I am not one to drink water from plastic bottles, but this might be a good reason to pick one up and keep it in the kitchen when you’re about to make that next batch of Hollandaise sauce for Eggs Benedict. I have no idea what she’s saying in the video, but it’s very cool and pretty self explanatory if you watch.
Archive | How-to RSS feed for this section
So I received the following email from someone asking about a VPN connection between remote offices and a main office:
I have a TZ170 with a static IP (dsl) at my so called corporate office (server resides in this office). The appliance is set for DHCP for the clients that are set up on the inside of the firewall at that corporate office.
All my remote offices have a basic setup with either cable or dsl (no static ip), behind a modem and a dlink router. When more than one person in the same remote office connects to the tz170 at corporate, both clients experience awful delays and disconnections. If only one client connects in that remote office it works great, but as soon as you add another person from the same office that try’s to connect forget it, nothing but problems. Is this because the tz170 is seeing to tunnels coming from the same ip (isp assigned)?
Will purchasing another tz170 for the remote offices solve my problem? Is there an additional configuration that I am missing in the tz170 that will enable me to do this ?
Here is the response I sent:
You are absolutely on the right track. The problem you are having is that more than 1 person from the same public IP address is establishing a tunnel.
There is not a good way to establish a tunnel using a VPN client from more than one client behind a NAT device to the same central VPN device. In this case, the user has a D-Link router as the NAT device. Some devices do a better job of handling the NAT for IPSEC VPN traffic, which is what the Sonicwalls use. The only thing he could try in this case, other than the guaranteed solution of implementing a remote-office VPN gateway device, would be to ensure that the D-Link is upgraded to the latest firmware and has the appropriate IPSEC pass through settings. The most reliable solution, though, would be a VPN appliance to maintain a site-to-site VPN device at each remote office.
I attended the MS Technet Briefing yesterday that covered both Virtual PC 2004, which I have already started using myself, and Exchange 2003 disaster recovery, which I have experience with in prior versions, but not so much in 2003. I was blown away by the new Recovery Storage Group feature. The presenter demonstrated both a recovery of deleted items from a single user’s mailbox, and he showed what was called a dial-tone recovery of an Exchange Information Store.
For anyone that hasn’t done an Exchange disaster recovery, this may not be that impressive, but here are the steps for a single-server disaster recovery of a crashed Exchange Private Information Store (not a crashed Windows server). This outline assumes a good backup from the previous night.
- Information Store crashes, and the administrator is notified.
- Administrator stops the Exchange services and moves the corrupt copies of the priv1.edb and priv1.stm files to another location for possible recovery at a later time.
- Administrator restarts the information store service and the IS fails to mount. Administrator mounts the IS and is informed that this will result in the creation of a blank database. Administrator accepts the warning. USERS ARE NOW ABLE TO SEND AND RECEIVE EMAIL. Total elapsed time to this point should be less than 10 minutes.
- Administrator creates a recovery storage group in the Exchange System Manager disaster recovery section.
- Administrator begins a restore of the Exchange database to the RSG. (This will happen automatically, since the mounted blank database that is currently in use is protected from overwrite by the restore process, and the RSG database is automatically configured to allow an overwrite by a restore. The administrator does not have to tell the backup to do anything other than restore the IS backup.)
- At this point, Administrator has a choice to either begin an Exmerge of the data from the restored database into the newly created blank copy, which could take a significant period of time, or to swap the recovered database with the live blank one and then import only the new items from the mostly blank database.
As anyone who has ever done an alternate server recovery with Exchange can attest, this process is incredibly simple and fast!!!! Way to go, MS. This feature, alone, is almost worth the cost of an upgrade to Exchange 2003.
Thanks to Roland Taglao for the pointer to this article by Phil Agre at UCLA. For people that do what I do, that is help people with computers, this article represents the same kind of things for us that we think users should know about the computers and don’t. If more people who help people with computers acted this way, I wouldn’t have as much success with my business cleaning up after the Saturday Night Live Computer Guy.
I found this article at Ars Technica. I have a couple of different clients that have this problem, and I am going to try their solution. We’ll see what happens.
