My iGeneration

Unconfigured Named MSSQL Instance

2012-01-10T09:53:00.002-05:00

If your installation errors out half way through, you may wind up with an unconfigured named instance. Short of uninstalling all of Microsoft SQL, you'll have to manually remove the instance in order to re-use the name.

Use regedit to find the following key:
HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Microsoft SQL Server

Under Instance Names\SQL, delete the value corresponding to your unconfigured instance. Now look for the corresponding MSSQLxx_xx.InstanceName key and delete it.

You will also want to delete the corresponding instance folder, the default path would be C:\Program Files\Microsoft SQL Server\MSSQLxx_xx.InstanceName.

Rooting the Coby Kyros

2011-12-18T01:41:00.000-05:00

This was tested with the MID7022. This newer version of the Kyros comes with Android 2.3 and APK Installer pre-installed. If you have a different MID version, stop at step 7 and refer to CyanogenMod site for correct ClockworkMod image file. I'm using Windows 7 for the following:

Install Android SDK per CyanogenMod instructions.
Replace the contents of C:\android-sdk-windows\extras\google\usb_driver with the Telechips, Inc adb driver.
Connect your phone to your computer via USB, put your phone in USB debugging mode and do not mount SD card to your computer.
Install SuperOneClick and click Root.
Once complete, remount Nand and then SD card in Settings.
Make a backup of your current recovery image:

Download dump-recovery-signed.zip and save as update.zip on your SD card.
Turn off USB storage and go to Settings-About device-System Updates-Install system update. Choose from SD card.
The stock recovery will automatically be dumped as dumped_recovery.img and /proc/mtd will be copied as dumped_proc_mtd.txt on root of SD card. Keep them in safe place.

Install ROM Manager (ClockworkMod) from AppsLib. "Flash ClockworkMod Recovery" option does not yet work for Telechips, so we'll flash it manually:

Download KYROS7022-8K-recovery-signed.zip and save as update.zip on your SD card. Do the same thing as you did in step 6 to install.
Open ROM Manager and choose Reboot into Recovery. Allow the superuser request. To navigate the recovery screen: MENU- = next(down), HOME/POWER = select, BACK = back
Go to Backup and Restore - Backup. This will create a backup of your entire system image onto your SD card. When it's complete choose Reboot system now.

Because CyanogenMod doesn't yet support a bunch of features (including the Kyros' strange software-based volume up/down control), I'm keeping the (rooted) stock image but I'd rather have Google Market. Here's how to install it:

Download Market.rar. Extract the contents of the Market folder to your SD card.
Remount your SD card to your tablet, open APK Installer and install in the following order:

OneTimeInitializer.apk (the one with the blank name)
SetupWizard.apk
GoogleServicesFramework.apk
Vending.apk (name shows as Market)

Turn your tablet off and turn it on again.
Next time you press the home button, a window saying “Complete action using” will appear, mark “Use by default” and choose the “Launcher”. Also, the camera shortcut on your home screen will stop working, all you have to do is delete it and re-add it.

Once you have Market on there, you'll find some things still missing. Here's a link to the latest Netflix app (1.5.2 as of 12/25/2011), Adobe Flash (10.3.186.3 as of 12/25/2011) and Pandora.

Renaming the root volume group in RHEL6

2011-12-13T10:30:00.002-05:00

renamevg vg_oldname vg_newname

edit /etc/fstab to reflect the change
edit kernel command(s) in grub (/boot/grub/menu.lst) to reflect the change

How to Remove Carrier IQ from your HTC Phone

2011-12-10T13:48:00.001-05:00

Carrier IQ is all the rage these days. Here are the simple steps to remove it from your HTC phone.

Root your phone. Google for "how to root xyz" where xyz is your phone make and model. For more HTC phones you'll use unrevoked or revolution.
Install Titanium Backup.
Go to the Backup/Restore tab.
Scroll to HTC IQAgent and tap it.
Press "Un-install !" button and confirm.
Do the same for IQRD.

