Andrzej's "IT Thoughts" Weblog

So a new season for mountain biking started. Actually for me it never ended, since we had a really mild 2007/2008 winter in Poland. But during the time from october 2007 till april 2008 I didn't do any racing, except one 2-day "orientation race" with my dad - it was lots of fun, but did not push myself to the limit like on standard MTB races, and 1 winter race.

So this year I already managed to start in 5 races. What can I say? I'm totally "hooked" into MTB marathons. And it's just beginning. In a week I am going to Istebna for a 4-day, probably polish toughest MTB marathon: http://www.mtbtrophy.com/ in a really magical place on the border of 3 countries.

One of the more fun races was in Karpacz, 1st of May, but we even had some snow on the way! What was also cool, was that professional cameramen where taking pictures during the race, and the end effect is really nice. You can check my photos at: http://alipka.spaces.live.com/photos/cns!FDED5C90960743BD!370.

I hope to write short story soon, but it will probably be in Polish and posted at our team site: http://www.xbox360biketeam.pl.

There is a tab when you select Tools->Connect from Ops Console that is called "Registered Servers". It actually will go into your AD and query for CN=SDKServiceSCP, which is a container that stores information on your Management Groups and RMS servers:

It normally works well, but there are a few gotcha's:

1. if you want to delete old management groups. You can find the RMS computer object in AD and from there find the corresponding SDKServiceSCP:

If you delete the SDKServiceSCP the server and corresponding management group will be gone from the registered servers view

2. If you have a clustered RMS you will notice that you have the SDKServiceSCP ONLY for the 1st node of the cluster. To have the correct cluster RMS name in the registered server, go into the SDKServiceSCP and modify these two attributes to the proper cluster instead of node name:

serviceBindingInformation and serviceDNSName

I was told unofficially that this should be fixed in next SP release for OpsMgr.

I thought I would post the above, as it seems this is nowhere documented.

Walter posted an interesting entry on his blog about OpgMgr 2007 qualifications and requirements: http://wchomak.spaces.live.com/Blog/cns!F56EFE25599555EC!969.entry.

I must say I fully agree with it and can add some additional input.

1. Responsibility and accountability

OpsMgr is a tool that is very powerful. With OpsMgr administrator privilege and agents installed on all Windows servers in the enterprise the OpsMgr administrator could easily halt the whole enterprise to a stop by malicious or inexperienced action, especially if the agent action account is local system or local admin (which is often the case due to overhead in maintenance when using domain accounts).

Therefore OpsMgr administrator must be a trusted, accountable and responsible individual

2. Communication and collaboration skills

Since OpsMgr monitors a lot of technologies the OpsMgr administrator should be able to communicate fluently with other Administrators that administer monitored services (AD, web applications).

Moreover he/she should also communicate well with external partners (consider external MPs) as well as in-house development teams (consider developing custom in-house MPs)

And finally communication with executive IT personnel should be good, since they often require specific reporting on IT services availability, performance and operations metrics.

3. Technical capability to use development techniques

OpsMgr is not only a product, it is a platform. OpsMgr administrator should understand the whole platform and be able to extend it. Building custom monitors, MPs, reports, tools, scripts requires:

• knowlege of scripting technologies (Powershell, VBscript, WMI)
• knowledge of XML (all MPs are XML based)
• knowledge of SQL Reporting Services (for developing custom reports)
• understanding of object oriented design and basics of .NET (OpsMgr is itself object oriented plus the OpsMgr SDK is a .NET library)

4. Knowledge and understanding of processes around Operations - ITIL and/or MOF

Implementing and running OpsMgr also means understanding ITIL/MOF processes related to IT management, mainly (but not exclusively) in ITILv3 terminology in these areas:

• Service Operation: Problem Management
• Service Strategy: Event Management
• Service Design: Availability Management, Capacity Management, IT Service Continuity

So just as Walter mentioned this role requires not only "depth" but also "breadth" of knowledge + quite a lot of "soft skills" (communication, collaboration, responsibility).

