From ASD Wiki

Bringing Up All Systems

Cheat Sheet

1. MSA1000 must be brought up first, if given the opportunity. If not, make sure that ASDFIBRE comes up after MSA1000. For instance, tonight I had to reset ASDFIBRE because MSA1k didn't come up before ASDFIBRE, and therefore FIBRE couldn't see it.
2. If clustered servers came up at the same time or before MSA1000, you will probably have to unload Clustered Services on each machine with the command ULDNCS and bring it back up again (AFTER MSA1000 and ASDFIBRE are up!!) with LDNCS. If they still don't see the shared storage, you will probably need to make sure that ion.ncf has been activated. If not, you will probably want to use ioff.ncf first to make sure the connection hasn't been made. CAUTION, don't use ioff.ncf before running ULDNCS, or you will drop in a poison pill and abend the server.
3. Check VMHOST and make sure that appropriate VMs are running.
4. Inside VMHOST, make sure that Compass server has started properly.
5. Make sure ASDPRINT has started up. Most likely, you will need to select DISTRICT and the print manager to get NDPS Manager to start properly.
6. Make sure MAIL1 and MAILUX have started. MAIL1 will autostart, but as of this writing, you will still need to start up MAILUX's Linux MTA. This can be accomplished by logging in to MAILUX and move to superuser at the console. Change directory to /opt/novell/groupwise/agents/bin and run the command "./gwmta @strtupus.mta &". This *should* start up the MTA. The GWIA will start as a function of the GRPWISE daemon in /etc/init.d. You should also startup the script "gwmon &" This will start up groupwise monitor which you can reach here.
7. Make sure ASDCAFE and ASDTERM have started up
8. By now, you should be able to check Clustered Services. Use ConsoleOne and click on "CL1.District". Once selected, choose VIEW from the pull-down and choose "Cluster State View". This will show you what services are running on which servers. For instance, currently, all volumes and services are running on ASDAPPS1, which is a bad thing. More on what should be running where, here: Nerd:Cluster.
9. Check eDirectory and make sure it doesn't need help resyncing (see below).
10. Make sure the Helpdesk server is up
11. Check SLPDA services. At ASDCOMM, type "DISPLAY SLP INFO". This will show the DA's and if they are running. If they aren't, type "SLPDA" at the consoles of AHSSERV1 and ASDCOMM. Reset SLP by typing "SET SLP RESET=ON" at ASDCOMM. Displaying SLP info again should list the SLPDA's now. This is a must if you want the clients to find the tree when logging in.
12. DHCP services is running on ASDPRINT now. Make sure that you can get a DHCP address (with pxe or whatever). If you can't, start the server (see below).

ZENWorks

Rogue Process Management

Logfile location: AHSSERV1/Sys/Rogue/rogue.txt
I gave RWCMF to root for Rogue folder. No erasy.
Added many processes from this site:

http://wiki.novell.com/index.php/Apps_to_Ban

The registry key I implemented is set to allow all "rogue applications" 
(applications started outside of the ZEN AppLauncher) to start without 
incident. The exceptions are listed separately inside a registry key 
"blacklist". So far, I have the following programs in the "blacklist":

pingfu.exe (the web tunneling program)
kazaa.exe (a Peer-2-Peer file sharing program)
limewire.exe (another Peer-2-Peer file sharing program, mainly for MP3s)
(I added a ton more after posting these three)

I did a quick Google search and added a few more, including the installer 
.exe files, so hopefully it won't even get on the computer.

I also found this link and added some of the newer ones listed:
http://wiki.novell.com/index.php/Apps_to_Ban 

If any of these programs in our block list are started, the process is 
terminated within 3 seconds, the user is notified and an instance is 
logged to \\ahsserv1\sys\rogue\rogue.txt. Supposedly, even if a user renames 
the application, it still uses the "Original Filename" property inside the 
application to identify it. 