If you really want all traces of it gone, additionally you should do the following:

from /system/bin, remove 'iqd' and 'iqfd'
from /system/lib, remove libhtciqagent.so, libciq_htc.so, libciq_client.so
from /system/etc, remove iqprofile.pro
from /data/data, remove com.htc.android.iqrd and com.htc.android.iqagent.

You should probably then clear the dalvik cache by booting into clockwork mod.

Export Users, Passwords, Server Roles from MSSQL 2005 to MSSQL 2008

2011-11-17T14:08:00.001-05:00

Microsoft provides a useful script to export users and passwords between the two versions: http://support.microsoft.com/kb/918992

But it does not transfer the server roles. To do that, you'll need to use the following script:

DECLARE @cmd varchar(max)
DECLARE @SERVERROLE VARCHAR(100)
DECLARE @MEMBERNAME VARCHAR(100)

CREATE TABLE ##SRV_Roles
(
  SERVERROLE VARCHAR(100),
  MEMBERNAME VARCHAR(100),
  MEMBERSID VARBINARY (85)
)

/*GET SERVER ROLES INTO TEMPORARY TABLE*/
SET @CMD = '[MASTER].[DBO].[SP_HELPSRVROLEMEMBER]'
INSERT INTO ##SRV_Roles EXEC (@CMD)

DECLARE SERVER_ROLES CURSOR FOR
   Select SERVERROLE ,
     MEMBERNAME
   FROM ##SRV_Roles

OPEN SERVER_ROLES
FETCH NEXT FROM SERVER_ROLES into @SERVERROLE,@MEMBERNAME

WHILE (@@fetch_status =0)
  BEGIN
   Set @CMD = ''
   Select @CMD = @CMD + 'EXEC MASTER.DBO.sp_addsrvrolemember @loginame = ' + char(39) + @MEMBERNAME + char(39) + ', @rolename = ' + char(39) + @SERVERROLE + char(39) + char(10) + 'GO' + char(10)
   --from ##SRV_Roles --where MemberName = @DatabaseUserName
   Print '--Login:' + @MEMBERNAME
   Print @CMD
  FETCH NEXT FROM SERVER_ROLES into @SERVERROLE,@MEMBERNAME
END

CLOSE SERVER_ROLES
DEALLOCATE SERVER_ROLES

Drop table ##SRV_Roles

Be sure to comment out the very first login for 'sa' as you cannot modify its roles.

Source

IBM BladeCenter Cisco Switch Default Password

2011-09-28T14:43:00.003-04:00

For anyone else looking for this: the default credentials for the switch's web interface is no username, password cisco. You can then set a telnet password on the Advanced Settings tab of Configure-Express Setup. And set the enable password by going to /configure/-/enable/secret/0/SomePassword/CR where SomePassword is what you want to set it to.

Tracing the History of your Indiana Home

2011-09-18T01:54:00.001-04:00

Tracing the history of your Indiana Home is possible, it just requires a little more detective work than in some other states. The place to start is your county government's GIS portal. This is usually listed somewhere on the county website. On the GIS portal, you can type in your address and determine your parcel ID, which you'll need later on in the process. In the Improvements section, the year the government believes your house was constructed will be listed. For older homes, it's probably a ballpark figure. Hopefully there's also a Transfer History section which will show past owners and most importantly, the book & page at the county recorder's office that the deed transfer was recorded on. Now that you have that information, it's time to head down to your local county government offices.

In the Plat Mapping office, you can give them the parcel ID and they'll give you a copy of a card on the back of which might have additional past owners. But the info from the card is probably just what was on the GIS portal. They can also give you the property's "legal description" if you don't already have it, which will help you when you're identifying deeds in the next step.

Now in the Recorder's office, you can look up the oldest book & page you know of. Hopefully your County Recorder has started digitizing these records, which will make the process much easier. Once you have the page, note the previous owner. The deed transfer entry may span pages, so if you're returned a few pages read them all carefully to make sure you've got the right deed. You can run a search on the previous owner's name and get all the deeds they're involved in for that county. The software they use is not the greatest at sorting, but you'll want to try and go back to earlier books and find where that owner was deeded the land. If multiple owners are listed (such as husband and wife), you may have to search on the other owners' name to find it. Repeat this step as far back as you can go and you'll have the names of everyone who legally owned the property.