My friend Tomek (http://blogs.dirteam.com/blogs/tomek/default.aspx) pointed me to the new tool for automatic diagram generation of SCCM and SCOM. It has just shown up on Microsoft Downloads: http://www.microsoft.com/downloads/details.aspx?familyid=521b3884-1eda-4b9d-8ad7-67d00fe9ce8a&displaylang=en&tm.

Haven't yet time to test it. If it is half as good as the AD Topology Diagrammer (http://www.microsoft.com/downloads/details.aspx?familyid=cb42fc06-50c7-47ed-a65c-862661742764&displaylang=en) it will be really helpful.

Currently the tool supports alerts and computers for OpsMgr and system information, user groups and patch information for ConfigMgr. It looks to be a good framework for extending it further to other information or systems.

There is a couple of things about OpsMgr 2007 notifications, that may not be so obvious at first sight. Here is a list from my experience.

1. Notifications audit trail

So you want to have a trace left about what notification was sent to whom? This is not performed by SCOM out of the box.

My suggestion is to use a command-line notification channel as described by Stefan http://weblog.stranger.nl/using_opsmgr_notification_channels or Steve http://blogs.msdn.com/steverac/archive/2008/03/14/command-line-notifications-and-variables.aspx.

Additionally I like to know to whom it was sent, so I snatch the information on that from Clive's blog:http://blogs.technet.com/cliveeastwood/archive/2007/10/12/a.aspx

and insert it into the event description: $Data/Recipients/To/Address/Address$

Another but similar approach is to log to a text file: http://contoso.se/blog/?p=265

2. Subject encoding

If you have a malformed subject line, this is most often due to the fact that the <SubjectEncoding> flag in XML and that the subject is over the limit for SMTP. With SP1 you can check the checkbox under subject in notification format. If you don't have SP1 you need to export the MP Microsoft.SystemCenter.Notifications.Internal and manually edit the file by removing <subjectEncoding> flag under the subject body.

3. HTML Mail body

You want to add HTML formating to the mail body? Again play with the XML this way (remember to set the IsBodyHtml to true), e.g.:

      <WriteActionModuleType ID="DefaultSmtpAction" Accessibility="Public" Batching="false">
        <Configuration />
        <ModuleImplementation Isolation="Any">
          <Composite>
            <MemberModules>
              <ConditionDetection ID="ContentGenerator" TypeID="Notification!Microsoft.SystemCenter.Notification.SmtpNotificationContentGenerator">
                <Content>
                  <From>MailFrom@domain.com</From>
                  <Subject>SERWER:$Data/Context/DataItem/ManagedEntityPath$ PROBLEM:$Data/Context/DataItem/AlertName$</Subject>
                  <Body>
<![CDATA[
YOUR HTML Mail body goes here

Tip for the URL - tricky with quotes:

<tr><td> URL for SCOM Console:</td><td> <a href='$Target/Property[Type="Notification!Microsoft.SystemCenter.AlertNotificationSubscriptionServer"]/WebConsoleUrl$?DisplayMode=Pivot&amp;AlertID=$UrlEncodeData/Context/DataItem/AlertId$'> Details of Alert in OpsMgr console</a></td></tr>

]]>
                  </Body>
                  <BodyEncoding>utf-8</BodyEncoding>
                  <IsBodyHtml>true</IsBodyHtml>
                  <ReplyTo>MailFrom@domain.com</ReplyTo>
                </Content>
              </ConditionDetection>

4. Mail priority

If you want to play with mail priority use Stefans' hints: http://weblog.stranger.nl/opsmgr_email_notifications_with_high_importance

The issue with modifying the Notification MP from XML level is to remember that you cannot modify it in the GUI (the relevant part only) as the wizards will overwrite the parts that are not exposed in the OpsMgr UI.

5. Targeting notifications to single alert

If you have a request to add a single notification you can use the attached script that bases on solution from Stefan Koell: http://code4ward.net/cs2/blogs/code4ward/archive/2007/09/19/set-notificationforalert.aspx. II wanted to be able to set subscriptions for different recipients, not always the same. The problem is that using a console task you cannot ask user for input, so I just slightly modified it into a standalone script:

Just use SetNotificationForAlert.bat.