I assigned this to all student containers in the MS and HS organizations. 
How do you feel about assigning it to all Staff containers on all campuses? 
I think it might be a good idea to help protect their machines, but I wanted 
your input before I committed that association. A good thing about having it 
logged to a file is that it would help us track the location so we can take 
appropriate action to protect those machines. This is a sample log entry from 
my test earlier:

"Process Terminated","81","2/11/2008 8:57:41 PM","HSTAFF.STAFF.HS.ALMA_SD",
"VM-XPP100:0C:29:FD:05:66.WORKSTATIONS.ZENWORKS.DISTRICT.ALMA_SD",
"10.0.151.33","ZENworks Process Management","","","","","C:\Program Files\LimeWire\LimeWire.exe",
"LimeWire.exe","5612","2952","Process terminated successfully","0"

It logs everything... time/user/workstation/ip/process

NetWare

Mapping Drives

To manually map a drive, you need to know two things:

1. An unused drive letter on your workstation
2. Server and folder location of the drive you want to map

The syntax is as follows:

map root [letter]:=\\[server]\[folder path]

So to map the print volume, you'd use:

map root P:=\\asdprint\printvolume\example_folder

NDS

Resyncing eDirectory

Occasionally, for whatever reason, wear and tear on DS, the replicas will get out of sync. You can check synchronization on any server with a replica by entering the following command at the Netware Console Screen:

dsrepair -a
Report Synchronization Status

This will cause all of the servers in the replica (currently, ASDCOMM AHSSERV1 AMSSERVER AISSERV1 APSSERV1 MAIL1 and ASDAPPS1) to report the last time they were synchronized. If they are out of sync, you will need to do the following tasks:

Start DSREPAIR -A on all of the servers in the replica
ADVANCED OPTIONS MENU > REPAIR LOCAL DS DATABASE > LOCK NDS... and F10 to begin scan
Do this on all servers, repeating the scan until you have 0 errors

Once you have repaired all replicas EXIT out of DSREPAIR on all servers and go to the Master Replica (ASDCOMM). From the Netware Console, type the following commands:

set dstrace=on
set dstrace=+s
set dstrace=*h

This is only done from ASDCOMM and none of the others. This basically tells ASDCOMM to broadcast to the other servers "I'm #1 server, synchronize your directory to mine."

I've seen this process take as little a couple of hours and as long as a day. When it's completed, you can run the Synchronization Status Report again from the DSREPAIR front page and everything should be synchronized.

Clearing Obituaries

This process is pretty much the same as above except you use the following DSTRACE commands instead:

set dstrace=on
set dstrace=+purge
set dstrace=*j
set dstrace=*f

This must be done on EACH server in the replica.

Console Commands

Starting DNS/DHCP

DNS

ASDCOMM: LOAD NAMED -DL 2
ASDPRINT: LOAD NAMED -DL 2

DHCP

ASDPRINT: LOAD DHCPSRVR -D2

Resetting SLP

SET SLP RESET=ON

Linux

Additional Repositories

openSUSE has a list of additional repositories on their website. There are a lot of application-specific ones like OpenOffice and Firefox. Also on the list are two of my favorites, GURU and Packman.

1-Click Install for SLED

I had to pull a few prerequisites before it would let me install the YaST module.

1. Download xdg-utils and install.
   
2. Make sure that Perl-XML-XPath is installed. (from GURU or PACKMAN repositories--see above)
   

These were the only two dependencies that were required when I installed the manager, but it's possible that you might have more. Once you have met all the requested dependencies, install the yast2-metapacakage-handler.

Console Commands

NCP

Info taken from here. Italicized copy below is direct pulls from this link.

ncplogin

Usage of ncplogin is required for every new tree the current user wish to connect to. If current user has been authenticated via NetWare using the pam_ncp_auth module, usage of ncplogin is not needed for accessing resources belonging to the tree to which the authenticating server belongs.

