WizGidget

March 9, 2011

Running out of Addresses, part IV (and Last)

Filed under: Belton Journal, Info Bytes, Tips & Tricks — pmckinley @ 12:00 AM

Last week I introduced IPv6, which is the eventual solution to the IPv4 address exhaustion issue.  Besides having a much, much larger addressing space (from 4 followed by 9 zeros to 3 followed by 38 zeros), IPv6 has a number of other features built in,  including some security features.  It’s more complicated than IPv4, which means us propellorheads will have to get bigger propellors.  Probably the biggest issue with IPv6 is that it isn’t compatible with IPv4.  If the equipment is designed for it, such as routers, computers and such, the equipment can handle it, but it’s like having two different networks on the same wire — they don’t talk to each other.

There is some equipment that can translate between the two protocols, but I suspect what will happen is that internet providers will at some point simply stop issuing IPv4 addresses and require using IPv6 addresses.  I don’t think that will happen for several years, but when it does it may mean buying a new router and/or getting some help with reconfiguring the network on our computers.

For instance, I run both a Cisco “Pix” firewall and a wireless router on my home network.  The Pix supports IPv6 at software version 7.0.  I can upgrade the software in the Pix, but I don’t think the model I have can be upgraded past the version I have already, so it’s likely my Pix will become obsolete.  That was an expensive piece of gear — about $500 new.  The modern replacement for the Pix will cost about $350 new, although I might could find a used higher-model Pix that could handle IPv6.  And then again there are probably other firewall-routers out there that would do the same thing(s) for much less.  The wireless router would also have to be replaced — it doesn’t handle IPv6 no way, no how.  Fortunately there again there are better wireless routers on the market that sell for $100 or so.  One of the things I really need the Pix for is getting a VPN (“Virtual Private Network”) connection into my home network from “outside.”  The newer wireless router I mentioned does VPN, so it might satisfy the need within a single network appliance.

It seems likely that most of us will have upgraded to IPv6 compatible equipment by the time it becomes a requirement.  Granted that I tend to run things for years past their normal lifetimes, but I have had to replace our wireless router about 5 years ago when the old one succumbed to a lightning storm, or getting too hot or something.  So, it seems reasonable that it will have to be replaced again eventually, and all the new ones support IPv6, and usually also are faster and more powerful.

IPv6 is supported on Windows XP and later, and although it needs to be turned on and may take some tweaking, after being turned on it should configure itself just as your PC does now when you connect it to your wireless or wired router.

Got a computer question or having an issue that would make a good article?  Contact the author through the Journal or WizGidget.com

March 2, 2011

Running out of Addresses, part III

Filed under: Belton Journal, Info Bytes, Tips & Tricks — pmckinley @ 12:00 AM

Last week the discussion left off with addressing for end users — people who are accessing the internet like you and me.  Many if not most internet providers use a live internet address for each of their subscribers, although the addresses are dynamic — a new address is issued every time you connect.  So, for instance if you had your computer connected directly to your cable modem or whatever, and I knew your internet address, I could conceivably connect to your computer.  This would be a Really Bad Idea, because if I can connect to it, then so can all the hackers in Korea, China, Russia and Romania.  This is why I recommend people use either a wired or wireless router between their computer and their internet connection.  So, if every internet user gets his own address, that could mean it would be difficult to get a connection to the internet.  You might try to connect, but couldn’t get anywhere because you couldn’t get an address.

It’s possible that your internet provider could use something called “NAT” (Network Address Translation) to expand their assigned addresses out, many users to one address.  The typical wired or wireless router for home use does this — it needs one address on the “outside” to connect to the internet, but it also handles having many addresses (computers, printers or what-have-you) on the inside.  The addresses on the inside aren’t true internet addresses, they’re usually on what’s called a “non-routable subnet”.  So for instance, your “outside” address might be 66.196.3.249, while the inside network is 192.168.123.(something between 2 and 254).  Your router would “live” at 192.168.123.1 on the “inside”, and your computer would be for instance 192.168.123.10.  When you access the network, the router translates between 192.168.123.10 and 66.196.3.249, so from a server on the ‘net, it looks like the connection is coming from 66.196.3.249.  The router tracks connections (each connection has a unique sequence numbers assigned) so that when whatever server responds, the router says “Oh, that’s connection number 1928734, that means the packet needs to go to 192.168.123.10 on the inside.”  The cool thing about this is there’s no way to create an inbound connection: the exchange has to be initiated from the “inside,” and this is why using a router provides security.

