Sometimes developers try to use Crystal directly to pull data from multiple database platforms - we understand the need to give you additional tips to increase Crystal report flexibility. Also common case when Crystal report should pull data from multiple database platforms. Assuming that you are creating Crystal Reports on Windows platform - we recommend you to use Microsoft SQL Server as a database access tool.

1. Use Crystal when you have your query tuned – regardless how good Crystal Reports is - it is reporting tool - not querying environment. Tune your query first and be sure that you are pulling the right data.

2. Use SQL Server Linked Servers. Use professional tool, such as Microsoft SQL Server, where you can create linked servers to Oracle, Ctree, Btrieve, MS Access, Unidata, DB2, filePro, Fox Pro, FileMaker Pro and other platforms.

3. Use Distributed Queries – in the case of SQL Server - research the possibilities with OPENQUERY, OPENROWSET clauses to connect to non-SQL Server linked servers.

4. Create Stored Procedures - when you need parameters to be transferred to the SQL statement - you need stored procedure. You need stored proc if temp tables are required to do intermediate job in your SQL query.

5. Report Flexibility – when you have report based on the SQL view or stored proc - if you need to modify the data query - you don’t have to change the report - you change your SQL view/ stored procedure.

6. Create SQL Query to probe the data – again - we always recommend tuning your query and see that you are getting adequate results – in any case – Crystal Report is just a nice tool to show the results of your query.

Happy customizing, querying and designing! if you want us to do the job - give us a call 1-630-961-5918 or 1-866-528-0577! help@albaspectrum.com
About The Author

Andrew Karasev is Chief Technology Officer in Alba Spectrum Technologies ( http://www.albaspectrum.com )– USA nationwide Microsoft Great Plains, Microsoft CRM, Microsoft RMS customization company, serving clients in Chicago, Los Angeles, San Francisco, San Diego, Boston, New York, Houston, Dallas, Miami, Atlanta, Minneapolis, Seattle, Canada, UK, Australia, Germany, Russia, Brazil and having locations in multiple states and internationally, he is Great Plains Dexterity, SQL, C#.Net, Crystal Reports and Microsoft CRM SDK developer.

akarasev@albaspectrum.com

Issue:

We installed a licensing server and then installed Metaframe 3.0 Enterprise on a clean Windows 2003 server. Added this server to our XPa farm and made the 3.0 server the Data Collector. Any time a user would click an icon on the web interface the IMA service would blow up on the 3.0 server. This occurred on additional 3.0 servers that were brought up as long as they were data collectors. Changing the farm so XPa servers were the data collectors fixed this issue. The event log logged this when the IMA service blew up Microsoft Visual C++ Runtime Library : Runtime Error

Solution:

I’m told that there is an issue in the SQL database that causes the 3.0 server acting as a data collector to blow up the IMA service. Our fix was to create a new SQL database. Create a new farm using the same name and republish all the applications. We then uninstalled and reinstalled Metaframe and pointed the servers to the new farm. Rather than an upgrade we rebuilt our entire farm into a clean Citrix Metaframe 3.0 Enterprise environment. While rather extreme this did resolve the issue. We recreated our published applications by hand. You may find a tool for this here http://www.citrix4ge.de/ctxsdk/expimppa.htm

Issue:

When we performed an upgrade on a server from XPa to Metaframe Presentation Server it only upgraded it to Advanced. When we used add/remove programs to add in the installation manager and resource manager the server stayed at advanced edition and never registered its proper edition with the license server. Thus users received errors that the system had reached its license logon limit.

Solution:

Open the management console. Right click the server and choose set edition type. Set it to enterprise then reboot the server TWICE. Check the startup licenses on the licensing server to see that your server is listed as enterprise. What bit us was the two reboots required. The system only asks for one but it took two for the server to register properly.

Issue:

Upon adding a Metaframe 3.0 Presentation Server to our XPa farm changes could only be made to published applications if you pointed your Citrix Management Console at the new 3.0 Presentation server. After the IMA blow up we removed all 3.0 servers from our XPa farm only to find we could change no application.