Establishes a permanent connection to a nds tree. Example:

ncplogin -l -S ASDCOMM -U admin.administration -P ***** -A 10.0.0.2

-l automounts (SYS, I think) in /mnt/ncp/*
-S server name
-U username & context
-P password
-A ip address

ncpmap

ncpmap supports all arguments of ncplogin/ncpmap except any argument related to user name or password, since it requires that user is already authenticated.

ncpmount

Ncpmount uses the same syntax as the above commands and is a command that authenticates and maps in one step. Downside is that it is a temporary action, as I've seen my mappings go away after a period of time. Using ncplogin/ncpmap is supposed to be a more permanent solution.

Interpreting vmstat

taken from http://rimuhosting.com/howto/memory.jsp

vmstat helps you to see, among other things, if your server is swapping. Take a look at the following run of vmstat doing a one second refresh for two iterations.

# vmstat 1 2
  procs                      memory    swap          io     system         cpu
r  b  w   swpd   free   buff  cache  si  so    bi    bo   in    cs  us  sy  id
0  0  0  39132   2416    804  15668   4   3     9     6  104    13   0   0 100
0  0  0  39132   2416    804  15668   0   0     0     0   53     8   0   0 100
0  0  0  39132   2416    804  15668   0   0     0     0   54     6   0   0 100

The first row shows your server averages. The si (swap in) and so (swap out) columns show if you have been swapping (i.e. needing to dip into 'virtual' memory) in order to run your server's applications. The si/so numbers should be 0 (or close to it). Numbers in the hundreds or thousands indicate your server is swapping heavily. This consumes a lot of CPU and other server resources and you would get a very (!) significant benefit from adding more memory to your server.

Some other columns of interest: The r (runnable) b (blocked) and w (waiting) columns help see your server load. Waiting processes are swapped out. Blocked processes are typically waiting on I/O. The runnable column is the number of processes trying to something. These numbers combine to form the 'load' value on your server. Typically you want the load value to be one or less per CPU in your server.

The bi (bytes in) and bo (bytes out) column show disk I/O (including swapping memory to/from disk) on your server.

The us (user), sy (system) and id (idle) show the amount of CPU your server is using. The higher the idle value, the better.

Creating ISO from CD/DVD

To create an ISO from a CD, use the following command:

dd if=/dev/cdrom of=[filename].iso

To create an ISO from a DVD, use this:

dd if=/dev/dvd of=[filename].iso

Erasing MBR

The following dd command will erase everything from the MBR up to the partition table information:

dd if=/dev/zero of=/dev/hda bs=446 count=1

This didn't work as we had hoped. Will continue to research a solution later.

Large File Management

Splitting Large Files

I ran into an issue recently where I had a ZENworks image that I needed to burn onto a DVD. The problem was that I couldn't have a file larger than 4GB on the DVD and the image was 8GB. Skip has noted that ZENWorks Image Explorer has this ability for this very reason, but Linux currently doesn't have Image Explorer. Regardless, this command will work with any file.

split -b [number][k|m] [file to split] [output name]

Use k for [*1024 bytes] and m for [*1048576 bytes]. The number is the size you want it to break into. It will create files based on that option into filenameaa, filenameab, filenameac, etc. For example:

split -b 1m filename.zmg splitfilename

will output the following filenames at 2 meg increments until it finishes:

splitfilenameaa
splitfilenameab
splitfilenameac
etc.

Concatenating Large Files

This section will describe how to concatenate files back together.

cat file1 file2 file3 > newfile

To recombine the zmg file mentioned above, use the following command:

cat splitfilenameaa splitfilenameab splitfilenameac > filename.zmg

I've successfully split and recombined a PDF file and it opened successfully.

Apache Configuration

Helpdesk default-server.conf

Scenerio...
/srv/www/htdocs/ - is webroot
/srv/www/cgi-bin/ - is script folder behind webroot in tree structure.

