Flexible Web Services Info Tech Support

Everyone probably already knows this, but I just discovered it.  I frequently  high-light my material and copy it into the windows clipboard when I go to save my work. To do this, I press Ctrl-A to highlight everything, then Ctrl-C to copy.  That way, if I lose the material when the save doesn't work, I can just paste it back in with Ctrl-V.  If the program won't open again, I can paste my work into almost any other application and I can save it in that software.  

Doing this over and over often means that there is something in the buffer already and as I sometimes do, I will press the wrong key.  That will destory the contents that is in the high-lighted area and since I didn't get chance to copy it, it's lost.  Gone forever.  Or so I thought.

I have long known that Ctrl-Z is the short cut in most programs that means Undo.  However, if it's not on the menu, I assumed it wasn't available.  Chrome doesn't even have a menu bar which compounds the problem.  Right clicking usually opens a dropdown menu for options in the area where you right-clicked.  It also didn't have the Undo option.  

Well, I just lost my data a few minutes ago, by accidently pressing Ctrl-V after I had highlighted the area and lost all my changes and in it's place put data from a previous cutting.  That was frustrating and I was ready to close the window and I decided to try the Ctrl-Z anyway.  And guess what???

It worked!

On my site, I installed the BP Blog 9.0 package into it's own subdirectory and only put a link to the blog from the main page.  The intent was to later incorporate content from the blog into the pages on my main site.  My first project was to add the 10 most recent subjects to main page as hyperlinks.  The hyperlinks would open the blog page and display the entire article exactly is it would appear if opened from the main blog page.

In order to do this, I had to open the database and pull the subject lines for each article and sort them in date order, most recent being at the top.  I chose to do this with a saved query or view.  As I mentioned in a previous blog, I used Postgresql database backend on my installation off BP Blog.  Further more, the database server is located at a separate site from where the web server is.  This doesn't change anything in my project, I am just describing my working environment.  I get a lot of flack for setting up database servers separate from the web server, but I have never seen any problem with it.  When you are viewing this site, the config is just as I described my working environment here.

I created the view to pull the ID and blog title from the main blog data base and sorted it by date, limiting the selection to include only the articles that were not in draft.  Once I had this view saved, I was able to access the view just like I would a regular table.  I chose to pull all of the records and limit the number of records in the asp script.  Limiting is done differently in different databases, for example, in MS SQL it would be "top 10" and in postgres it would be "limit 10"

I then opened one of the asp pages that is normally used to display the blog text, default.asp (the one in the theme directory) works fine.  I copied the lines from the top and bottom of the asp scripts and pasted them into the appropriate places in my main page file.  This provided the connections file and the initial database pull. Then I replaced the SQL query text with my select statement which pulled from the view I had created and limited the output would be the 10 most recent entries.

I already had a list of projects on my main page which I was going to replace with this list.  The primary difference being that the new list would change everytime I added an article to my blog.  With this change my main page would now have changing content.  It's the most frequently hit page of all my pages and also the most frequently abandoned page.  The bullet format was already there, I just needed to replace the text and make them links.  Something like this is done in the blog pages, so I opened default.asp again (same one in the themes folder) and extracted the loop script.  I cut out the code that displayed the summary and changed the loop criterion to list everything instead of using the Dreamweaver pagination script that is in the default.asp.

That completed my project.  The only problem I ran into, besides typos in my editing was understanding how the script worked to pick the themes in BP Blog 9.0.  The main blog directory contains files that have exactly the same name as the themes directory.  The asp files in the main directory only open the database to determine the theme and then re-direct to the same named file in the theme directory.  Since my package was installed in a separate subdirectory, I had to modify the script copied to include the subdirectory I copied it to.  For example:  

<a href="/blog/template_permalink.asp?id=<%=(rsBlogSite.Fields.Item("BlogID").Value)%>

The first run I did, didn't work because I left it without the directory in bold.  I then put a link directly to the subdirectory which turned out a page without any of the CSS formating.  That didn't look good.  So I use the default.asp page to get the theme info and set the session variable in my main page, but that wasn't needed.  Each same named file in the blog directory also determines the theme you are using and sets that session variable and then calls executes the correct file in your themed directory.

That completes this project.

I just finished reinstalling a laptop and needed to install the Verizon cellular modem.  I found the drivers online at http://www.vzam.net/download/supported.aspx which lists several models.  The one for the USB727 was downloaded and installed, but every step of the way, I got pop ups that indicated it couldn't find some part of the hardware.  I have always installed USB device software first, then inserted the USB device.  However, this time, that apparently doesn't work.

I'm not sure if it would have recoginized the installation if the usb727 was inserted or not, but after it was all installed, It still didn't recognize the device.  I looked at the device manager and found 4 items under other that were not recognized.  They were data ports. 