The NAT thing could work for normal users, but some of us propellor-heads have servers we run from our “home” connection.  Those servers have to have a live address in order to be useful, so that the connection can originate from the “outside.”  So, there’s no way of getting around it, sooner or later we’ll need more addresses.  What we need is a new addressing mechanism that has more addresses, and that solution already exists, in the form of “IPv6.”  IPv6 uses 128-bit addresses, so the new address space supports 2128 (3.4×1038 or 3 followed by 38 zeros) addresses.  Weren’t we talking about Really Big Numbers just last week?  IPv6 technology is already available on most operating systems like Windows 2000 and later, MacOS, and I would expect all versions of unix from the last 10 years or so.  Next week will finish up with what IPv6 means to us end-users.  Be sure to send in your questions for future articles.

February 23, 2011

Running Out of Addresses, Part II

Filed under: Belton Journal, Info Bytes, Tips & Tricks — pmckinley @ 12:00 AM

Last week I introduced the internet address exhaustion as an issue, and explained the difference between an internet domain and an internet address.

As mentioned, the current addressing scheme uses four numbers separated by dots.  Actually the address is four 8-bit numbers (remember everything computer is binary at some level and “bits” are digits in a binary number), where an 8-bit number can have values between zero to 255.  Four of these numbers translates to 32 bits (8×4=32), ) which is effectively 2 times itself 32 times, a number that translates into 4,294,967,296. There’s some overhead in how the addressing works so the actual available addresses is maybe a half percent less than that.  The point though is that it’s a somewhat limited number, about 4 addresses for every 6 people on earth.

I can imagine a scenario where every person needs several addresses, such as one for your phone, one for your PC, one for your laptop, and so forth, so despite the fact that there’s lots of people in the world who don’t use the internet, that number is made up by all the web and email servers in use. Eventually we’re bound to run out, and that’s what happened February 3, 2011, at least the last of the main blocks of addresses were allocated to regional internet registries.  At this point either your eyes have glazed over, or your propellor beanie is really spinning, so let’s get down to what it means to us civilians.

In terms of domain names, it doesn’t really mean anything.  Remember we have DNS (Domain Name Service) to provide mapping between URL’s and the actual address.  Internet domain names have their own limitations, but it’s more of a practical limitation than a functional one — domain names can be up to 63 characters (a-z, 0-9, and “-”) long, not including the www or the .com.  That’s a LOT of permutations (think 37 characters raised to the 63rd power, or 6 with 98 zeros, a Really Big Number), but typing in 63 characters can be a bit cumbersome, especially if they’re random characters, so that cuts down on the practical list of domain names.

Also, it’s possible to run multiple domains on the same address.  The http protocol specifies that when you type in a URL, the packet of information that gets sent to the server to request your webpage includes the URL of the page.  It’s not difficult for the server to sort out which web server to send the request.  The web server where wizgidget.com lives has something like 30 different domains/websites.  It’s quite common to have hundreds or more, although the web hosting services that load too many suffer from poor performance. Your website’s cheap hosting service may mean that they’ve leveraged their server a bit much so your site may be a bit slow to load at times.  So, address exhaustion could possibly be an issue for websites, but we’ve kinda handled that.

Another possible issue would be end users  like you and me — people who are trying to access the internet.  You have to have an address to get to the internet.  Next week we’ll continue with this issue — addresses for end users.

February 9, 2011

Secure email

Filed under: Belton Journal, Tips & Tricks — pmckinley @ 8:00 AM

Last week I mentioned that email is one thing you could do safely at a public WiFi, IF you have your email configured to use SSL. SSL stands for “Secure Sockets Layer.” It builds a virtual “tunnel” through the network, much the same way that the VPN connection  does. The data is encrypted in such a way that someone snooping the network stream would have a very, very difficult time cracking the encryption so that they’d have access to your information. The entire data stream is encrypted, including login information such as username and password.  Without SSL, someone on the same local network as yourself could easily “snoop” the network stream and read all of your information, including usernames and passwords.  I would recommend using secure email even if you only use email from home.

