with the MBR since it uses a GPT by default but it still could render your system unbootable. As such I obviously take no responsibility for your actions. You break it, its your fault, not mine.

First make sure that you have the rtl8187 driver installed and loaded by running (do NOT plug your Alfa in yet, if you already have it plugged in, unplug it, disable any existing networking equipment, restart the wireless by right clicking on the network manager and unchecking “Wireless”, then recheck it)
Code:

sudo modprobe rtl8187

Now plug your Alfa in. You should see that the Network Manager identifies the dongle as a “Realtek RTL8187″. Go ahead and associate with a network if you can.

Now I was using the Alfa with an 8dBi antenna. You do the math to realize how much power that is, and you’re probably getting a bunch of reflections and echos. We need to turn the power down.

Code:

sudo iwconfig wlan0 txpower 50mW

Where wlan0 is the name of your dongle.

I also had an issue where the dongle was identifying the signal strength as much lower than it actually was, so it set my bit rate really low (like 1Mb/s)

To fix this, run
Code:

sudo iwconfig wlan0 rate auto

And it should read 54 Mb/s when you do another iwconfig

Now you need to reassociate to the network. You do this by going to your network manager and right clicking on the network you’ve already associated to.

After doing this my networking was screaming fast. Also continued to work after reboot, but the power went back up. I left it alone and things seemed stable. Held a stable ping over 2000 sequences with around 1% packet loss.

Hope this helps.

Part 1: Understanding the Google Algorithm and computer searching

The internet is a big place.  What amasses in the petabytes (1 petabyte = 1,048,576 Gigabytes; possibly even larger) of information, culminates one of the largest data archives in the known universe, and best yet, it’s actively archived by search engines such as Google, Microsoft Live, and Yahoo.  However, in it’s initial days, search engines were loaded down with poor information and bad sorting algorithms.  Fortunately, this is no longer the case.  Ever since Google came into being, the original revolutionary search engine sorting algorithm, pretty much everyone has updated their algorithms to be more efficient, and thus more useful.

However, despite this, the ever growing Google index can be difficult to navigate and find useful information.  With literally trillions of entries (link) in the index, searching through it to find anything from a niche topic to the latest big news article can be an absolute nightmare (and just imagine what it’s like to INDEX all of this).

First however, you need to understand a little about searching.

For the most part, computer search programs index information based on key words.  Indexing, is the action of taking significant (important) information, id est, key words, and adding the key words plus their location to a massive database called an index.  Like the index cards at your local library, the index contains significant information and it’s location, along with some other information (Google actually keeps a cache of the sites the crawl, id est they keep a copy).  However, unlike your index cards, indexes can span trillions of entries across terabytes of drive space.

Computers, however, are able to take this information and process it based on rules (aka an algorithm) and update the index with this information.

This introduces a couple of problems.

First two are technical.  The first of the two is searching.  Its extremely inefficient to just go from A->Z on the index list.  Computer’s on their own are incapable of just “skimming” through the index like we are to find the information they’re looking for.  Therefore more efficient sorting algorithms such as Binary searching are used (although I highly doubt Google uses this mechanism, binary searching is a lot more efficient than “linear” searching, but is still inefficient in it’s own respect, id est serving up the kind of traffic Google serves).

The second is keeping all of this information up to date.  Sites, such as this blog, are constantly updated with more or less of a frequency.  Wikipedia for instance gets updated very rapidly, as do popular forums.  Other sites, such as the W3C’s site or a personal “about” page, are updated less frequently.  There are quite a few initiatives working to resolve some of these issues but more often then not you’ll notice it takes a few hours/days/weeks/months for you to see changes in a search engine’s index because of the time it takes to “crawl” (crawling is a term used to indicate what search engine “spiders” do, they’re computer programs that crawl the web parsing information and updating the index) your site.

The third is probably the most obviously and also arguably the most important.  How do you group all of that significant information together in a way that makes sense and is easy to use?

Google pioneered this thought with their revolutionary algorithm, of which the byproduct is a “Page Rank”.  Google’s method was to use both the archaic method of taking your keywords and finding the site with the most of those keywords (however this resulted in quite a few “spam” sites), with a new method that measured how many sites linked to that given site with the given key words.  Exempli Gratia; say LifeHacker links to an article on Gizmodo with the key words “cell phone”.  This increases Gizmodo’s “Page Rank” because another site linked to it with given key words, so that when you search for “cell phone” you’re more likely to be linked to Gizmodo than some other site with just the key words “cell phone” and no site linking to it with those key words.

Confused yet?  Well it gets worse.  In this given example, LifeHacker has a high Page Rank, and thus it’s link to Gizmodo has more “weight” (id est, influence) on the Google Spiders than say if you opened a site on Geocities and linked to another site with the key words “cell phone”.