The folks over at the Exchange weblog, You Had Me at EHLO! have a great article on adding an RSS feed to a Sharepoint website. More importantly, their article has an XSLT for RSS feeds that you can use in other applications.
As I have been helping folks with Typepad domain mapping, I have run into a couple of problems lately, and I wanted to share the information I have gathered with others that might be trying to map their domains themselves. Here are the problems of which I am currently aware:
- 404 Error after mapping everything correctly – this error can be caused by a problem on the Typepad side. In order to troubleshoot where the problem lies, you need to do the following if you are running Windows XP, 2000 or NT:
- Click start then run and type cmd, then press enter.
- At the resulting command prompt, type nslookup and press enter.
- At the nslookup prompt, type the mapped domain name you used, e.g. www.mydomain.com, and press enter.
- If the response you receive looks something like this:
then the DNS side of things is set up properly. If you get a request timed out, then type the domain name again and press enter. If you get any sort of different response than the one above, then you don’t have DNS set up properly.
If you are running some other OS, then you can use an internet-based DNS lookup tool, such as this one from zoneedit.com. On this tool, type the domain you are looking up in item 1, change item 2 to “Return All”, and your response should have a line that looks like this:
www.mydomain.com CNAME yourtypepadaddress.typepad.com
Assuming that your DNS settings are resolving properly, and you still receive the 404 error, then the problem is with the update to the Typepad web servers. According to the Typepad help, it should take about 2 hours from the time that you complete the domain mapping options in your site control panel, but my experience on this has ranged anywhere from a couple of hours to several days. I don’t know what is going on on the backend with the necessary updates on Typepad’s part, but I find the response for the update to be basically unacceptable because of its unpredictability.
- Mapping an uncommon domain name, i.e. www.somedomain.co.uk – I don’t know what to say here. When you are setting up the mapping on the Typepad control panel, it strips out part of the domain if you type in www, but it also seems to cause a problem if you don’t physically type the www in the box (even though it is typed for you in front of the box). The workaround to this seems to be typing the domain like this: www.www.somedomain.co.uk. I don’t really understand the logic behind the coding in the Typepad interface on this one. I hope they change it soon.
- Domain forwarding with some providers – some providers offer the ability to forward your domain name for you. DO NOT SET THIS UP IF YOU WANT TO DO DOMAIN MAPPING IN TYPEPAD. The way the providers set up forwrading is to put the website to which you are forwarding, in this case your Typepad weblog, in a hidden frame inside your browser. Think of this like driving down the road and going to someone’s house, looking through the window, and seeing a completely different house inside. If you set up this type of forwarding and try to do the domain map inside Typepad, the links that Typepad generates will never work properly.
- Mapping to a site, or a blog, or a photo album – I haven’t yet actually mapped a domain straight to a photo album, but I have mapped to both an entire Typepad site and distinct weblogs within a site. The only caveat here is that if you map to your Typepad site as a whole, and you want to track inbound links in a tool like Technorati, you may find some unusual results. When you map to the site, Typepad generates links to your individual items and archive pages as http://www.yourdomain.com/blogname/actual_file.htm. The reason for this format is that if you have additional blogs on your account, you would be able to access them through your domain by simply putting a /blogname on your URL. Technorati sees these individual links, though, as a separate blog from your main domain, so you may not actually see all the inbound links to a particular post or item if you do a search for the link cosmos on just the base domain name. This is just something to keep in mind.
I hope this is helpful to folks that are trying to do domain mapping on their own. Keep in mind that I still offer to set all this up and troubleshoot any issues on your behalf for $25 US per domain, and if you find my site particularly helpful I have followed someone’s suggestion and put a donate button under my affiliations list on the right side of the page.
I have helped a number of home users with their computing needs, and I think it might be time to share some basic ideas with home users about home networking. There are a number of factors that have contributed to the need for more complex computer setups at home, among them are broadband Internet access and the need for more computers in multiple-child households. In this article, I will attempt to introduce novice and intermediate computer users to home networking concepts with future posts relating to specific how-to articles on these technologies. You can click on any of the images in this post to see a larger version.
First, let’s talk about what a network really is. A network is any group of devices that are connected together, like the digram on the right. In most cases today, networks are connected with wires through something called a hub or switch. The whole idea of networking is to allow the connected devices to share their resources with one another. These resources might include things like files on their hard drive, printers to which they are connected, access to the Internet, or perhaps even access to some program that is running on one of the devices.