I needed to have helpdesk.almasd.net/ work so I didn't have to include /pdesk.cgi or /cgi-bin/pdesk.cgi. My ultimate error was in creating symbolic links for the cgi files in htdocs/ used in conjunction with the apache directory options "FollowSymLinks" directive. I deleted the sym links I created out of the webroot and the site started working. I guess FollowSymLinks makes it's own path to those cgi-scripts. Below is the webroot directory configuration:

<Directory "/srv/www/htdocs">
 Options FollowSymLinks ExecCGI
 AllowOverride None
 Order allow,deny
 Allow from all
</Directory> 

Using ScriptAlias wouldn't work right with DirectoryIndex, so I had to use Aliases instead:

Alias /images/ "/srv/www/htdocs/images/"
Alias / "/srv/www/cgi-bin/"

And finally, I had to set up the cgi-bin directory to execute the perl scripts:

<Directory "/srv/www/cgi-bin">
 DirectoryIndex pdesk.cgi
 AddHandler cgi-script cgi
 Options FollowSymLinks ExecCGI
 AllowOverride None
 Order allow,deny
 Allow from all
</Directory>

Now I don't know if I need the DirectoryIndex listed inside there or not, but it is working, so I'm leaving it alone. :)

Windows Apps Replacements

I love finding tools to replace common stuff I do in Windows. I hereby dedicate this section to helping eliminate dependency of Windows.

Minicom - A HyperTerminal Replacement

Install minicom with yast2

# minicom -s

You will get Configuration screen - Choose Serial Port Setup 

1. Choose A - Serial Device : /dev/ttyS0 (If your serial cable is connected in COM-1)

2. Choose E - Bps/par Bits : Again Choose E (ie.9600)

3. Choose F - Hardware Flow Control : No

4. Esc from setup screen

5. In the configuration screen Choose - Save setup as df1

Linux Hyperterminal is ready.

You will need to start it with the following

minicom df1

or create a sh script to do that for you.

minicom -s df1

will allow you to edit that configuration file again. Use this one to exit out of Minicom:

CTRL-A Q

Rdesktop - A Windows Terminal Client Replacement

Install rdesktop & tsclient with yast2. It's pretty much ready to go when you install it, plus tsclient is a GUI front-end and will show up in your Applications list.

SLED10

Things noted to get dual-boot and linux graphics working

Resizing Partition

1. Must run "chkdsk /r" before using Partition Magic 8 to successfully resize disk

3D Desktop Effects

1. Download and execute this as root to compile the ATI driver against your specific kernel. Make sure you have installed the kernel source. (Caveat: Should you upgrade your kernel through the update service, you will need to recompile a new driver. Found that out the hard way. Unfortunately, I couldn't get the SLED ATI install source's driver to work, which would have negated a recompile.)
2. Run aticonfig with the following command as root:

init 3
aticonfig --initial --input=/etc/X11/xorg.conf

1. Run sax2 with the following command:

sax2 -r -m 0=fglrx
init 5 (or just restart)

1. Enable desktop effects

--- One last thing, should you get in a situation where you can't log in graphically, you have to do the following:

1. At the boot screen, enter "3" and the default choice to start up in init level 3.
2. Log in as root and enter the following:

sax2

1. Click OK and reboot your machine

This should get you running again where you can compile a new driver.

Installing Novell Linux Client 1.2

1. Download the installer here and unpack it somewhere.
2. Read the instructions here. (I used the Add-on Package utility in yast)
3. Reboot then set up system preferences from the system tray icon right-click
4. After setting up SLP and defaults, reboot again. (DA's are 10.0.0.2,10.0.2.1)(Scope is ALMA_SCOPE)

Installing Novell Linux Client 2.0 Beta

1. Download the ISO here.
2. Use yast2 to choose an Add-on Package and use the ISO as the source.
3. Installation instructions are here.

NLC 2.0 Beta also allows integrated login so if you set up CASA (via the client advanced tab), you will get your login script after initial login. Finally!

Desktop Wallpaper

Geeko Samurai - 1680 x 1050

NDISWRAPPER

You can use ndiswrapper to install a wireless card's Windows driver in Linux. Obtain the Windows driver and extract it into a folder. When you have, locate the *.inf file. This is the file you will use with ndiswrapper. Open a terminal shell and log in as root. Type in the following:

ndiswrapper -i [location & name of file].inf
modprobe ndiswrapper

This will install and start the wireless driver.