Short little trip off the path.  You probably are thinking “can’t this be exploited?” and the answer is yes.  Exploiting the Google algorithm is called “Google Bombing” and was very prominent in the 2004 US Presidential election (search “failure” or “waffle” and you’d see George Bush’s website and Michael Moore’s website linked to the prior and John Kerry’s website linked to the latter).

Now for a little more information.  Google has recently started keeping a history of your web searches (and web history if you use Google Toolbar or Google Chrome).  This adds to a decent level of personalization in your search results as Google “learns” how you search (how you phrase your search terms) and what information you look for.  However the extent to which this is effective is rather debatable.

Now that you have a better understanding of how Google sorts its data, lets work into part 2, which explains more of how to fully utilize the algorithm to find information.

Part 2: Utilizing the Google Algorithm (aka, good searching techniques)

First off, most common mistake is to ask Google a question.  This theory is due to A.Natural habit (you have a question, we’re all trained to ask it a specific way) and B. Ask Jeeves, which is (misleadingly) getting users to ask it questions.  Google is smart enough to omit some of the commonly used “question keywords” such as “how do I ___?”  It will extrapolate that phrase and guide you to a site that has instructions on how to do whatever.

However, a better method is to completely remove the “how do I?” part, and just go with the ____ part.  Exempli Gratia, if your question is how do I register to vote you’d enter register to vote in Google and it will likely lead you to a site that has registration information.  Also, this search is further enhanced by adding your state of residence, such as register to vote Texas.  Also keep in mind that Google does not keep your search terms linear unless you specifically ask it to.  So register to vote Texas is the same as Texas register to vote which is the same as register Texas vote to.  If you want to signal Google to keep all search terms (not omit any search terms that you inputted) and in that specific order, use quotation marks around your search term.  Keep in mind that this will seriously hamper the amount of search terms returned, and while useful for honing in on specific things such as quotes, will be a lot less useful for generic information because other people might have phrased their information differently.  Also keep in mind that words like “to” and “for” and words like that are largely unnecessary in a search, but can sometimes be helpful if you need to reference back to that search later.

Second is probably isn’t what a lot of you are going to want to hear, but, while decent search terms yield decent to good answers, great search terms will often times lead you to exactly what you want.  Great search terms often times require research, unless you already know about the given topic.

Like the Substitution Method in Integral Calculus, finding great search terms is more or less a trial-and-error process, all the time while you refine your search terms.  On average, I usually do 3-4 searches in Google before I find the information I was looking for.  Good search terms will often lead you to sites with good information that helps you refine your search terms.

However, I will most often (probably 95%-98% of the time) find my answer (or information to refine my search terms) on the first 2 pages of search results.  The thing about the Google algorithm is the information grows more irrelevant as you proceed farther down the search results.

This can usually be a huge time saver for those of you used to pouring over all those pages of miscellaneous and seemingly irrelevant information (often times it is).  If your top results are irrelevant to your topic, chances are you need to change your search terms.

But what is a good search term?  Well like I said, it’s largely a trial-and-error method, but once you get used to doing it you can usually come up with good to great search terms off the top of your head in seconds.  It also has a lot to do with how much you know on a given topic.

However lets take an example.  I know a lot about Windows Mobile, media formats (such as H.264), and the AT&T Tilt.  Someone asked me a question about H.264 playback on the AT&T Tilt.  Personally I had never tried this (I usually use my laptop for video playback if it’s necessary), so I have no idea what techniques work and which don’t.  However, I know that the Tilt in all logic should be able to run SD H.264 videos with decent framerates.  I searched “Windows Mobile H.264″ (because Google handles search terms based on key words, not on the actual question), and then scanned through the search results.  Turns out my answer was the fourth link down “Hackzine.com: HOWTO – iPod and PSP movies on Windows Mobile“.  I happened to know this was correct information because I know the iPod and PSP use H.264 as their primary codec.  I also checked the description of the link which read “Last week I mentioned that you can use TCPMP on Windows Mobile Smartphones and Pocket PCs to view H.264 encoded MP4s”.  Notice how my search terms are bold-faced?  That description was exactly the information I wanted or at least had further information on it within that link.

How do you refine your search terms though? This can sometimes be more ambiguous but will often times require you to read an article.  However, reading entire articles can be lengthy and time consuming, something we as students have very little of.  So I’d like to introduce you to a keyboard shortcut.  I know this works in Firefox (my favorite method) and in Internet Explorer (a more annoying version with a popup box).  Ctrl+F.  This is the “find” shortcut (you can go into your toolbar, and Find is usually under “Edit”, sometimes called “Find in this page” or “Where is”) and can be a life saver.  Remember how I referenced the link descriptions in Google?  How they bold-face your matching search terms to the key words on that page?  Well this is where that comes really in handy.  Typically you can see a short preview of the information on the page before you even enter it, allowing you to discern if it’s even worth looking at, but sometimes you need to read that entire paragraph of information in order to comprehend what it’s talking about.  So you enter the page, Ctrl+F, enter your search terms that were bold faced on Google, and this will take you directly to the location of those keywords in that page.  Keep in mind that, while Google’s search tool is not linear, Firefox’s search (and probably all of them) ARE.  So in Firefox, Windows Mobile is not the same as Mobile Windows, yet they will yield the same results in Google if you search the two.

