I am not much of a writer and I never will be, but here is a solution that I searched for long and hard.  It fixes problems when using automysqlbackup.sh to backup all databases, including “information_schema” and “mysql”.

I had to change this line:

OPT=”–quote-names –opt”

to this:

OPT=”–quote-names”

There is only one other change needed; I had to add an -o (or) statement, just replace this section:

# Database dump function
dbdump () {
touch $2
chmod 600 $2
if [ $1 = “information_schema” -o $1 = “mysql” ] ; then
NEWOPT=”–skip-opt –skip-lock-tables ${OPT}”
else
NEWOPT=”–opt $OPT”
fi

if [ -z “${USERNAME}” -o -z “${PASSWORD}” ] ; then
mysqldump –defaults-file=/etc/mysql/debian.cnf $NEWOPT $1 > $2
else
mysqldump –user=$USERNAME –password=$PASSWORD –host=$DBHOST $NEWOPT $1 > $2
fi
return 0
}

Kaltura 3.0 Storage Locations

Posted March 17th, 2011 by bryanr

The Kaltura open-source community is in desperate need of support from the people who are now profiting from their open-source software.  As you search through the Kaltura forums, you will continue to find unanswered posts with basic questions about software.  Kaltura may be useful software, but they need to continue to support the open-source community that got them to where they are today.  With that being said, I would like to provide some new configuration information for KalturaCE 3.0.

In versions previous to 3.0, a user specified directory called myKalturaStorage was used for all uploaded and converted media.  This location is no longer used by Kaltura 3.0, well it is not mentioned in the installation guide, which appears to be copied and pasted from the 2.0 instructions with a few omitted sections.  Kaltura mentions that the 3.0 version has an upgrade path from 2.0, but the upgrade instructions are not available.  So, I assume that the myKalturaStorage directory is used for legacy media when an upgrade is performed.

In the end you are left wondering where the media is stored for Kaltura 3.0, since it is no longer a user specified option.  It is now located here:

<kaltura_root>/web/content/

If you are like me and don’t want to worry about your system partition filling up, your easiest option is to move the content to another partition and then create a symlink from the original folder.

A co-worker recently shared this article.

http://blog.hubspot.com/blog/tabid/6307/bid/9481/Top-6-Reasons-You-Should-Fire-Your-Website-Designer.aspx

I have long preached that even ugly websites can sell items. The key is the items for sale, whether E-book’s Information or other products. If people like it they will buy, no matter how ugly the site. While I agree with the article there are two points I have a varying opinion on.

First, in my experience, the lack of clear goals and outcomes for the site is usually what creates the perception the designers holding the site hostage. Many organization form committees to tackle the re-design of a site and the end result is months and months of setbacks and internal battles of which department wins the front page. A Designer often low on the corporate food chain is caught in an internal struggle of doing what is a right and what they are told to do. Which may in cause them to fall prey to some of the no no’s in the article.

Secondly, the article implies that Banner Images are a failure. Hmmm, let me think for a moment. Is that not what AdSense uses? If graphic banners do not sell you or your product. I think you are mistaking. Users eyes are first drawn to the top of your page, a well designed inticing banner will draw and keep users attentions and usually lead them to the rest of your site.  So in this case aesthetics do make a difference.  What if you are an Art Gallery or Artist? Isn’t the “web” a representation of yourself and your work. If you throw up a website from the late 90’s, all text and hyperlinks. Do you think it will help draw your customers?

SEO is important and I am not suggesting that you throw it out the window for a pretty site. However, you can accomplish both.  Anyone aware of a hack going around that had been affecting WordPress and other websites, that when search brought up information on gambling startegies?  All the rest of the relevent information on your page was basically rendered hidden. Which means entire page and source optimization may in fact be more relevant to SEO than the Graphic’s banner or the newly formatted <h1> tag with your Companies name.  So web designers do not panic your jobs are safe for a while. Well except for the CMS’s out there, but that is a topic for another article.

Now let’s talk SEO briefly. The idea of throwing meta-data on a page and winning the war is not going to work. Review everything from hidden source tags, related information, cross links, etc. As a company using SEO as tools to create brand awareness, and lead generation consider this. If you are investing the in the stock market you hire someone or pay very close attention to the ups and downs in the market. The SEO expert will do this too.  They will monitor rankings and keywords and find you the best result for your market. Be aware the best keywords for your industry might not be exactly what you think as an insider.