Solution:

We had to add 3.0 back to the farm and then instruct admins to point consoles at the 3.0 server.
About The Author

John Gall is a full time IT Manager and IT Consultant in Minnesota. He runs several websites as a hobby featuring content similar to that in this article.
http://www.gallconsulting.com
http://www.sandalspictures.com
http://www.napavalleypictures.com
http://www.winesilove.com
http://www.sturgistravels.com
http://www.gallconsulting.com/blackhills

In this article we will look at what SSI is, why it makes life so easy, and exactly how to set up SSI on your website…and then I’ll share two little bonus tricks I’ve discovered.

What is SSI?

SSI actually covers a number of features to improve your website. I am going to speak here of just one critical improvement, referred to as an “include” file.

Essentially, an “include” file is a separate file that your web page can include as if it was part of the page file. Let’s take a real-life example. My site at http://www.vitamin-supplements-store.net uses SSI in a number of places.

If you are used to using FrontPage or have recently learned HTML, you probably assume that every web page is a single html file. The example above is actually five files. There is the main HTML file for the page. There is a style sheet (CSS) and there are three SSI “include” files — one for the left navigation menu and two for the two navigation menus across the bottom. I could have, perhaps even should have, used more “include” files, but you can be the judge when you’ve finished reading this article.

Why use SSI?

SSI makes it easy to bring changes to your website. There are some changes you will need to make to every page or to every page in a specific section. Your website might start out with just 10 pages, and you might figure that it is no big deal to cut-and-paste a change to the navigation menu 10 times.

But each time you add a page, you have to update that menu. And with each new page added, you have to paste one more time. When you reach 50 pages (Never thought your website would grow that big? You’re not alone.) it becomes very tedious to update.

For instance, when I wanted to add the website monitoring logo and the link to the navigation menu at http://www.thehappyguy.com, I had to change just one file, and — presto! — the change appeared on every page of the site bearing that navigation menu. It was so very much easier than past updates before I began using SSI.

There are two other benefits to SSI “include” files. Because a single line of code replaces what might have been several dozen in each HTML file, your files are much smaller, taking less space on your server. And, because the “include” file has already been loaded with the first page a visitor sees, the next page is much quicker for visitors to load.

How to set up SSI?

You need three things to set up SSI.

1. Configure your server for SSI
2. Set up your “include” file
3. Call up your “include” file into your web page HTML file

First, you need your server configured for SSI. Ask your web host if this has already been done. Also ask if it has been set up to parse .html extensions to read SSI.

If your server has been set up for SSI, you are one step closer. If it has also been set up to parse .html extensions to read SSI, you are two steps closer.

Not all hosts support SSI, but most do. If yours does, but it has not been set up for your account, look for the .htaccess file in your root directory (where your index.html or home page file is stored). In my experience, this is not usually viewable using an FTP process; you have to find it through your control panel.

The .htaccess file is a text file. If you do not already have an .htaccess file on your server, you can create one in NotePad or even in Word (just save it with a .txt extension), but whatever you do, make sure not to write over a .htaccess file already on your server. If it is already there, just add the following lines to the file, being careful not to erase anything that is already there:

AddType text/html .shtml
AddHandler server-parsed .shtml
Options Indexes FollowSymLinks Includes

So far, so good. But this will recognize only .shtml files for includes, and you probably don’t want to change the .html extensions to .shtml on all your pages, if for no other reason because it will mess up all your inbound links to those pages.

So add this line. In theory, it should be the second line, but careless me has gotten it to work at the end, too:

AddHandler server-parsed .html

Save the edited .htaccess file to the root directory of your server.

As an aside, there are several ways to configure your server for SSI. This one has worked for me across several hosts.

So much for the techie stuff, now you need the second element: the “include” file itself. Suppose your “include” file is the navigation menu. You don’t need “head” and “title” and “body” tags. Just type in your text and code just the way it would appear in your webpage. It’s that simple. Save it as an HTML file, and load it to your route directory or to its own directory.

All you need now, is to call up your navigation menu in each file you want it to appear in. To do this, a simple line suffices where a whole column of code and text once stood.

Here is the line to place in your code:

This assumes that nav1.html is the name of the “include” file and that it is in the same directory as the page it is being inserted into. If you place the file into a directory called “nav”, you would have to include code like this in your web page file:

That’s all you really need to know. But here are a couple bonus tricks that will make SSI “includes” work even better for you.

Bonus Tip #1

You don’t need to stuff everything into a single SSI file. What if some elements you want on some pages and others not? Let me offer two examples of situations that call for splitting the SSI “include” file in two.