Most email providers have the ability to support SSL-encrypted email. There’s two streams for email, both the incoming which is either IMAP (Internet Message Access Protocol) or more typically POP (Post Office Protocol), and the outgoing stream which is SMTP (Simple Mail Transport Protocol). SMTP is the protocol used to send email; it’s also the protocol used to send email between email servers. From a network perspective, each protocol uses a standard “port”, for instance, POP normally uses port 110 and SMTP normally uses port 25. SSL encrypted POP and SMTP normally use port 995 and 465, respectively. IMAP normally uses port 143, while SSL imap uses either port 585 or 993, depending on whether it’s normal IMAP tunneled through an SSL connection, or a newer IMAP standard that incorporates SSL. Your email provider should be able to tell you what ports to use for their email servers. Often they provide this information on their web page. For instance, Yahoo has a web page with instructions for setting up your email application (such as Outlook or Outlook Express); the URL is http://help.yahoo.com/l/us/yahoo/mail/yahoomail/mailplus/pop/pop-35.html.  Alas, Yahoo’s pop-able email account is a fee-based service.  Google on the other hand has free pop-able email.  Their instructions are at http://mail.google.com/support/bin/answer.py?answer=75291

Basically, setting up secure email in Outlook or Outlook Express involves turning on the SSL layer, setting the correct port numbers, and of course testing that your email still works.  For Outlook, click the Tools menu, then pick Email Accounts.  The “View or change” radio button is selected by default, just click the “Next” button, then pick the account to change, and click the “Change” button on the right.  In the window that pops up, click the “More Settings” button, then the Advanced tab.  Check both of the SSL checkboxes — POP should automatically reset to port 995, but you’ll have to change SMTP to port 465 if that’s what’s required.  Outlook Express is similar: click Tools, then Accounts, then the mail tab.  Pick an account, and click the properties button.  Next click the Advanced tab, and make the same settings as Outlook.

Spouse Ann wants to know what’s all the hoopla in the news recently about the Internet running out of addresses, so we’ll explore that next week.  Have your propeller beanie ready.

February 2, 2011

Surf Safe Part II

Filed under: Belton Journal, Tips & Tricks — pmckinley @ 8:00 AM

Last week we talked about some of the ways that your internet surfing can become risky from a networking perspective.

The first thing for keeping safe at home is either to use a wired network or to make sure your wireless router is using WPA or WPA2 security with a “Pre-Shared Key” (PSK) which is basically a really long password.  Being an engineer I like using some form of scientific number like Planck’s Constant that I’ve memorized  as the PSK.  Other people like to use a phrase or lyric that means something to them and is thus easy to remember.  It doesn’t matter what you use, but the more characters or numbers it has the more difficult it will be to guess.  Using WPA security effectively makes your wireless session undecipherable to other computers, even ones that are connected to your wireless network.  It also makes it fairly impossible for someone without your PSK to connect to your wireless network.  This is important, because as we discussed you really don’t want strangers and neighbors to have access to your network – it’s like leaving your front door unlocked when you’re away.

Most wireless networks at coffee shops, airports & hotels don’t use WPA security because that would make it difficult for their customers to use the network.  So, while you’re connected to that network, your computer is exposed to others who might find your information useful.  You could just have fun browsing the web, but using a web-based email reader, ebay, facebook, or anything that requires a login is not a good idea.  In other words you can’t do anything really useful. You can be somewhat useful catching up with your email… if you’ve made sure your connection to your email server is encrypted using SSL.  Most email providers and email applications like Outlook can handle an SSL connection, if you configure it properly.  But that’s only going to help if you’re using your own computer (laptop or tablet since you’re at a coffee house, right?), AND you’ve configured your email properly.  We’ll discuss that next week.

Pretty much anything you’re going to do in a web browser needs similar security, and fortunately there is a solution in the form of a “VPN” service.  VPN means Virtual Private Network – it’s a software method for encrypting your entire internet session so that nobody between your computer and your VPN provider can snoop your web session.  VPN works by configuring a virtual encrypted “tunnel” between your computer and the VPN provider.  Just do a search for “VPN provider” and you’ll find a number of companies that provide this service for varying prices.  I recommend checking out any company that you pick, because while your neighbors won’t be able to snoop your session, the VPN company can.  Google the company name or domain and see if anything nasty shows up before you sign up.  For instance, when I was doing the research for this article, at least one of the sites listed on vpnpreviews.com was flagged  by my antivirus program as being a fraudulent site.