Just something to consider.

One small project that has been lingering on my list is to setup vpn for our internal network and allow employees to authenticate with their ldap credentials.  I decided to use a PPTP vpn server because OSX has that client built-in to the OS, which makes things easier for me.. and everyone else.  The PPTP server I have setup to use provides two auth methods: local database of users or a radius server.  After some reading, I found you can setup a freeradius server to query a specified ldap server for user accounts… sweet!

Setting up freeradius on an Ubuntu server is very simple if you use apt-get.  So that’s what I did, and after tinkering with the server settings for the good part of a day I was still having problems successfully authenticating to the server with my ldap credentials.  There is a test utility that comes with freeradius called radtest and makes testing your login very easy.  The error logs were of no help either.  So I gave up.

Not really, but I went to a Jerry for ideas and he said “dude, I think there is a radius server built into OSX server”.  Come to find out, there is a freeradius server pre-installed in OSX 10.5 and 10.6 servers, but out of the box it is setup to manage authentication for airports on your network.

After some searching on the net, I found that Apple’s install of radius is actually a complete freeradius server that has been configured to Apple’s liking.  By default, the radius server tries to authenticate users against the /etc/passwd file, but that can be changed to auth against open directory very easily.  You will need to change the following in /etc/raddb/users.

DEFAULT Auth-Type = System
Fall-Through = 1

Change this to:

DEFAULT Auth-Type = opendirectory
Fall-Through = 1

And since our LDAP is already running on OSX 10.5, all I had to do is make the change and start the radius service.  I used the radtest script to verify it was working properly.

To allow our VPN server to auth against the new radius server I had to add the VPN server address to /etc/raddb/clients.conf as an allowed client.  I finished setting up the VPN server with the new authenication method and now all I have to do is train our staff how to use our new VPN connection…

If you have problems, you are able to enable some extra logging features for the radius server.  Run the following commands:

$ sudo radiusconfig -setconfig log_auth yes
$ sudo radiusconfig -setconfig log_auth_goodpass yes
$ sudo radiusconfig -setconfig log_auth_badpass yes

Update:

With the release of Mac OS 10.6, you no longer have to edit the /etc/raddb/users file.  The built-in freeradius server looks at open directory for users by default.  You still need to edit the clients.conf file according to the included instructions.

If you happen to have multiple certificates on your server, you will have to manually choose the certificate under the Radius service Settings.  If you do not, the configuration assistant will open when you try to start the service, which requires you to add Airports to continue.

MySQL database exists… but wait, no it doesn’t exist?

Posted January 13th, 2011 by bryanr

I recently updated our Mysql servers to version 5.1 and moved the installation to a 64 bit servers.  Afterward, I had some minors problems that led me to run the “mysql_upgrade” script on one of the servers.  Everything seemed functional after running the script, which checks the databases for incompatibilities.  Here is an explanation:

http://dev.mysql.com/doc/refman/5.0/en/mysql-upgrade.html

But no, one database was completely empty after performing the check.  So, I tried to restore from my handy backup.  Something strange happened and it created a new database with the same name instead of overwriting the existing schema.  I decided I wanted to remove both of the databases of concern and re-import, creating new schema.

When I tried to drop the database, I received the following error:

“Can’t drop database ‘dbname’; database doesn’t exist”

And when I tried to create a new database, it told me:

“Can’t create database ‘dbname’; database exists”

That just didn’t make any sense!  And the Mysql Administrator application still listed the database in the Catalog list.

The fix was simple; navigate to the /var/lib/mysql/ directory and remove the folder for the problem database.  After that, I was able to restore the database from my backup and everything worked.

To work around this behavior and display devices when you click Show hidden devices:

  1. Click Start, point to All Programs, point to Accessories, and then click Command Prompt.
  2. At a command prompt, type the following command , and then press ENTER:
    set devmgr_show_nonpresent_devices=1
  3. Type the following command a command prompt, and then press ENTER:
    start devmgmt.msc
  4. Troubleshoot the devices and drivers in Device Manager.NOTE: Click Show hidden devices on the View menu in Device Managers before you can see devices that are not connected to the computer.
  5. When you finish troubleshooting, close Device Manager.
  6. Type exit at the command prompt.Note that when you close the command prompt window, Window clears the devmgr_show_nonpresent_devices=1 variable that you set in step 2 and prevents ghosted devices from being displayed when you click Show hidden devices.