The first is on my vitamin site. Return to http://www.vitamin-supplements-store.net and look at the links across the bottom. The first row is typical website stuff: contact, privacy, etc., which one wants access to from every page of the site. The second is the links directory, which a webmaster does not typically want linked from every page. For example, see the bottom of my humor article at http://www.vitamin-supplements-store.net/articles/eggs.html. The row of links directory is not there.

The second example is this article on search engine tips: http://www.thehappyguy.com/SEO-tips.html. Notice again that there are two different navigation consoles. The first is generic to the site, the second is specific to that section of the site. So a second SSI “include” file, using just one additional line of code, can provide interlinking for the section, without affecting other sections of the website. This is very handy for large sites.

Bonus Tip #2

You can also use an SSI “include” file to hide some of your source code. I will show you a ridiculously simple way to do this.

There are at least two legitimate reasons why somebody might want to hide parts of their code. The most obvious would be if you are running a proprietary script. The second is if you are running a script that displays content, but shows up as a script in the source code. This was my challenge.

At the bottom of the navigation menu at http://www.thehappyguy.com/self-actualization-articles.html, there is a “Happiness Quote of the Day”. In fact, it is a random quote that reloads when the page reloads. I had included the random-text script in the SSI “include” file, but the search engines were seeing the script in the source code, not the text that human visitors were seeing. So I placed the script in its own “include” file, which I inserted like this into the main “include” file:

Now the search engines see the same things as humans see. Why would I care? Search engines visit more frequently pages that change more frequently. That does not mean the site will rank higher, unless frequent change is part of the search engine’s ranking algorithm, but it does mean that other changes will be indexed faster.

To sum up, SSI “includes” can save you time and headaches when changes need to be made to your site, plus they reduce the file size and increase the loading time of your pages. Once you’ve set up the server to read SSI, all you need to do is create an “include” file and call it up in your web page using one line of code.

I can hardly believe I wasted so much time cutting and pasting before I learned about SSI. You won’t believe it either!
About The Author

David Leonhardt is a freelance writer, and an online and offline publicity specialist. Contact him at: mailto:info@thehappyguy.com
For a copy of Don’t Get Banned By The Search Engines:
http://www.thehappyguy.com/SEO.html
For a copy of Get In The News:
http://thehappyguy.com/publicity-self-promotion-report.html .
To promote your website with an ezine:
http://www.thehappyguy.com/ezines.html
Info@thehappyguy.com

This article discusses a number of atomic timing references and how they can be used to provide synchronisation of computer equipment and networks.

National Radio time and frequency broadcasts transmit accurate time and date information from a radio transmitter. A number of radio time and frequency broadcasts are available, such as: WWVB, DCF-77 and MSF-60. The time and frequency radio transmissions are referenced to a precise atomic clock time source. By utilising a relatively low-cost radio receiver PC and NTP server systems have access to precise timing information. Additionally, radio antennas can generally obtain a good signal indoors close to the host time server making installation very straight-forward. However, radio broadcasts have a finite range and are generally only available to a regional audience.

The DCF77 time broadcast is generated from from Frankfurt, Central Germany. The transmission is a long-wave signal broadcast at 77.5 kHz. The transmitters are maintained by T-Systems, a division of Deuche Telecom. The time and frequency broadcast is referenced to precise atomic clocks located at the German National Standards Laboratory (PTB) in Brunswick. The DCF-77 signal can be received using a low-cost radio receiver and when decoded provides a precise timing reference for computer time servers.

The MSF radio time and frequency transmission is broadcast from Anthorn, Cumbria in the United Kingdom. The signal is broadcast as a long-wave radio transmission at 60 kHz. The transmitted time and date information is referenced to atomic clocks sited at the UK National Physics Laboratory. The transmitters are maintained by VT Communications.

WWVB is the US National Time and Frequency radio broadcast. It is a 60 kHz transmission broadcast from Fort Collins, Colorado. The transmission is referenced to atomic clocks installed at the US National Institute of Standards and Technology (NIST). WWVB provides a time and frequency reference to within 100 microseconds of the correct time and has been in operation for 45 years.