So once your search terms are highlighted in the page (you may have to hit “Next” a few times to find that paragraph, but you may also need to find further information in that page) you can read up a little more information and may even be able to refine your search topic, or even find your answer.

Now I know that sounds like a lot to learn and practice.  It does take quite a while to learn well.  I’ve been using Google for years and years, and it’s taken a long time to work out my technique.  However, with all the searching we do as students, you can work out a method rather quickly.

Obviously it uses GTK for it’s window engine.  Logic tells us that it would use GTK themes in the same way.  This is correct.

You can grab your themes from Gnome-Look (only from the GTK 2.x category) and then extract them.

One you have your files, grab the folder that contains a folder with the name “gtk-2.0″ (id est, you would find the “gtk-2.0″ folder then hit “go up one level” one time, and then take that folder) and put it under C:\Program Files\Common Files\GTK\2.0\share\themes or C:\Program Files (x86)\Common Files\GTK\2.0\share\themes if you’re on 64-bit Windows.

This is the default location of GTK, if you installed it to a different you’ll need to go under /Share/themes from the root GTK install directory.

Then go to your start menu, find the GTK+ folder and launch “Theme Selector”, select the theme you chose from the “Global” list, hit ok, and now restart any GTK applications (like Pidgin).  You should have your new theme.

This method, in my experience, would only function if IIS5 Isolation Mode was enabled.

I’m not sure why but the ISAPI plugin started up just fine after enabling IIS5 isolation mode.

Check this thread out.

Microsoft has this to say about the modes of operation.

Here’s how to enable IIS 5.0 Isolation mode.

If you’re not into running Isolation mode (I wasn’t) I setup Apache 2.2 on a different port specifically to run JSP apps.  Instructions after the break…

If you want to do either way, grab yourself a cup of coffee (or another favorite) and clear a couple hours out.  The documentation on getting this done is sparse and annoyingly contradictory.

Here are a few good pages I polled info from:

Good website, very accurate, follow these configuration instructions (a bit short though)
Good information UNTIL you get to configuring Apache (do NOT follow these configuration instructions)
Old, outdated, but still somewhat useful Apache documentation

When you go to test your configuration, put

JkMount /jsp-examples/* neosmart

Into your httpd.conf file where neosmart is the worker from worker.list name defined in “workers.properties”.  Id est:

worker.list=neosmart

That should get you through it

Bear in mind that with the JkMount directive, it’s in the format “JkMount /directory/* worker” where directory is the directory to be hit by Tomcat under the default Tomcat install directory -> webapps.  Id est, in the case of Windows, C:\Program Files\Apache Software Foundation\Tomcat 5.5\webapps by default.

The star after this is a wildcard telling Tomcat to parse (well, execute, but that’s semantics) everything UNDER that directory.  Again the worker directive is the worker that you’re using for that configuration, which would be neosmart in my case.

Now that I’ve thoroughly confused you due to my enhanced lethargy and exhaustion brought on by the night’s discoveries, I hope you’ll be able to take something out of this.

IMPORTANT! Flashing your ROM is a potentially dangerous activity and I take NO responsibility for you messing up your device or anything else.

Please read this post on TiltMobility for important information about the video prior to flashing your ROM.

Well the way this systematic attack is approached is via an arbitrary browser exploit.  But what if browser exploits were cut off from the main machine?

Meet ZoneAlarm’s new security tool that’s been a bit long in coming in my opinion.  It’s called ForceField (and with good reason).

What the product does is put your browser inside of a virtual environment that’s totally cut off from the underlying OS.  It also cuts off the OS from the browser (helping negate the effects of keystroke loggers and such).

Now earlier they were giving away free CD keys for ForceField and I registered for a copy (1 year subscription).  It doesn’t have Vista 64-bit support unfortunately but it is on my laptop (32-bit Windows XP) and I’d say that it runs rather well.  My laptop is a machine that’s gaining a little age (although it’s far from old) and typically has some trouble running Firefox 3.0 with all my normal extensions.  However even running virtually, it still runs rather fast (negligible performance decreases).

So aside from negating this proposed attack, it can also protect against a wide variety of other problems, and it’s existence (albeit probably inevitable) is certainly welcome.  It will certainly be nice to be even less concerned about visited websites leaving cookies, ads containing exploits, and the like.

By the way, Microsoft Live Writer(and a lot of the Live suite of software) is very cool(using it right now for this article)

Also….I’m on the President’s Honor Roll for last semester, WOOHOO!