One of the main reasons people set up a network in their homes is to allow more than one computer to connect to their broadband Internet connection simultaneously. In order to accomplish this a router of some sort is needed. Think of a router like an old fashioned traffic cop – they look at the information on the network and decide how to send it along where it needs to go based on the signal that is contained in the information. A lot of home networks use something in Windows called Internet Connection Sharing, which turns one computer on your network into a router for the others. Other home networks, like the one in the diagram to the left, use one of the multitude of commercial router appliances that are available on the market today from manufacturers like Linksys, DLink, and Netgear.
Finally, some home computer networks also use wireless networking to allow them to connect. Typically, this wireless networking will be based on a standard called WiFi. This general term really refers to a group of standards for wireless networking that were created by the IEEE, 802.11. The most common form of WiFi connection is 802.11b, which uses the 2.4ghz wireless spectrum and maximizes the speed for network traffic at 11mbps.
I know that last part introduced a couple of acronyms and words we haven’t discussed, so I will try and explain them a little more here:
- The IEEE is the Institute of Electrical and Electronics Engineers. It is a standards body that writes standards for a lot of the technology used in the computer world so that devices from different manufacturers will work together.
- 802.11 is the standard number for wireless networking. The IEEE uses numbers to refer to the standards it writes. For instance, the standard for Ethernet, an almost universal form of wired networking that is used in most homs is number 802.3.
- 2.4ghz is a portion of the radio spectrum that is used by a lot of home wireless devices, including many cordless telephones and 802.11b and 802.11g wireless networking devices.
- And finally, 11mbps means 11 megabits per second, which refers to the speed of the network. For comparison purposes, a typical wired computer network allows for 100mbps, and some today are even allowing 1000mbps, which is also referred to as 1 gigabit per second.
We have discussed here some basic ideas of how a home network can be set up. In the next article in this series, we will discuss a number of different Internet connection methods, including DSL and Cable, along with some specifics around how to share one of these connections between computers.
If you are at all interested in web design, then I suggest you learn about CSS. There is a book on my bookshelf to the left that I am currently reading that gives a very thorough exploration of the subject. In brief, the idea of CSS is to separate the content (for instance, the text you are reading here) of a web page from the design aspect (for instance, the color, size, orientation, font, etc. that applies to said text).
One of the best ways to explain CSS to the first timer is to see it in action. One of the best sites I have found for a demonstration is the CSS Zen Garden.
I am by no means an expert on the topic of CSS, for that go take a look at Eric Meyer’s site, but I do think it is something that any fledgeling, or experienced for that matter, web designer should know. If you are using a Typepad site, then your whole site is driven by a combination of CSS and the Movable Type template tags. I will be playing with the style of this site over the holidays. If anyone has suggestions, please let me know.
Many people are unfamiliar with the protocol called NTP (Network Time Protocol), so I thought I would comment on it and give some basic information on its use. If you would like to read the full technical details of the protocol, you can go to the IETF website and read RFC 1769.
NTP allows computers to synchronize their time over a network. One computer uses another as a server and requests information on the current time and date settings. Several complex calculations are performed that measure latency and other information between the two computers, and the client computer determines if an update is required and then updates its time based on the server’s settings. All versions of Windows since Windows 2000 have had a reliable time service that runs by default and can be configured to synchronize with a time server of the users’ choice.
If your Windows 2000 or XP computer is a member of a Windows domain, then it is automatically configured to synchronize with one of the Domain Controllers on your network. DO NOT CHANGE THIS SETTING. Windows 2000 and later uses an authentication protocol called Kerberos that uses the time as part of its process – if the workstation and Domain Controller times are not in sync, access to network resources may not be possible.
In Windows 2000 or XP, you can modify your NTP settings with the following commands at the command prompt (start->run->cmd[enter]):
net time /querysntp – this command will display the current default time server; many Windows machines come preconfigured to synchronize with time.windows.com.
net time /setsntp:serverlist – this command allows you to set the list of time servers with which you wish to synchronize.
net time – this command will synchronize the time on your computer with a time server from the sntp list.