The GPS system is a Global Positioning System intended for worldwide navigation. The GPS system consists of 24 satellites in high orbit. By utilising triangulation, the GPS system can provide highly accurate positioning information anywhere on Earth. In order to calculate position, each orbiting satellite has an on-board atomic clock timing reference. Atomic time is constantly transmitted from each satellite and is available to a GPS receiver. The GPS signal is available subscription-free anywhere on the face of the Earth. By utilising a low-cost GPS antenna and receiver accurate timing information can be made available to NTP server and computer time synchronisation systems. A GPS antenna does however need to be installed ideally on a rooftop with a good 360-degree view of the sky.

The Network Time Protocol (NTP) is the standard means of achieving network time synchronisation. NTP is used to distribute accurate time around the Internet and other computer networks. The standard NTP server distribution for LINUX\UNIX is available free-of-charge under the GNU public licence and provides reference clock drivers for many radio and GPS receivers. Radio and GPS receivers are available with serial or USB ports that can interface to a PC or NTP time server to provide an accurate external timing reference. Depending on time source, NTP servers can synchronise to within a few microseconds of the correct time. Depending on network traffic they can synchronise network time clients to within a few milliseconds.

To summarise, there are a number of national and global time and frequency references available for synchronising computers and computer networks. Many can provide synchronisation to within a few microseconds of the correct time. Utilising accurate radio or GPS timing references, precise time is freely available without the expense of installing an atomic time clock.
About The Author
David Evans is a highly experienced technical author who specialises in time server and atomic clock time synchronisation systems to ensure accurate network time. Click here to find out more about atomic clock and GPS NTP server systems.

http://www.timetools.co.uk/ntp-servers/gps-ntp-server-s5100.htm

* When emails are received in Lotus Domino, a trigger in Domino should check if email has a GUID in the header and, if so, should check on the object in MS CRM with this GUID. If the object is found in MS CRM, email activity should be created in CRM and be attached to the object.

* The incoming message is processed before delivery by a Java Agent registered on incoming mail boxes (Inbox folder) and sent mail (Sent folder).

* The Purpose of the Java Agent is to transfer a kernel of communications to MS CRM (the information in the received messages such as addressees, recipients, bodies of messages etc).

* For incoming messages, it is necessary to transfer the necessary message RFC-headings for an analysis by the Java Agent. The analysis consists of checking the Subject heading (CRM: {GUID}:MRC).

* For outgoing messages, the agent should modify the heading Subject by adding a similar line at the end of the existing data. In this case, the Outgoing Message Connector will carry out message transfers from MS CRM to Lotus Notes/Domino

* The Java agent carries out the communications with Web Services Front-end by means of asynchronous calls to a web-service. This will prevent a big number of blocking methods when processing messages and it will also allow reducing the load on the Lotus Domino Server

* The Web Services Front end accepts the information and transfers it to a COM + Back end. The COM + Back end is composed of objects which carry out the communications with MS CRM using MS CRM SDK and direct access to MS CRM database

* The COM+ Back end consists of objects written in .NET C#. These objects contain methods for searching users in MS CRM, creation of Email Activity and queue updates, among others.

You can always have us to help you. Give us a call: USA: 1-866-528-0577, 1-630-961-5918, Europe: +49 231 4387600, help@albaspectrum.com
About The Author

Boris Makushkin and Andrew Karasev are consultants at Alba Spectrum Technologies ( http://www.albaspectrum.com ) Microsoft CRM, Great Plains, Navision, Microsoft RMS, Lotus Notes/Domino customization company, serving clients in Illinois, New York, Florida, Arizona, California, Minnesota, Virginia, Texas, Georgia, France, Canada, Germany, Australia, Mexico, Russia.

akarasev@albaspectrum.com

Sometimes developers try to use Crystal directly to pull data from multiple database platforms - we understand the need to give you additional tips to increase Crystal report flexibility. Also common case when Crystal report should pull data from multiple database platforms. Assuming that you are creating Crystal Reports on Windows platform - we recommend you to use Microsoft SQL Server as a database access tool.

1. Use Crystal when you have your query tuned – regardless how good Crystal Reports is - it is reporting tool - not querying environment. Tune your query first and be sure that you are pulling the right data.

2. Use SQL Server Linked Servers. Use professional tool, such as Microsoft SQL Server, where you can create linked servers to Oracle, Ctree, Btrieve, MS Access, Unidata, DB2 and other platforms.