NDISWRAPPER @ startup

Fired up with my latest Linux training (thanks to SwitchDesk's Chris Clark), I decided to take on a recent issue regarding the startup of NDISWRAPPER at boot time. This is assuming that you have used ndiswrapper to install the correct INF file for your wireless adapter. Afterward, I did the following things:

1. Create a file in /etc/init.d/ called "wlan"
2. Use a text editor (joe, vi, etc.) and add and save the following script to the empty file (The CR after the command is important...):

#!/bin/bash
modprobe ndiswrapper
 

1. Change the permissions to make "wlan" executable:

chmod 755 /etc/init.d/wlan

1. Make the wlan script start up in runlevel 3 and 5:

chkconfig -a wlan

1. reboot your computer.

This will cause modprobe to add the ndiswrapper for wireless (if you've already set up ndiswrapper with the wireless card's .inf file) when the computer boots up. I've had to turn off the wireless and back on to get DHCP to kick in, but it works!

Compiz-Fusion SLED101

Added the following sources:

• http://ftp.stw-bonn.de/pub/opensuse/repositories/X11:/XGL/SLE_10/
• http://download.opensuse.org/repositories/X11:/XGL/SLE_10/
• http://download.opensuse.org/repositories/X11:/XGL/SUSE_Linux_10.1/

Ran:

zypper in compiz-fusion-gnome

That failed. Currently, XOrg 7 is required for SLED to run fusion, and Xorg 6.9 is installed.

Firefox

Annoying Trackpad Fix

In Firefox, type “about:config (without the quotes) in the location bar
Change mousewheel.horizscroll.withnokey.action to 0
Change mousewheel.horizscroll.withnokey.numlines to 1

Teaming & Conferencing

NERD:tandc Teaming & Conferencing

PRINTING

Restarting Print Manager

This should be set to startup automatically in the autoexec.ncf file, but should you have to do it manually, enter this command at the Console:

ASDPRINT: load ndpsm

Choose the District container and then the ASD_DISTRICT_MANAGER. This will start up the manager and bring all of the printers online.

Reading Counts/SRI 3.0.3 labels

3.0.3 requires a patch located here to print labels correctly.

DNS

A Records

Thought I'd start a section on commonly asked-for DNS entries.

GroupWise Instant Messenger - messenger.almasd.net port 8300
Alma Liferay - portal.almasd.net
Groupwise client - mail1.almasd.net

GroupWise

Linux

Status & Startup

1. Use putty and log in to Mailux (10.0.7.15) as root.
2. To check the status of Groupwise, type in the following command as root:

rcgrpwise status

The commands to be used with rcgwavaman are:

rcgrpwise status

rcgrpwise start

rcgrpwise stop

NetWare

GWAVA

Status & Startup

1. Use putty and log in to Mailux (10.0.7.15) as root.
2. To check the status of GWAVA, type in the following command as root:

rcgwavaman status

The commands to be used with rcgwavaman are:

rcgwavaman status

rcgwavaman start

rcgwavaman stop

rcgwavaman restart

Learning Feeders

I set up IMAP folders for the Ham and Spam feeders for GWAVA to use. GWAVA analyzes the content and adjusts its rules accordingly to teach itself what constitutes good and bad email, essentially adjusting to your email habits.

Ham

1. Check "Mailbox" of the Ham user and make sure that SPAM wasn't released into it.
2. Move messages to "Ham" folder. This folder is being IMAP'd by GWAVA.

Spam

1. Check "Graveyard" folder for the Graveyard user and make sure the all listed are truely spam messages. "Graveyard" folder is a shared folder.
2. Move messages to the "learnspam" folder. This folder is being IMAP'd by GWAVA.

Exceptions

Sources (Whitelist)

Sources can be added to the exception lists. Adding addresses alone will not create a source. Once you add the address, you have to scroll down to the bottom and open it (single-click) and choose an item to except (i.e. Antispam Heuristics). Once excepted, click Save at the top of the webpage. The list below is current as of 5-14-07:

List located O:\Technology Stuff\Whitelist\whitelist for gwava.txt

Destinations

VMWare

SLED 10 SP1

Download & Install

Get it here. Install it and run vmware-config.pl as root.