January 12, 2011

Find Out What’s Going On, Part II

Filed under: Belton Journal, Tips & Tricks — pmckinley @ 8:00 AM

Last week I left off with a description of the Processes tab in Task Manager, which is where we’ll continue this week.

I mentioned last week to be careful when you kill (“End”) a process.  There are some processes that will render your computer unstable or cause it to reboot if they are killed.  Svchost.exe is a good example – you may have noticed that there’s a bunch of them.  You may be able to kill one or two of them off, but you’re playing russian roulette — you don’t know which one is going to blast your computer’s brains and cause a hang or reboot.  McAfee published an update to their antivirus suite awhile back that clobbered the actual  svchost.exe file — after which the PC would not boot.  Fun, eh?  I, fortunately, don’t use McAfee.

Another thing that’s handy to do with the process list is to become familiar with the processes that are running there.  If you see a process that you don’t recognize, you can search it on Google.  You’ll most likely find information on what that process is and what software package it goes with.  You might even find one that’s an actual virus, although often these days viruses disguise themselves as legitimate processes — like svchost.exe.  I actually found one once, when I’d connected my Windows-2000 laptop up to a client’s network with no firewall.  Within 30 minutes my laptop had been infected with the MSBlaster virus.  I discovered it when my laptop started behaving peculiarly, and noticed the msblaster.exe process running.  That looked suspicious, so I checked it out and eliminated the virus.

You can also use the Processes tab to figure out which processes are being memory hogs.  Firefox, while my favorite web browser, seems to have a memory leak, and especially if it has a website loaded that has flash images.  The only solution for that is to occasionally kill Firefox so that it will free up the memory.

Lastly, there’s a Performance tab, that shows a graph of CPU and Page File (PF) usage, which is a measure of how much memory you’re using.  The PF graph is something to watch, because if the memory usage approaches or exceeds the amount of physical memory your computer has, it will slow things down considerably.  How can the memory usage exceed the memory available, you might ask?  Because Windows will swap pages of memory out to disk so that it can effectively have more virtual memory than it has physical memory  That’s why the graph is called the “Page File” graph instead of “Memory” graph.  The problem is that swapping memory pages in and out to disk is horribly, interminably slow, at least in terms of computer time — disk acess is maybe a thousand times slower than physical memory access.  How do you know how much memory you have?  There are several boxes with numbers underneath the PF graph, one of which is labled Physical Memory, which also lists the “Total Memory”.  Divide that number by a million to get physical memory in Gb.  Mine has 2612588,  which translates roughly to 2.5Gb of memory.

January 5, 2011

Find Out What’s Going On

Filed under: Belton Journal, Tips & Tricks — pmckinley @ 8:00 AM

Windows has a nifty tool called Task Manager which is handy for a number of things.  You can access Task Manager with a right-click in the taskbar, somewhere that there’s not an icon already showing. You can also get the Task Manager with a control-shift-escape key sequence (press and hold the control key, then simultaneously press and hold the shift key, then press the ESC key).

One of the more common uses for Task Manager is to stop a program that won’t stop when you click the window’s “close” button (the X in the upper right corner.)  Once Task Manager is running, the Applications tab will show the list of active user programs you have currently running.  There are probably other programs running that don’t show in the Applications tab, but we’ll cover those later.  Once you find the program you need to stop from the list, you can right-click the program and then select “End Task.”  You may get a dialog box if the program doesn’t want to stop quickly.

The next tab is the Processes tab, which lists *all* of the programs running, including many programs that are set to run as a “service”.  This will often be a fairly long list.  You may recognize some of those processes, such as firefox, Outlook, or iexplore.exe (Internet Explorer).  You can also stop programs from this list, but be careful: killing the wrong process can cause your computer to be unresponsive or even reboot.  “svchost.exe” is one of those, and you’ll probably see several “svchost.exe.”  They act as sort of a “wrapper” that in turn run other programs like your networking services.