After researching I found this site: http://www.evdoforums.com/thread7612.html which had basically the same problem I had.  I right clicked each unrecognized item in other and selected the auto install.  It found the drivers and installed it for each one.  There are 4 separate dataports. 

And, no, there is not software on the USB for this model.

Found the solution to this problem at the Bullzip forums.  See the link below.

 http://www.bullzip.com/phpBB/viewtopic.php?t=6635

The problem was to change or add a setting to settings.ini, but there was no indication where to find it.  The above article shows it's in

C:\Documents and Settings\<user>\Application Data\Bullzip\PDF Printer.

Cargowise has a split in upgrade paths depending on the SQL server you are running.  We were left out of a new upgrade because we were using SQL 2005, so we decided to make the switch.  The instructions from Cargowise EDI for the migration simply are:

1. backup the database
2. copy the backup files to the new server
3. Install 2008 R2
4. restore the database
5. Re deploy the Controller and web interface.

Their instructions start with installing 2008 server first.  These didn't work very well for me because I'm limited on servers and I didn't want to install SQL 2008 on a server that already has SQL 2005 installed on it.  I rearranged the steps to those listed above, but added to uninstall SQL 2005 after the backup was done.

The backup software used to backup Cargowise is the one provided by Cargowise.  It's found on the SQL server in the C:Program FilesCargoWise ediediEnterprise ServerDistributionApplication directory.  To correctly access the database, the server/instance had to be provided in order to list the databases.  It shows all of the tables in the database, but there are no instructions on which table to backup.  So I selected the main table, in this case enterprise.  It performed a backup of every table.

I had severe problems uninstalling SQL 2005 because the original install path was missing.  I had archived the installation files for Cargowise after the installation was complete.  Once the uninstall started, it did not role back after the abort and left me with a half uninstalled SQL 2005.  I couldn't reinstall it from a new download of SQL 2005 because the installation was still looking for the original install path.  So I had to go through and manually remove the components of the old server.  The installation for SQL 2005 exposed the installation log which showed the conflicting key.  Removing those keys only made the installation worse.  After further reading, I found that the reverse of the first portion of the key also needed to be removed.  So if the key was {432832-43233d32f4-2342234}, I had to find the keys that started with 238234.  I checked the folder to make sure I was removing the same information as the original key problem and that info matched.  I also removed the parent folder where ever I found either key.

 Once the SQL 2005 was uninstalled, I was able to install SQL 2008.  However, there were several pre-requisites that I needed to install before I could continue with the 2008 install.  One was an updated installer and the other was powershell which I don't know why it wasn't included in the SQL 2008 installation package.

After installing the database, I restored from the backup.  It went farily smooth, however; I changed the name of the main table from enterprise to edienterprise.  That became a real problem because no one could connect.  I thought when I specified the table to restore that I was entering the instance name.  To fix that problem, I had to go to ediload.ini and change the name of the table there.  On the server it was located at: C:Program FilesCargoWise ediediEnterprise Server.  It was in a slightly different location on the local network.  Once this was done everyone could get back.  

The final steps were to re-install the Controller, though I probably wouldn't need to do this if I kept the database name the same.  re-installing the controller is done at the workstation that has this service installed.  Go to the Admin menu and select scheduled tasks.  In the upper portion of the screen where you normally restart the controller task, you will see those options greyed out.  The only option is install.  I didn't want to press that because everything should have been the same.  Installing it again actually put two enteries on my services list, one as controller for the old database and one for the new.  Only one was running.

Then I went to the web server and ran ediload.exe -web deploy, but that wasn't enough.  I later had to go into the IIS settings and into the folder we setup for track and trace.  Select properties and then select the ASP.NET tab, then click Edit Configuration...  In application settings, the lower window has the database name in it.  Double click it and correct the name.

To-date, we stil have one major problem.  The backups are not running.  Everynight I get an error that says the user cannot login (it was the same error when I was trying to connect before the database name was changed, but the database name here is correct.)  As soon as I get more info I'll correct this post.

The Controller PC manages all the print jobs in Cargowise EDI.  Printers that are directly connected to the Controller PC are available to all the workstations.  The printers do not have to be setup on each workstation.  Cargowise EDI has it's own interface for selecting the printer.  If you have a shared network printer, it will not automatically be seen by Cargowise EDI.  The Controller software runs as a service.  In order for the printers to be seen, they need to run with the same permissions as the Controller service.  I'm not sure you can setup printers up to work in the Local Service user group, so what we did was make the Administrator account the one that runs the Controller service.  Then to add a printer to Cargowise, we just log in as the Administrator on the Controller PC and add the printer to that account.  Then log out and reconnect with the regular users creditials.  If  you haven't rebooted to make this switch, you will need to restart the Controller service.  Then when you print a job from Cargowise, you will see the newly added printer.