Lock-up Fix

I guess this really should go in the Linux section, but I wanted to note a fix for getting VMWare Workstation 5.5.4 to work properly in SLED 10 SP1. I found this solution in the VMWare community forums.

Dalton Kole wrote: A simpler solution to the browse issue:

1. Open a terminal and su to root.

su

2. Change to the libpng12.so.0 directory.

cd /usr/lib/vmware/lib/libpng12.so.0

3. Rename the libpng12.so.0 file within that directory.

mv libpng12.so.0 libpng12.so.0.old

4. Create a new soft link for this file.

ln -sf /usr/lib/libpng12.so.0 libpng12.so.0

5. Now change the the libgcc_s.so.1 directory.

cd /usr/lib/vmware/lib/libgcc_s.so.1

6. Rename the libgcc_s.so.1 file within that directory.

mv libgcc_s.so.1 libgcc_s.so.1.old

7. Create a new soft link for this file.

ln -sf /lib/libgcc_s.so.1 libgcc_s.so.1

8. Restart VMware and you should now be able to use the browse option.

Linux Host Shared Folders

Noting this because it kicked my butt a few months ago...

1. VM > Settings > Options tab > Shared Folders
2. Add... shared folder on the host
3. Browse to here on the guest:

/mnt/hgfs/

Kaspersky Antivirus

Configuration & Installation

Net Agent

Silent Network Install

I copied the Network Agent installer from the AdminKit into the Z:\KAVLicense folder. I set up a ZEN object called "Install KAV Agent - CMD" that silently launches the following command:

Z:\KAVLicense\agent\lsexec.exe /S

I have this running on every staff machine that doesn't have the file klnagent.exe. After this has been installed, I'm able to see the workstation inside my groups.

Reissuing Cert

On machines that we've installed previously, the network agent fails to connect to the new AdminKit server. For these machines, I've set up a routine with ZENWorks to issue a new certificate. I created an object called "Install KAV New Cert". This object does the following:

1. Checks for the following file that I manually insert on their computer:

C:\Program Files\Kaspersky Lab\NetworkAgent\newcert.txt

2. If it finds the file, it immediately deletes it and runs this command to issue a new certificate:

C:\Program Files\Kaspersky Lab\NetworkAgent\klmover.exe -address 10.0.7.32

I've assigned this object to all Staff.*, SpecEd.* and ADM.* containers. Inserting a blank "newcert.txt" file in the above directory on any of those machines will force run next time they refresh.

ZENWorks Workstation 6 MSI Defaults

Scans set to run

• Weekly Scan => Saturday night @ 10:00pm
• Critical Areas (Mem, Win32 folder, etc.) => Daily @ 10:00pm
• Startup Objects (when programs run) => Every time a program runs

Updates

• Scheduled to check for updates every 2 hours
• Checks Administration Kit Server first (local network)
• If fails, downloads from Kaspersky Update servers (internet)

Spyware Module

• Turned off Popup Blocker
• Turned off Anti-Banner
• Turned off Anti-Dialer
• Left Anti-Phishing running
• Left Anti-Spy Running

Web Anti-Virus

• Security Level => Recommended
• Scans basic ports (HTTP, FTP, 8000, 8080, etc.) but not all ports

File Anti-Virus

• Security Level => Recommended

Password

Administration Password is set to prohibit disabling/pausing scanning.

NetWare Antivirus

Updates

• ASDCOMM scheduled to check for updates hourly via AdminKit first, then web
• After updates are downloaded, it rolls out updated virus definitions to the following servers:
  • ASDPRINT
  • ASDAPPS1
  • ASDAPPS2
  • AHSSERV1
  • AMSSERVER
  • AISSERV1
  • APSSERV1
  • ASDTEMP
  • MAIL2

Linux

• ASDSYNC scheduled to check for updates hourly on the web via a crontab:

* 0-23/1 * * *  root  /opt/kaspersky/kav4fs/bin/kav4fs-keepup2date

PAC-related

Choice Setup

Icon Setup

Navigate to T:\VFP and right-click vfp8.exe and send to desktop (create shortcut). Do it twice. Rename one "VFP" and the other "Ticket #" (depending on machine).