3. Use Distributed Queries – in the case of SQL Server - research the possibilities with OPENQUERY, OPENROWSET clauses to connect to non-SQL Server linked servers.

4. Create Stored Procedures - when you need parameters to be transferred to the SQL statement - you need stored procedure. You need stored proc if temp tables are required to do intermediate job in your SQL query.

5. Report Flexibility – when you have report based on the SQL view or stored proc - if you need to modify the data query - you don’t have to change the report - you change your SQL view/ stored procedure.

6. Create SQL Query to probe the data – again - we always recommend tuning your query and see that you are getting adequate results – in any case – Crystal Report is just a nice tool to show the results of your query.

Happy customizing, querying and designing! if you want us to do the job - give us a call 1-866-528-0577! help@albaspectrum.com
About The Author

Andrew Karasev is Chief Technology Officer in Alba Spectrum Technologies – USA nationwide Microsoft Great Plains, Microsoft CRM, Microsoft RMS customization company, based in Chicago, Los Angeles, San Francisco, San Diego, Boston, New York, Houston, Dallas, Miami, Atlanta, Minneapolis, Seattle, Canada, UK, Australia and having locations in multiple states and internationally (www.albaspectrum.com), he is Great Plains Dexterity, SQL, C#.Net, Crystal Reports and Microsoft CRM SDK developer.

akarasev@albaspectrum.com

Sometimes developers try to use Crystal directly to pull data from multiple database platforms - we understand the need to give you additional tips to increase Crystal report flexibility. Also common case when Crystal report should pull data from multiple database platforms. Assuming that you are creating Crystal Reports on Windows platform - we recommend you to use Microsoft SQL Server as a database access tool.

1. Use Crystal when you have your query tuned – regardless how good Crystal Reports is - it is reporting tool - not querying environment. Tune your query first and be sure that you are pulling the right data.

2. Use SQL Server Linked Servers. Use professional tool, such as Microsoft SQL Server, where you can create linked servers to Oracle, Ctree, Btrieve, MS Access, Unidata, DB2 and other platforms.

3. Use Distributed Queries – in the case of SQL Server - research the possibilities with OPENQUERY, OPENROWSET clauses to connect to non-SQL Server linked servers.

4. Create Stored Procedures - when you need parameters to be transferred to the SQL statement - you need stored procedure. You need stored proc if temp tables are required to do intermediate job in your SQL query.

5. Report Flexibility – when you have report based on the SQL view or stored proc - if you need to modify the data query - you don’t have to change the report - you change your SQL view/ stored procedure.

6. Create SQL Query to probe the data – again - we always recommend tuning your query and see that you are getting adequate results – in any case – Crystal Report is just a nice tool to show the results of your query.

Happy customizing, querying and designing! if you want us to do the job - give us a call 1-866-528-0577! help@albaspectrum.com
About The Author

Andrew Karasev is Chief Technology Officer in Alba Spectrum Technologies – USA nationwide Microsoft Great Plains, Microsoft CRM, Microsoft RMS customization company, based in Chicago, Los Angeles, San Francisco, San Diego, Boston, New York, Houston, Dallas, Miami, Atlanta, Minneapolis, Seattle, Canada, UK, Australia and having locations in multiple states and internationally (www.albaspectrum.com), he is Great Plains Dexterity, SQL, C#.Net, Crystal Reports and Microsoft CRM SDK developer.
akarasev@albaspectrum.com

If your like many IT Administrators or Managers you may be wanting to use VMWare ESX Server for virtualizing servers but may meet resistance from others in your department or business unit. Many people are gun-shy on putting business critical applications on VMWare Virtual Servers and may resist if you try to go whole hog virtualizing these applications in the face of such resistance. We found a successful introduction to VMWare ESX and virtual servers could be done by choosing to virtualize older systems and utilize VMWare initially to perform other functions besides replacing physical hardware for critical systems.

Consolidate older systems and stand alone processes: We had many large older servers that needed to be running but were not worthy of equipment upgrades. Everyone in IT wanted these boxes out of the server room but the data needed to be available in case of an audit or historical need. Some systems performed low level processing on a scheduled basis. We imaged theses systems to VMWare ESX virtual servers using Ghost and an in place upgrade to get them booting again. We were able to run eight servers on a two processor Dell Blade server.