Various brick walls you may encounter:

Husband and wife listed, but you can't find where either one was deeded the property. Try searching on the wife's maiden name if it can be determined.
After the home was built, you can't find when an owner purchased it. Using the transfer date you know of as a reference, step back in time using US Census data. The census is recorded every 10 years, so this will get you close to the date they moved in. Using the oldest census you could find them living in the house on, step back further in time using old city directories available at your local library. This method may also give you a more precise date that the home was built than available at your county offices.

Now that you have names, you probably want to know more about them. When Google fails, at your local library you can search obituaries and local history collections for the names. The stories you find may answer questions you have about your home!

If you think your home is haunted, first contact a local ghost hunters group. In addition to being the right thing to do, they may have someone available to do this research for you. In Southern Indiana, try The Anomaly Response Network.

Irish Genealogy: Bypassing the County Genealogy Center

2011-09-18T01:49:00.001-04:00

Note: I first published this summary of my experiences researching my Irish genealogy on the Geni Forums four years ago. Geni has since discontinued the forums, so here is the post in its entirety.

Finding your Irish Ancestors for Free
Sean Feeney - 7 July 2007

Note: This concerns my experience tracing ancestors in County Sligo, Ireland. Your experience may vary, and I'm sure each county is a bit different in terms of where they store their records but this should serve as a good guide.

Yes, the records are mostly free. All you’re paying for at the center is for someone to do your research for you. There is a reference library down the road from the central library which has the property records, like Griffith Valuations, and the census records. The central library doesn’t have anything. You’ll need to know the surname and exact location (not just the county). To get to the reference library, walk east along Stephens St. from the central library and turn right on Bridge St. The reference library will be on your left, about half way down the block. Hours: M-F 9:30-12:45, 14:00-16:45.

If you are unsure of the exact location, it might be listed on your ancestor’s birth record. To get a copy of birth, death, or marriage records, you’ll have to go to the Markievicz House (Health Services Board - HSE) up the road from the central library. Walk west along Stephens St. and take the first right up Holborn St. Veer left at the fork in the road and Markievicz will be on your left in a fenced off parking lot. It will cost you 6 euro for a research copy of a record. The location and the mother’s maiden name is about the only thing you’ll find of use on a birth record from the late 1800’s in Ireland – they did not include the parents’ dates and places of birth like we did in the United States. These records also only go back to 1864.

If you want maps, you’ll need to contact the Valuation office in Dublin. They can issue you a map based on the information in the first column of the Griffith Valuation for your ancestor’s land. It is located in the Irish Life Center on Lower Abbey St.

Source Material at the Sligo Reference Library

If you know your ancestor’s religion, you can consult church records going back to 1762 in some parishes. This may include births, marriages, and deaths and you can find the parents names (including mother’s maiden) on the baptismal records.
1901 Census of Ireland.
1876 Return of Land Owners Co. Sligo.
--This would only be for rich families, as landlords hadn’t given over the land to their tenants yet.
1857 Griffith Valuation.
1841, 1851, 1861 Censuses Co. Sligo.
--These do not have names – only numbers by townland.
1846-1851 The Famine Immigrants (6 Volumes).
1842 Sligo Borough Valuation.
1837 Sligo Voters Lists.
--There was not universal suffrage of any kind, so you’re unlikely to find someone here unless they were very influential and wealthy.
1823-1837 Tithe Applotment Books.
1795-1797 Voters Lists for Co. Sligo.
--See voter lists note above.
1749 Census for the Diocese of Elphin.
1670 Books of Survey and Distribution for Co. Sligo.
1664 Hearth Money Rolls.
1659 Census Co. Sligo.
--This has surnames with number of people using them by barony.
Graveyard inscriptions sorted by graveyard and indexed by surname for Sligo County, mostly taken from the late 1980’s.