I discovered today that \'s won't save in the database.  Right now I have to double it to make it work, but will need to modify the code to replace all \'s with double so I don't have to do it and so it will save it every time. 

 I followed the suggestions on this site: http://supportforums.blackberry.com/t5/BlackBerry-Desktop-Software/8830-Syncronization-Error-code-0x8004fceb/m-p/44843 which recommended deleting the intellisync folder from the appsdata folder, however; those instructions were for a Windows 7 PC, possibly a Vista machine also.  This is the path I used to fix the error on a Windows XP PC with Outlook 2007 installed...

c:\documents and settings\<me>\application data\Research in Motion\Blackberry.  

The intellisync folder is in that last folder.  Remove the folder and everything inside it.  When you start the Blackberry Manager and connect the phone, it will re-create the intellisync folder.  You will have to setup the folders to synchronize again.  After doing that the error went away.

It was irritating me that once I un checked remember me, that I had to keep un checking it.  So I looked in the code.  On line 176, the input line for the checkbox has this string in just before the conditional check of the cookie: 

checked="checked" <%If... 

I couldn't find any documentation for this attribute in the <input> html.  However, if that checked key word  is present, the word "CHECKED" provided in the conditional statment will never be used.  Taking checked="checked" allows the login screen to remember that I don't want it to remember.

Getting a Unable to load the 'ThemeStrings.english-utf8' language file. error message when I try to access the forums on slapbackguitar.com.  This is the stable forums package being offered by the same guy that wrote bp blog.  Not a secure feeling.  I can't tell if it's a hack or an error in design.

Hope it's backup soon.

Brighttree is a web-based DME/HME Billing application.  My customer is currently using the app to handle the shippment and billing for medical supplies to individuals covered my military medical plans.  

My customer contacted Pitney Bowes to install Send Suite Shipping to handle their shipping needs.  Unlike other Pitney Bowes products this one is not stand alone and not easy to setup.  A product specialist came out to setup the computer (which we needed to provide) and connect the hardware to do the shipping.  The requirements for the computer were Windows XP (won't run on Windows 7), 2 Gig Ram, 40 Gig HD, Network and internet.  The internet was needed to update the shipping costs and transmitting info to the various carriers they support.  The network was required to interface with the billing software to get the shipping addresses from the system.  However, Brighttree is web based and has no direct connection to the database.

The product specialist called me to find out the odbc connection info to Brighttree, which doesn't exist.  So we called Brighttree.  Our solution was to run an AD HOC report to extract the customer database to a file.  At first we tried exporting to excel, which didn't work.  For some reason, when we sent the Excel file via email to someone, they couldn't open it.  They'd get a message that stated ... file.htm is not available.  The location of that file was in the originators temporary internet files folder.  That didn't make any sense to me, but apparently the original web page was embedded in the exported Excel Spreadsheet.  I didn't want to use the Excel Spreadsheet anyway.  The zipcodes or any number field with a leading zero gets truncated in Excel if no special formating is provided.  i.e. 09273 becomes 9273 in Excel.

The easiest file to create was the CSV file which can also be opened with Excel.  I'd recommend leaving it alone though.  Excel could possibly remove the leading zeros on the zipcodes if the file was saved.  We're still waiting on Pitney Bowes to give us the goahead and start using the CSV file.

This site had some installation instructions for pure-ftpd.  Note line 3 has spaces between the hypen and words in pure-ftpd.  Do not put the hypens in.  Yum will not download the correct package.  I've made some modifications to the instructions to give additional guidance.  See: http://forums.fedoraforum.org/showthread.php?t=147172

Note also that pure-ftpd does not allow for matching the date stamp of the source files.  This may be critical because if you restore those files, all the dates will be as of the date you FTP'd them to your storage location.  I believe this will also cause problems with design packages like Dreamweaver.

1. install pure-ftpd using yum: yum install pure-ftpd

2. run: /usr/sbin/pure-config.pl /etc/pure-ftpd/pure-ftpd.conf (this will create the default conf file for use with pure user database

3. start service: service pure-ftpd start

4. create the pure user database: pure-pw mkdb

5. create users on the fedora box: useradd <uname> and create a password for that user userpw <uname>

6. create a pure-ftpd user account: pure-pw useradd <uname> -u <uname> -g <usame> -d /home/<uname>

Running this command will prompt you for a password.  I used the same as the name I setup in linux, however; I think it can be different.

You can also check user info in the pure-db file by typing pure-pw list.

Note: installing pure-ftpd does not configure iptables to allow access on port 21.  You will have to do this manually, I don't have the notes for that here yet, but to test, I just turned off iptables: service iptables off