Use VMWare for a Development and Staging Environment: Today most organizations are moving to a structure where they utilize development and staging servers in addition to their production servers. VMWare Virtual Servers running on ESX make great environments to use Dev and Staging servers without adding a bunch of servers to your data center.

Using Citrix Metaframe Installation Manager? Use VMWare Virtual Servers for your packager. If you use Installation Manager for Citrix Metaframe Enterprise you will need a clean system running the same operating system as your Citrix server. Using VMWare Virtual servers as your packager allows you to keep a clean packager without adding hardware and rebuilding servers in your data center.

We used the ideas above to get value from VMWare ESX Server and Virtual Servers in a “safe” environment. This provided a proof of concept for the gun-shy doubters and has paved the way for us to use VMWare ESX Server in more critical environments.
About The Author

John Gall is an internet hobbiest with many web sites his most recent project is http://www.whoinventedmotorcycle.com.

I know, there are all kinds of brand new blogging scripts, forums, CMSes, and other stuff to make your site easy to update. But on the other hand there are just as many desktop-based programs that churn out thousands of static HTML files for you to put on your server.

What do you do when you have to change these files later? Edit them by hand, one at a time? I don’t think so.

You can save a lot of time down the road using Server Side Includes (SSI).

The point of using Server Side HTML is so you can include other pages on your regular HTML pages, things like a list of links you regularly update OR a header and footer.

Try it out. Take one of those static HTML pages on your site and rename it to something like “filename.shtml” instead of “filename.html.” All we’ve done is changed the “html” at the end to “shtml.”

Now, take the part of that site you want to appear on every page (the header graphic, link sidebar, etc.), copy and paste that part and save it into a file named “header.shtml.” REMOVE that text from your filename.shtml file.

Now, in the filename.shtml file, place this at the top:

[!–#include virtual=”/header.shtml”–]

(IMPORTANT: Replace the brackets [ and ] above with the greater-than and less-than “HTML” brackets… hold shift and press the “comma” and “period” keys on your keyboard.)

Upload filename.shtml and header.shtml to your server and load header.shtml. If everything went well, the page should look exactly as it did before. The only difference is: you only need that one “include virtual” line to show the top of the page, AND to change that top HTML later on all you have to do is edit header.shtml.

If you had this sort of code on 20,000 pages of your web site all you’d have to do was edit that ONE header.shtml file, and your entire site would change as soon as you uploaded that header.shtml file.

Not bad, huh?

Now do the same thing with the bottom of your web site, only save the new file as footer.shtml and at the bottom of filename.shtml place this:

[!–#include virtual=”/header.shtml”–]

Again, not that exact code… replace the hard-brackets with HTML brackets.

But wait, your fancy page generator software will only save the files as .HTM or .HTML, and specifically links to certain files. You can’t just go in and edit everything to link to .SHTML files.

Htaccess to the rescue.

Place this in a new text file:

AddHandler server-parsed .shtml .html .htm

Save that as “htaccess.txt”, upload to your web server, and rename that htaccess.txt file you just uploaded to “.htaccess”, WITH the dot in front. Now try renaming filename.shtml to filename.html. Guess what, your server side includes now work on .html and .htm files too.

Hopefully your page generator software allows you to edit the template and place these server-side include tags in there. (If not, look into a program called BKReplaceEm.) Don’t worry, the pages will look funny and broken on your hard drive but once they are sitting on your web server everything will turn out ok.

If you find yourself doing a bunch of repetitive HTML editing it might be time to look into Server Side Includes to make instant changes.
About The Author

Robert Plank

You know what else, you can use a similar bit of HTACCESS code to make PHP scripts act as HTML files. That, plus 35 more “Why Didn’t I Think of Those” secrets are revealed in my upcoming ebook: http://www.affiliatebattleplan.com

articles@jumpx.com

For this example we’ll have two servers, one named Johnny and another named Cash. Both are running openssh. Our goal is to have a program on Johnny login to Cash and run a program on Cash. To make the task a little more complex we’ll be using different users on each machine.