If all you’re doing is records research, it may be easier to base yourself out of Dublin. Aside from the Valuation Office, it also has the National Library, General Register Office, National Archives, Registry of Deeds, and various church libraries. Most of the Sligo Reference Library microfilms are just duplicates of the ones at the National Library.

My ancestors lived out in the country, not in Sligo town, and they were farmers. This being the case, the sources most helpful to me were the parish records, 1901 census, Griffith Valuation, tithe applotments, and graveyard inscriptions.

Overcoming the 1864 Brick Wall

Due to English invasions, most records prior to this time were destroyed. All we have before this time are the English records and the 1659 census, which just gives us an idea of where our surname was located at the time. Your best bet for overcoming this brick wall is using the parish records for as far back as they go and studying the graveyard inscriptions, which can easily go back to the late 1700’s, and talking to the people in the area.

Even if your name is a fairly common Irish name, I found that if you go to the area where your last known ancestor lived, you can find descendants whose word-of-mouth information can get you a good bit past 1864. The Irish have a long-standing tradition of oral story telling to relate information about ancestors and this continued into the twentieth century, albeit by this point the stories no longer involved magical folklore. This is good for genealogists as this lends more credence to the information nonetheless. The people in Ireland are incredibly nice and don’t mind you asking about family information, at least in the rural communities where everyone still waves to everyone as they drive by.

In my case, I just drove to the townland where I knew my ancestors lived and I started asking in the shops and pubs that proudly displayed my surname. Some customers pointed me in the right direction, and the third door I knocked on had a long-lost relative in it (from their perspective, I suppose, I’m the long-lost relative since the family that stayed in Ireland continued on after my grandfather came to America). Finding this relative allowed me to really grow out my tree, both upwards as I expected from this trip and to the side and downward, as I did not expect. Although I’m sure this cannot happen for everyone, if it happened for me it could happen for you.

Have you been successful finding information about your ancestors in Ireland using a source or method not listed in this article? Please add a comment about your experience!

Vigo County, Indiana 1930 Census

2011-09-15T18:59:00.000-04:00

This might be helpful for anyone else looking...

You can search the 1930 census for free using your Vigo County Public Library card number at HeritageQuest. Terre Haute starts on Roll 634, Page 33.

Install Starcraft 2 on Ubuntu 11.04

2011-06-12T18:48:00.004-04:00

Put in the SC2 CD.
sudo mount -o remount,unhide /dev/cdrom
wine /media/SC2-L100-D1/Installer.exe

The updates will take forever, but they will complete eventually. Afterwards you have a link on your desktop that will allow you to play. Also see Regedit video performance tweaks on WineHQ.

Organize Launcher Icon Order in Ubuntu 11.04

2011-06-10T21:16:00.002-04:00

Left click and hold an icon and you will be able to move it, similar to what you do on an Android phone.

rhn.rpclib.InvalidRedirectionError

2011-06-08T15:13:00.002-04:00

Remove everything from /var/spool/up2date/ and try again.

Customizing Email Alerts in Nagios XI

2011-06-06T14:58:00.008-04:00

In Nagios 3, you could define a custom "macro" on services or hosts with information to be included in host or service alert messages. You can still do this in XI, but the message template is now a part of each user's profile and you have to modify the related xi notification handler command.

Here's how you can include additional macros from Nagios Core in your notification messages:

1. For service notifications, modify the "xi_service_notification_handler" command and add the following to the end of the command definition:

--newvar="$_{SERVICE|HOST}SOMEMACRO$"

...where $SOMEMACRO$ is a valid Nagios Core macro (either a standard macro or a custom macro). This passes the Nagios Core $SOMEMACRO$ macro to Nagios XI as a "newvar" variable that can be used in service notifications. Custom macros are defined in the Free Variables section of the Misc tab on host and service definitions. When defining a custom macro, you leave out the SERVICE or HOST prefix in the free variable definition. That part is added automatically by Nagios before it is passed to the notification handler.