When you right-click on a process in the Processes tab, you may also notice an option to “Set Priority.”  This allows you to define which programs are going to get preferred treatment by getting more time to run on the CPU.  If your computer is a bit slow, and you have a program running you want to run a bit faster than others, you can use this setting to help.

Another option from a right-click on a process is the Set Affinity. If, like son Jesse you now have a quad-core CPU (4 cpu’s in one CPU chip that run simultaneously) you can tune your applications by telling them which CPU core to use.  Windows does a pretty good job of divvying out the workload so it’s not usually useful, likewise if your computer only has one CPU core.

Finally, the Processes tab is handy for figuring out who’s being a cpu or memory hog.  You can click on the column headings to get Task Manager to sort those columns.  The first click is going to sort it from low to high, which isn’t very useful.  Click it again and it’ll sort from high to low, which makes the hogs float to the top.  Sometimes if my computer is being sluggish I’ll sort the CPU or Memory column and kill off the worst offender.

More on Task Manager next week.  This week’s article was suggested by spouse Ann, I’d welcome suggestions from other readers as well!

December 29, 2010

2011 Computer Resolutions

Filed under: Belton Journal, Tips & Tricks — pmckinley @ 8:00 AM

Christmas has come and gone, the new year begins Saturday.  Are you one of those that make new year’s resolutions?  This could be a good habit, just from the point of view of acknowledging that we all can develop new habits to make life better for ourselves.  I thought this would be a good opportunity to review some things and make recommendations for the new year, things that you can incorporate into your computer habits to make your computer use safer, more enjoyable and less stressful.

My first recommendation is to make sure you have good antivirus software on your PC.  Although Microsoft is supposedly coming out with some reasonably good free antivirus software, I’d recommend against the free stuff.  Let’s face it, it takes a LOT of time to watch for new viruses and do the updates to the anti-virus to handle each new virus strain as it comes out.  The cybercriminals are constantly looking for ways to circumvent the current antivirus solutions, so its a continuous task to keep the antivirus solution up-to-date.  Since nothing in this world is truly free, follow the money: who’s paying for the work to keep it up-to-date?  You are though your subscription if you’re using a paid, commercial product.  I used to use CA, but had problems with that when I upgraded the hard disk on my laptop: I’m currently using TrendMicro, which seems to work reasonably well and doesn’t noticeably slow things down.

Why is this so important?  First and foremost, it protects your personal data, because that’s what many of the current viruses are designed to do: collect your personal data for cybercriminals to use.  Secondly, because another common purpose for computer viruses is to use your computer to commit cybercrime such as sending out tons of spam (see wizgidget.com/typhoid), or mounting a denial-of-service attack.  You may have heard about denial-of-service attacks recently in the news — they were used against the British government as well as Visa (the credit card company)  in retaliation for the Wikileaks founder Julian Assange being arrested for alleged sexual assault.  Whether or not you agree with the Wikileaks thing, I’m thinking you’re not OK with having a stranger use your computer to make political statements.

Secondly, set up regular backups for your PC.  If you don’t currently run backups on your computer, go out and get one of those external USB hard drives, and check with the salesperson to make sure it includes backup software.  One of mine actually has a utility that allows me to make a boot CD that will then completely restore my laptop… if I’ve taken a good backup image first.  I have mine set to take an incremental backup at night every couple days, so that it catches any new or changed files.

Finally, I recommend being careful about what information you give out on the internet.  Use disposable credit card numbers (see wizgidget.com/virtualcard) when you order stuff online.  Make sure your facebook and other online profiles either don’t include your birthday or have a bogus year.  I recommend 1982, that would make you what?  29?

Santa brought spouse Ann a Kindle e-book reader this year.   Look for e-book reader tips in future articles.

December 15, 2010

Disaster Recovery

Filed under: Belton Journal, Info Bytes, Tips & Tricks, Warnings — pmckinley @ 8:00 AM

A computer disaster can come in many forms.  It might be a disk failure, or the computer is lost or stolen, or you just accidentally erase or overwrite a file you needed.  I wonder how many of my readers have taken steps to deal with a computer disaster?  Making backup copies is good practice.

Traditionally, backups were done with some form of tape drive.  Most large companies still do some form of tape backup, although backup-to-disk has also become popular especially with new deduplication technology.  Backup to disk basically means the data gets copied to a secondary disk device.