Happy Easter to all:)

[update!]: issue is now corrected in 6.0.6277.1 MP, thanks to Marius for quick reaction on this: http://blogs.msdn.com/mariussutara/archive/2008/03/18/update-mp-for-monitoring-cluster-services-mscs-finally-web-released.aspx.

Unfortunately that MP has a breaking change which means you need to delete it previous version before importing new one. To delete an MP you cannot have any MPs referring to it. Read more on Marius blog: http://blogs.msdn.com/mariussutara/archive/2008/04/03/remove-old-management-packs-after-breaking-change-was-introduced-in-newer-version.aspx. Actually I used Boris Override Explorer to locate overrides and create new ones in a separate MP. This is one of the reasons why we suggest to use SEPARATE unsealed MPs for storing overrides for each sealed MP.

[end of update]

I would like to inform anyone using the cluster MP for SCOM just released (6.0.6277.0).

There is a noisy (due to suppression configuration) alert rule:

The suppression is not in place:

<Configuration>

<Priority>1</Priority>

<Severity>2</Severity>

<AlertMessageId>$MPElement[Name='Microsoft.Windows.2003.Cluster.Node.DiskResource.Alert.AlertMessage']$</AlertMessageId>

</Configuration>

Hence for every file that is reported there is a new event. In case of my disk full issue on 1 server the environment created over 30'000 alerts.

I workaround by disabling this rule (I am monitoring disk free space by monitors anyway).

If you need to close the alerts by alert name here is the solution:

Get-Alert | where-object {$_.name -eq "Disk resource space alert."} | resolve-alert

Together with my colleague Marek Kuzminski, we have slightly modified Brian's scheduled maintenance mode MP (Scheduling Groups of Objects for Maintenance Mode) to add following features:

• Override by group display name
• Add additional rules to allow 10 different maintenance schedules
• Add a run as profile to the MP to facilitate permissions to schedule maintenance mode
• Added simple error handling in maintenance mode script for troubleshooting

I wasn't able to re-use that MP to do different schedules just by overrides. So the design is still rather 'quick', and limits your number of schedules to 10 (number of rules). To use overrides for different schedules and computer groups bound to the same rule, the MP would have to be redesigned.

Thx to Brian for posting his great MP, and Marek for his help.

Ever tried mass-installing a lot of agents from console? If yes, then you probably miss some feedback information. Once you close the running task window you loose the information. The information is kept, sadly enough in SUCCEEDED Task status - browsing through that to find failed installations in UI is hard (the failed info is kept in Task output not task status, so no easy filtering here). To get it by Powershell you can use something like:

To HTML:

get-taskresult | where {$_.Output -notlike "*<ErrorCode>0</ErrorCode>*" -and $_.Output -like "*MOM.MOMAgentManagementData*"} | ConvertTo-Html | set-content "D:\Scripts\GetTaskResult\FailedTasks.htm"

To CSV:

get-taskresult | where {$_.Output -notlike "*<ErrorCode>0</ErrorCode>*" -and $_.Output -like "*MOM.MOMAgentManagementData*"} | export-csv "D:\Scripts\GetTaskResult\FailedTasks.csv"

You can easily extend those to do some more elaborate filtering as the Output property is XML, and Psh can parse XML well.

Thx to Bogdan, my colleague from MS Poland, for giving me some background on where we keep the data. The rest was easy.

[UPDATE]:

I have update the MP in recovery part. It was not possible to override the recovery properly, because using a standard script response does not allow overriding the ARGUMENTS parameter of the script (actually it does allow, but it does not work as expected - EventLogName is not resolved, since the target is passed as a string not an object).

So basing on Brian's (http://blogs.technet.com/brianwren) input I created a custom write action module that allows you to override the COMPRESS flag and the Backup log destination folder as separate parameters. Attached is the corrected MP now (version 1.0.0.26). It also has some built in groups to let you divide your servers based on the disk on which you want to store local event log backups (see overrides on the recovery for more details).

I have done some small work to update my backup event logs MP for MOM 2005 to 2007 (http://blogs.technet.com/alipka/archive/2006/04/09/monitoring-and-backing-up-event-logs-with-mom.aspx). I attach it to this blog post. The MP has been roughly tested on SP1 RC, but should work with SP1 RTM, for SP0 you need to change probably only the reference section.