2. In Nagios XI, go into your account page and select the "Notification Message" link. Customize the "Service Alert Message" text to include the new variable (enclosed in % signs). For example:

Custom Variable: %newvar%

Note that the name of the variable in the notification message is "newvar" and that is will be replaced with the value of "$SOMEMACRO$" that you passed from Nagios Core to Nagios XI.

Similar process for host notifications. You have to make this change to every profile (via masquerade) you want to receive the custom macro in their alert messages.

Note: As of XI release 1.9, the xi_service_notification_handler and xi_host_notification_handler commands are overwritten during an upgrade. Be sure to backup these commands before an upgrade so you can re-implement them afterwards.

Ubuntu Live CD Username and Password

2011-06-05T17:14:00.002-04:00

Ubuntu LiveCD may occasionally lock up and want you to login. The only option is "Other...". Click this and you'll be prompted to enter a username and password. The username is ubuntu and the password is blank.

Great Apps for Rooted Android Evo 4G

2011-06-04T00:36:00.004-04:00

Instructions to root the Evo 4G

I may occasionally update this with more essential apps:

Android Wifi Tether - Sprint will try to charge you to use the 4G hotspot found via settings; this will turn your phone into a hotspot for free.
BusyBox - JRummy16's BusyBox Installer includes essential unix apps like cp, grep. I chose 1.19 and /system/bin. You can uninstall this installer after running it and busybox will stay. This version is better than Stericson's because it symlinks some additional things (like which) that Stericson's doesn't.
ASTRO - GUI file browser
ConnectBot - Extra features when rooted
Titanium Backup - Backup and delete stock apps
Terminal Emulator - Jack Palevich's "Android Terminal Emulator" works best with the Evo's touch screen.

Bash on Android Evo 4G

2011-06-04T00:07:00.001-04:00

Download bash in your Android browser.

su
mount -o rw,remount -t yaffs2 /dev/block/mtdblock3 /system
cat /mnt/sdcard/download/bash.txt > /system/bin/bash
chmod 0755 /system/bin/bash

Now in Terminal Emulator you can set "Command line" (your shell) from /system/bin/sh - to /system/bin/bash -

Remove Peep

2011-06-03T20:37:00.002-04:00

Root your Android, backup peep using Titanium Backup in case you ever want it again, open terminal emulator
su
mount -o rw,remount -t yaffs2 /dev/block/mtdblock3 /system
rm /system/app/*Twit*
And you're done!

You may also be able to remove Peep and the following in Titanium Backup without using command line.

Related stock Evo apps...
*FriendStream (/system/app/FriendStream.apk)
*FriendStream Widget (/system/app/com.htc/FriendStreamWidget.apk)
*Facebook for HTC Sense (/system/app/HtcFacebook.apk)
*Twitter for HTC Sense (if you did not use wildcard command above)

TSM VCB Proxy Setup

2011-05-23T18:39:00.000-04:00

Planning

There are different ways to backup your data - over the network, direct from SAN, or "hot add." Your proxy server can itself be a virtual machine if you choose over the network or hot add modes. Hot add is supposed to be very efficient, as it gives a direct access path to the other VM's disks, similar to SAN-based backup, but I just didn't trust mounting production drives across multiple VMs. So I chose over the network for this walk through. As long as your proxy VM and the VM it's backing up are on the same physical hosts, the physical network won't actually be involved. It's actually recommended practice that you only have one ESX host per virtual proxy.

Setup

You need a service account, preferably in an active directory domain, to run the backup process as. This service account's password can have NO SPECIAL CHARACTERS as of TSM 6.2.2. While vcbMounter can handle a password with special characters, TSM cannot. A bug report has been filed. This service account needs to be setup in your vCenter with the following permissions:

Log on to the VI Client as a user with Administrator privileges.
From the VI Client, click Administration in the navigation bar.
Click the Roles tab.
Click Add Role. The Add Role dialog box appears.
Type a name for the new role, for example, VMware Consolidated Backup User. THE EXISTING SAMPLE ROLE IS NOT GOOD ENOUGH.
Select the following privileges for the new role. Click the plus (+) signs to expand the lists, as needed.