Open properties on the shortcut "Ticket #" and modify the following items:

Target: T:\VFP\vfp8.exe ctsw0# -t

and

Start in: T:\VFP\choice

You should now be able to open the "Ticket #" shortcut and log in to the choice server.

Ticket Printer Assignment

Please see section below first. Once the printer has been installed, log in to Choice Ticketing and choose "Maintenance > Maintain the workstation file > Change Records". Once this is selected, choose the correct box office machine and continue. Change the ticket printer and the receipt printer to match the correct printer.

Ticket Printer

1. On Ticket 1, create a new printer for LPT1 and use the "Generic" printer type and "Generic / Text Only".
2. Share the printer so the other computers can print tickets to it. If you are on Ticket 2 or 3, install the shared printer so you will be able to print to it.

XCharge

You will need to contact Choice so they can install it for you. It is important that you grab the configuration file along with the transaction/history folders before you reimage the machine. The configuration file is a really long filename in the xcharge directory on Ticket 1.

BES Server

Install

• !!!Install GroupWise Client FIRST!!!
• Also install Groupwise Messenger
• When running install, choose first option: Blackberry Enterprise Server and Groupwise choice to side.
• Default choices for everything until choosing Groupwise 6.5 or 7 connector.
  • If 7, make sure SOAP is installed.

Wireless Application Installation

Adjusting Application Polling Cycle

You have to create the following Registry entry:

HKEY_LOCAL_MACHINE > Software > Research In Motion > Blackberry Enterprise Server > Blackberry IT Admin Server

Inside this item, create a DWORD entry called AppProvisioningPollInterval with a decimal value in minutes. After you restart the Blackberry Policy Service, it will start this new cycle.

Error Pushing Wireless App to Device

POLC log shows:

[40000] (01/01 00:00:00):{0x970} {user@domain.com, PIN: XXXXXXXX, ID:4}RequestHandler::PollForMissingApps - 
Transaction Id already queued for package <Application_Name>.

Solution: You have to back up the blackberry configuration database:

C:\<Anywhere>osql -E
1> backup database BESMgmt to disk = "C:\BlackberryConfigurationDB.bak"
2> go
Processed 1344 pages for database 'BESMgmt', file 'BESMgmt_data' on file 1.
Processed 1 pages for database 'BESMgmt', file 'BESMgmt_log' on file 1.
BACKUP DATABASE successfully processed 1345 pages in 0.409 seconds (26.929
MB/sec).
1> quit

And you have to run the following MSDE commands:

C:\<Anywhere>osql -E
1> use BESMgmt
2> UPDATE syncserverstate SET State =1 WHERE Type ='PolicyServerLastGroupingNumber'
3> DELETE FROM ITAdminqueue WHERE GroupingNumber <> 0
4> go
(1 row affected)
(19 rows affected)
1> exit

On the next application polling cycle (ours is currently set to 30 minutes), the application will be pushed out to the device.

TroubleShooting

IE URL Crash

Problem

IE loads home page. Clicking bookmark or manually entering URL and pressing enter crashes browser. Happens in Firefox as well.

Suspicion

Believe it has to do with KAV. Only affects computers that haven't had a fresh image in a couple years. Also of note is that these computers have had eTrust antivirus installed in the past. Uninstalling KAV resolves the issue.

Course Of Action

No e-Trust

• Started with clean machine and installed KAV, updated, rebooted and tried to crash IE with description above. No such luck.

e-Trust

• Started with clean machine and installed e-Trust, updated, rebooted and gave it a full system scan. Took Snapshot of VM to test this install of KAV: Uninstall eTrust first, then let KAV agent roll out KAV6.

Results

After uninstalling eTrust and letting the agent and adminkit rule push KAV6, I was unable to reproduce the error. I will wait until it happens again to try to reproduce it so I can get more input first.