These days tape technology has just about become obsolete for the typical home or small business user.  I have a two-drive tape library that holds 10 tapes, with each tape holding something like 40 to 60 Gigabytes (Gb) per tape depending on how well the data compresses (some data compresses well, other not so much).  A few years ago when a 100Gb disk drive was really huge, having 10 tapes in the library was ample capacity.  Nowadays the disk drive capacity for the typical PC has grown to the point that even my strato-geek 10 tape library is struggling to keep up.  Spouse Ann’s new PC has a 500Gb drive.  I just upgraded the disk in my laptop to 500Gb.  That’s a terabyte (Tb) of data to store on tapes that might only get 40Gb per tape — I’d need something like 25 tapes just to take one full backup if both disks were reasonably full (fortunately they’re not).  And that doesn’t include another 100Gb or so spread across two Unix servers and a Windows server that I run.

Fortunately there is a good solution available for the casual user.  Not only is the capacity of disk drives  increasing, but external disk drives have gotten quite cheap.  It’s not uncommon to find an external 1Tb disk drive that will run off your USB port for something in the range of $100.  Many of these disk drives come with backup software that is quite adequate for the casual user.

Even though I have all the fancy equipment to do backups, I still need something to use when I travel.  Recently I bought a 1Tb external drive to use for backups when I’m on the road.  I paid a bit more for it because it was a model that has a dock accessory available.  It’s about the size of a largish deck of cards.  When I’m home, I plug the drive into its dock.  When I’m on the road, I take it with me, and use a regular USB cable to connect it at night, so that backups (I use the included backup software) will be done while I sleep.  Half my brains are on my laptop, if it were destroyed or lost, I would have a really hard time supporting my clients.  Having a backup of my data means that in a worst case scenario I could go buy a new laptop and restore the important data within a day or so.

November 24, 2010

Back to the Future after a System Restore

Filed under: Belton Journal, Tips & Tricks — pmckinley @ 8:00 AM

Last week we discussed using the Windows System Restore tool, and I commented that I’d let you know how it went with finding the issue with backups.  It turned out that the problem stemmed from the “Connection Manager” software I’d installed to support my broadband wireless card.  This week I thought it might be helpful to go over the process I used to isolate and resolve the issue.

The Windows System Restore tool is found on Windows XP by clicking the Start button, then All Programs, then Accessories, then System Tools, then System Restore.  The first screen on the System Restore window gives three choices: Restore, Create a restore point, and Undo my last restoration.  I picked Restore and clicked Next.  The next screen shows a calendar, with today’s date selected, and the list of restore points (if any) that were created for today.

Because I wanted to be able to put things back the way they were, I started with the current day and worked backwards, recording any software that had been installed, in the order it had been installed.  I kept going back day by day until I found the backup software install.  That way I developed a list of applications I would have to reinstall on the way back to the current setup.  Some of the restore points were listed as “Software Distribution Service”, which are clusters of updates done by the Windows Update or Microsoft Update.  I didn’t need to worry about which updates those were, I could just let Microsoft Update take care of tracking which updates needed to be reinstalled.

Once I had the list of software that had been installed after the backup software, I went ahead and did the System Restore, which requires a reboot.  I immediately tested the backup software to see whether it worked or not.  Joyfully it did, and so I knew that something that came after that broke things.

Even though the Windows and Microsoft updates had been interspersed with other installations, I thought it might be good to go ahead and get all the updates installed first.  I have Windows set to download, but not install updates, so most of them I could install that way.  Others I went to the Windows Update website (there’s a link on the Automatic Updates tool, available from the Control Panel) and forced the install.  After all the updates were installed and I’d tested backups, I installed each application, one at a time.  I tested backup after the application install, then rebooted, and tested it again.  When I installed the Connection Manager, backup still worked, but after the reboot it no longer worked, so I knew I’d found the culprit.  It turns out that  the broadband card manufacturer Sierra Wireless also has a connection manager tool separate from the cell phone provider, so I tried installing it instead, and that one doesn’t break backups.  It doesn’t have the nice bargraph showing how much bandwidth I have remaining for the month, but I can live with that.

« Newer PostsOlder Posts »

Powered by WordPress

Bad Behavior has blocked 49 access attempts in the last 7 days.