Global > Cancel Task
VirtualMachine > Inventory > Create
VirtualMachine > Inventory > Remove
VirtualMachine > Interaction > Power On
VirtualMachine > Configuration > Disk Lease
VirtualMachine > State > Create Snapshot
VirtualMachine > State > Remove Snapshot
VirtualMachine > Provisioning > Allow Virtual Machine Download
VirtualMachine > Provisioning > Allow Read‐only Disk Access
VirtualMachine > Provisioning > Customize
VirtualMachine > Provisioning > Allow Disk Access
VirtualMachine > Provisioning > Allow Virtual Machine Files Upload
Resource > Assign Virtual Machine to Resource Pool

Click OK to complete the process.

Back in Inventory-Hosts and Clusters, click the top level name of your vCenter server and go to the Permissions tab. Right click in any white space and select Add Permission. Under Assigned Role, select the new role you just created. Under Users and Groups click Add. Select the Domain where your service account resides. In users field, type Domain\ServiceAccountUsername. OK, propagate to child objects is fine, OK.

Note: If this is the same account you logged into vCenter as, you'll lose your administrative privileges. Any other member of the vCenter's Windows Administrators Group will be able to get you back in.

Now that permissions are sorted out, let's build your proxy server. You need a staging partition whose size is equivalent to the sum of the sizes of all of the VMDK's of your single largest VM. I actually built in additional space to provide for growth. During the backup process, the VM being backed up will be copied to this partition before being sent off to TSM. It's a sequential process, which is why this partition only needs to be as large as the largest VM you plan on backing up.

After installing Windows and configuring to your liking (disabling IPv6 is a good idea to help with address translation issues), you'll need to install the VCB Framework from VMware. Now let's configure it by editing C:\Program Files (x86)\VMware\VMware Consolidated Backup Framework\config\config.js:

BACKUPROOT="E:\" --The staging partition
HOST="vCenter.somedomain.com" --FQDN of your vCenter Server
USERNAME="__sspi__" --Use the currently logged in user
PASSWORD="" --Password is passed from TSM
TRANSPORT_MODE=nbd --Network-based backup

When installing TSM Client 6.2.2, choose Custom and select the VMware option. After the install is complete, open the GUI and go to Edit-Client Preferences and click onto the VM Backup tab.

Backup Type > VMware Full VM (or whatever you want)
Domain for VM Backup > Domain Backup Types > Domain Full VM
Domain for VM Backup > VM Options > ALL-VM
Domain for VM Backup > VM host, VM folder or VM name > ALL-VM
Press Insert
VMware Virtual Center or ESX Server > Host: FQDN of your vCenter Server
VMware Virtual Center or ESX Server > User: Your Service Account
Datastore Location: The staging partition
Apply, Ok, Ok

You should narrow the Domain in a large environment as discussed in the Planning section. Now modify dsm.opt to suit your environment. If you choose to do a file level backup of your VCB Proxy, be sure to exclude the staging drive. Now set the password for your service account within TSM:

c:\program files\tivoli\tsm\baclient\dsmc SET PASSWORD -type=VCB "vCenter.somedomain.com" "Domain\ServiceAccountUsername" "password"

Explicitly give the service account full control of C:\Program Files\Tivoli\TSM\baclient and C:\Program Files (x86)\VMware.

Testing

You can test the ALL-VM domain or just one VM at a time (by specifying optional vmname) with the following command:

c:\program files\tivoli\tsm\baclient\dsmc backup vm vmname -vmbackuptype=FULL

Scheduling

Once testing has completed, you can schedule the backup by running your standard scheduler service on the VCB Proxy and creating a special schedule on your TSM server that kicks off the following client command:

'"C:\Progra~1\Tivoli\TSM\baclient\dsmc backup vm -vmbackuptype=FULL"'

The quotes must be exactly as so.

Troubleshooting

If you get IP address lookup failures, check what hostname it's trying to hit. Ensure proper DNS translation is occurring on the VCB Proxy server for that hostname. You can use the Windows hosts file as needed.