This is an excerpt from http://support.microsoft.com/kb/315539

Our organization hosts a Blackboard Learn server, but I have never really been tasked with maintaining the software.  I was recently asked to install the 9.1 update on the Blackboard test server so we are prepared for the production upgrade in a few weeks.

Let’s just say that I am not sure I want to be the full-time Blackboard admin.  In short, the installation failed when it reached the content system step.  It asks for a password for the Blackboard file storage database user password, which is pre-populated and hidden, and a content location.  But when you click next to continue it reports and error stating “Password is too long.  Password must be 20 characters or less”.  If you try to enter a shorter password, you just get the same thing.  I searched high and low in the Blackboard knowledge base, but in the end I found no resolution.

To fix the problem, you need to edit the bb-config.properties file in you Blackboard installation.  Change the following line to reflect your system user password:

bbconfig.cs.db.cms-user.pass=_UNUSED_DEFAULT_BY_INSTALLER_

It is hard to believe the I was unable to find a solution for this simple problem on the Blackboard support site.  I guess that’s why they want you to pay for support.  Bryan

View processor info on a linux OS

Posted November 10th, 2010 by bryanr

Curious to know if your linux install is recognizing all of the processors that you have allocated to it?

Run cat /proc/cpuinfo to list all of the processors.

There have been times that I have added processors to a virtual machine after it is created, and wondered if the operating system applied the change.

Google Logo

Posted September 6th, 2010 by admin

I realize today was Labor Day, and the staff at Google need the day off too. I was just shocked to hit the Google site today and noticed the Logo was the standard Google Logo. No flag hanging from the “l” no people napping in the “o”‘s nothing. After a quick mouse move I saw an American Flag on the lower part of the page. Hmmm, wonder if it was budget cuts? I wonder if this is a sign of the times. Is Google afraid to offend someone who does not celebrate Labor Day, or is it becuase it is not celebrated in all countries. Will we see anymore creative spins on the logos, I hope so it is refreshing.

Good old fashioned rant – re. iPad

Posted May 17th, 2010 by admin

I’ve been pretty bad about posting here, I’m going to strive to do better. I give up on the idea of just posting new, unique, and insightful stuff.  I’m gonna go for a good, old fashioned, blog rant!  Ok.. subject for the day – the iPad.

Very cool little bit of tech that probably is totally game changing. Revolutionary, Apple says.  We’ll see.  I’ve had a bit of time on them, quite impressed. I really think they will be perfect for consuming content, and to a degree creating content as well. Thats all  the computer power most folks will ever need.

I heard somewhere that people were needing to get from the ground floor to the third floor metaphorically, and we geeks gave them a jet fighter to get there…   what they really needed was an elevator!  In my years of supporting tech,  I’ve noticed that most users, God love ’em, really don’t “get” it…  they go through the motions, they basically know some behaviors that get their work done, but they have no idea what they are really doing.

So heres the meat of the rant.  People malign the iPad for many reasons, but I’ve heard two mentioned often – no file system that you can see at least, and no printing.  At first as a techie, I thought  – Apple this is stupid, omg this isn’ t a real computer!   Thinking as a techie…  which most vendors do.  Ok.  No file system – GOOD!  Many users don’t know how to navigate and place and retrieve files anyway!  Go ahead and abstract it away, thats a good thing for the user.   The biggie, no printing.   As mentioned, I have many years of tech support and the scars, mental and physical to prove it…  The problems that deeply scar my psyche are from dealing with PRINTING!!!!   WHY DO WE STILL NEED TO PRINT STUFF????   Sorry for yelling…  anyhow, no printing, GOOD!!!  There are dozens of ways to get information shared with others.  We should try to end printing on other platforms as well!

By now you realize the depth of depravity my tortured tech soul is capable of…  this may be a rant, but think about it.  I’m sure the file system “Problem”  and the printing “Problem” will be addressed soon.  My $.02 – it’s ok as it is.