The first thing we’ll need to do is generate public and private keys on Johnny. So, logged into Johnny as user ‘boy’ we create public and private keys by creating them in the .ssh directory as follows:

Johnny$> pwd
/home/boy/.ssh
Johnny$> ssh-keygen -t rsa -f sue
Generating public/private rsa key pair.
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in sue.
Your public key has been saved in sue.pub.
The key fingerprint is:
8d:e9:c0:g1:c7:1f:e3:b3:2f:38:12:aa:b5:3b:2e:b3 boy@Johnny
Johnny$>

In the example above we picked an arbitrary name, sue, to identify the files that hold the generated keys. When prompted to enter a passphrase (twice) we simply hit enter twice.

As the output of ssh-keygen indicates, the public key has been saved in a file named sue.pub.

The output of ssh-keygen implies, but doesn’t indicate directly, that the private key is in a file named sue (Yes, a user named boy created a file named sue.)

Johnny$>ls -l
-rwx—— 1 sue suegrp 887 Oct 17 14:27 sue
-rwx—— 1 sue suegrp 223 Oct 17 14:27 sue.pub

The private key file, sue, will remain on Johnny for the reaminder of this exercise, but the public key must be moved to the remote server, Cash. Note that the .ssh directory itself, as well as the sue and sue.pub files should have permissions of 700.

Now you’ll need to ftp the sue.pub file from Johnny to Cash. The user on Cash that we’ll login as is user ‘named’.

Johnny$> ftp Cash
Connected to Cash
220 Cash - Propery of Xyz. - Authorized users only
Name (Cash:boy): named
331 Password required for named.
Password:
230-Last unsuccessful login: Fri Oct 17 13:12:55 2003 on ftp from Johnny 230-Last login: Fri Oct 17 16:02:11 2003 on /dev/pts/1 from Johnny
230 User named logged in.
ftp> cd .ssh
ftp> mput sue.pub
mput sue.pub? y
200 PORT command successful.
150 Opening data connection for sue.pub.
226 Transfer complete.
224 bytes sent in 0.000781 seconds (280.1 Kbytes/s)
local: sue.pub remote: sue.pub
ftp>bye
221 Goodbye.
Johnny$>

Now we’ll telnet into Cash and concatenate the sue.pub file into /home/named/.ssh/authorized_keys file.

Cash$> pwd
/home/named/.ssh
Cash$> cat sue.pub >> authorized_keys
Cash$> exit
Connection closed.
Johnny$>

Let’s recap what we’ve done so far.

1) We’ve created public and private keys on Johnny.

2) We’ve ftp’d the public key file, sue.pub, from Johnny to Cash.

3) We’ve telnetted to Cash, and concatenated the contents of sue.pub into authorized_keys

We’re now ready to manually login from Johnny to Cash using ssh.

Johnny$> ssh -i /home/boy/.ssh/sue named@Cash
The authenticity of host ‘Cash (xxx.yyy.zzz.aaa)’ can’t be established.
RSA key fingerprint is 65:11:7d:ef:ed:a3:cc:34:d1:b5:ba:c9:16:22:31:23.
Are you sure you want to continue connecting (yes/no)? yes
================================================================
*** NOTICE TO ALL USERS ***
================================================================
Cash$>exit
Connection to Cash closed.
Johnny$>

Now on Johnny, create a shell script called ‘boynamedsue.sh’ with the following one line of contents and chmod the script to 777.
ssh -i /home/boy/.ssh/sue named@Cash /usr/bin/ls -l

Next, execute the script on Johnny.

Johnny$> boynamedsue.sh
================================================================
*** NOTICE TO ALL USERS ***
================================================================
-rwxrwxr-x 1 named namedgrp 10020 Oct 17 14:35 namedfile1.txt
-rw-r–r– 1 named namedgrp 680 Aug 14 16:18 namedfile.html
-rw——- 1 named namedgrp 1148 Aug 18 09:51 mbox
drwxr-xr-x 2 named namedgrp 512 Jun 17 13:38 old
Johnny$>

You just executed a program on Johnny, that logged into Cash and ran a program (unix ‘ls -l’).

The next step you’ll want to take is to replace the ‘/usr/bin/ls -l’ command in the boynamedsue.sh program with the path and name of the program that you want to run.
About The Author

C.S. Deam is a small business owner. His eBook Computer Nuggets: Non-Techie Internet Tips For In-Laws, Out-Laws, and the Rest of Society is a great gift for non-techie family members and is available at www.LinkertonPublishing.com where you can sign up for FREE E-Courses & Newsletters to help you on your path to self-employment.