To snapshot 2008 & 2008 R2 VMs, you may need to set the following flag while the VM is offline:
Edit Settings-Options-Advanced-General-Configuration Parameters
Add Row:
disk.EnableUUID = FALSE

To troubleshoot the scheduled command, change the schedule to STARTTIME=NOW on the TSM server and then run it in the foreground by executing the following on the VCB Proxy:

dsmc sched -optfile="C:\Program Files\Tivoli\path\to\dsm.opt"

View KML or KMZ on Android

2011-04-06T22:29:00.000-04:00

So you want to create a custom map and view it on your Android device, maybe showing your location on it?

Create a KML or KMZ file of your map. A good service that links into Google Docs, geocodes a spreadsheet of addresses and provides you a KMZ file is MapAList. At step 5, there's a disk icon under the map, click that to download the KMZ. Alternatively you can use Google Earth to create and save your map in either of these formats, or directly create your map via Google's My Maps, bypassing this step all together.
Import the KMZ into Google's My Maps. Click My Maps - Create New Map - Import. Name it and save it.
Open the native Google Maps application on your Android. Click the Layers icon (the middle icon at the top right). Click More Layers at the bottom right. Click My Maps. Select your map.

Voila. Your map, with your position (more or less accurate depending on whether you have GPS enabled at the moment).

Install TSM 6.2.2 on RHEL6 64-bit

2011-03-31T17:11:00.001-04:00

Download the tar from IBM and extract it:

tar -xf 6.2.2.0-TIV-TSMBAC-LinuxX86.tar

Grab dependencies from RHN and then install:

yum -y install compat-libstdc++-33 compat-libstdc++-33.i686 libstdc++-4.4.4-13.el6.i686
rpm -U gskcrypt32-8.0.13.4.linux.x86.rpm gskssl32-8.0.13.4.linux.x86.rpm
rpm -U gskcrypt64-8.0.13.4.linux.x86_64.rpm gskssl64-8.0.13.4.linux.x86_64.rpm
rpm -ivh TIVsm-API.i386.rpm TIVsm-API64.i386.rpm TIVsm-BA.i386.rpm

Add EPEL Repository to RHEL6

2011-03-31T12:29:00.001-04:00

Enable EPEL repo:
wget http://download.fedoraproject.org/pub/epel/6/i386/epel-release-6-5.noarch.rpm
rpm -ivh epel-release-6-5.noarch.rpm
Enable the 'optional' RHN Channel:
1. Go to RHN Website
2. Click into your system's profile
3. Click Alter Channel Subscriptions at the bottom left next to Subscribed Channels
4. Check RHEL Server Optional (v. 6 64-bit x86_64) and press Change Subscriptions
Change all https to http in /etc/yum.repos.d/epel.repo and epel-testing.repo

Change Keyboard Language At Windows 2008 Login Screen

2011-03-15T09:42:00.000-04:00

If you're only given one keyboard layout option at the Windows login screen, and it's not English but you need some special characters only available on the English layout, login via RDP (which will allow you to pass the correct characters from your client) and fix the following registry key:

[HKEY_USERS\.DEFAULT\Keyboard Layout\Preload]

You should see string values like:

"1"="00000407"
"2"="00000409"

407 = german, 409 = us, etc. You can add more languages 3, 4, 5, etc. Now at the login screen press CTRL+ALT+DEL. When you click [Options] you see the default language from key 1 (in this example DE) in the lower left corner. When you Press ALT+SHIFT you can switch all languages you have defined under this registry key. via TechArena

Export MSSQL Database Structure

2011-03-14T13:46:00.000-04:00

Right click the database in Management Studio and select Tasks-Generate Scripts. Run through the wizard, selecting what you want exported. Assuming you only want the structure, without data, in the Scripting Options tab click Advanced and make sure that General-Types of data to script is set to Schema only. You can save to file, clipboard, or a new query window.

PHP Google Maps Geocoder Tutorial

2011-03-06T01:18:00.000-05:00

The goal of this tutorial is to plot a series of points on a map based on their addresses and optimize for viewing on an Android phone via the Android Browser.

Background:
"Geocoding" is the process of turning the addresses into their corresponding latitude and longitude coordinates, which can then be plotted on a map. Something to keep in mind is that your server is limited to 15,000 geocoding requests per day using the free Google geocoding service. If you have a large number of addresses to map, or expect a lot of traffic, you should cache the latitudes and longitudes in your database along with the addresses so that you don't hit this limit. Caching would also improve speed.

Assumptions:
You're comfortable accessing the addresses in your database and manipulating object-oriented PHP code.

Before getting started, you need to sign up for a Google Maps API key.

Helper classes which shouldn't need editing:
point.php
placemark.php

Classes you need to edit:
genxml.php
At a minimum, insert your Google Maps API key on line 7, your database connection information on line 36, and your table name on line 40. Modify the SQL query to match your database and your needs, as well as the associative names in the while loop. You can also include more XML attributes for each address, which could then be printed out on their onClick pop-ups later on.
dmap.php
At a minimum, insert your Google Maps API key on line 8. If you added more XML attributes, add them to the createMarker function and define them in the GDownloadUrl inside function which reads the XML. If your application needs a zoom level less than 10, edit line 33.
geocoder.php
The only function you'll care to edit here is the lookup function (line 40). All of my addresses were for one city, so I appended the city and state to the end of the address before geocoding. You could just as easily read these out of your database. Just pass them as new arguments to the lookup function, update the $search string and, back in genxml.php, the call to the lookup function by $placemarks on line 18.

What's going on here?
Most of the tutorials I found online involved just JavaScript. Well you're not going to be reading values out of databases using JavaScript. The trick is the geocoding, which you need to put into XML format (or some other format that Google supports, like JSON) before you can map it. In this tutorial, this is done by genxml.php. Visit it in your browser to see what the XML output looks like. dmap.php reads the XML, outputs the map using JavaScript, and formats it for the Android Browser using CSS and the viewport meta tag. During the output process, a bounding polygon is drawn around the points which is then used to determine the optimal center point and zoom level.

Download the source files for this tutorial.

Known Bugs:
The pinch to zoom feature that we're all so used to with mobile Google Maps does not work properly in the Android browser. There is no known work around, but the word is that they're working on adding the feature in Android 2.3.

Another thing to keep in mind is that this tutorial uses Google Maps API v2, which is depreciated but still supported. I would have used v3 but there weren't as many resources for it as for v2. You should probably look into using v3, especially if you want pinch to zoom to work whenever that Android update comes out.

Sources:
Geocoding with PHP and the Google Maps API
Using PHP/MySQL with Google Maps
Google Maps API v3: Developing for Mobile Devices
Google Maps JavaScript API V2 Reference

Recovering a Corrupt Windows 7 Profile

2011-03-04T11:07:00.001-05:00

This morning when I logged into Windows 7 all of my icons were blank. So I rebooted. This time when I logged in it was like I was a completely new user, and some things were broken (for example, Folder View Options list was completely blank). Corrupt profile, I figure. So here's how I recovered:

Since I was logged in as some TEMP profile (check: Go to My Documents. It's blank. Now go to C:\Users\username\My Documents. It's not blank. You're not logged in as username even though you should be), and was a member of the local Administrators group, I went ahead and copied C:\Users\username to C:\username.
Run regedit. Go to "HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\WindowsNT\CurrentVersion\ProfileList" and expand it. You'll see several S-1-5-# keys under it. Go through each one, checking the value of "ProfileImagePath" to find username's profile. When you find it, delete the entire S-1-5-# key.
Reboot. Log in as username again. A new default profile will be created. When it's done loading, log out.
Log in as a local Administrator other than username. Copy C:\username back to C:\Users\username. Choose Yes to all merge and overwrite prompts. Choose Skip to any permission problem prompts (as long as you're a local administrator it doesn't matter).
Next time the user logs in everything should be back to normal. Some shortcuts may have to be re-established (for me: Taskbar Toolbars), but the bulk of the profile is as it originally was before corruption.

via interworks