By default it performs following:

- discovers all event logs on all computers

- monitoring is enabled by adding an override for the group "MCS EventLog monitoring computer group" to enable the discovery "Event Log File Discovery" which is disabled by default

- after monitoring is enabled you can do some customization of thresholds (80 and 90% by default AFAIR) and you have a recovery task to perform local backup (you can override script parameters, which include folder to which to backup and compression bool value - for compression I use compress.exe from Windows ResKit - it needs to be locally on the server in target location of backed up event logs).

You can then expand the MP further to add automatic recovery, shipping saved logs remotely etc.

It also ships with some default views.

Here is a preview:

Enjoy. As always please do thorough testing and customization before using it on production.

I am posting a set of console extensions that will add right click functionality similar to that in SMS 2003. It is basing on great scripts from myitforum posted by Cory (http://www.myitforum.com/articles/8/view.asp?id=7099). The vbs scripts are taken unmodified from his solution, I only provide XML files and some additional actions based on WMI. I will try to add SCCM specific actions soon (e.g. Software Update detect cycle force etc.). Here is a copy to get you going.

Aside from the above basic actions:

Re-Assign the Site Code
Restart the SMS Agent Host service
Regenerate the SMS Client GUID
Re-Run Advertisements without modifying the advertisement
Discovery
Software Inventory - Delta and Full
Hardware Inventory - Delta and Full
File Collection
Software Metering Usage
Refresh Machine Policies
Evaluate Policies
Update Windows Installer Sources
Change Port Number
Change Cache Size

It also has:

Ping

Get physical memory

Open client log folder

Disclaimer: I have only roughly tested these extensions, prior to production use internal testing and customization is advised. For deploy refer to Readme.txt

[update]:

An MP update is now available which corrects this issue http://www.microsoft.com/downloads/details.aspx?FamilyID=008f58a6-dc67-4e59-95c6-d7c7c34a1447&DisplayLang=en

http://blogs.technet.com/cliveeastwood/archive/2008/03/26/mp-refresh-active-directory-management-pack-6-0-6278-10.aspx

There is an issue with SCOM 2007 SP1 RC version. If you are using ADMP to monitor your Active Directory, and the DCs have non EN-US regional settings the value returned from * Last Bind (e.g. PDC last bind, DC last bind) Monitors is x1000. This means you will get many alerts, even though your infrastructure is in healthy state. If you run the AD General Response diagnostic task you see that the response time is way below thresholds. Thanks to Pontus (http://blopon.blogspot.com/) for posting this on newsgroups - thought I would share it here as well.

Sample Alerts affected:

"The AD Last Bind latency is above the configured threshold"

"The Op Master PDC Last Bind latency is above the configured threshold"

"The Op Master RID Master Last Bind latency is above the configured threshold"

So you are migrating from MOM 2005 to 2007 and do not want to go through the hassle of changing your firewall ACLs? Default port for agent to Management Server communication is 5723. To modify it to MOM 2005 default (1270) or any other of your choice, perform these tasks:

• OperationsManager database – this modification is needed for new agents to get properly installed from UI console: GlobalSettings table find the row with value 5723 and change it to your port
• RMS server: HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Microsoft Operations Manager\3.0\Server Management Groups\CPDMGM change “Port” setting
• Management servers: HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Microsoft Operations Manager\3.0\Server Management Groups\CPDMGM\Parent Health Services\0 change “Port” setting
• In case of manual agent install either specify through command line parameter SECURE_PORT=your port or after default install you can always modify it in registry: HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Microsoft Operations Manager\3.0\Agent Management Groups\CPDMGM\Parent Health Services\0 change “Port” setting

Thanks to Anandha Ganesan and Gavin Sharpe ffrom Microsoft or their help and input on this!

When you get to deploying SCCM, there is a couple of issues you may see. Below I summarize some of my experiences - hope this helps:)

1. For all SCCM RTM sites installed after Dec 1, 2007 you will not see any status messages

This is due to a bug in date calculation. A fix is available and must be installed on all sites experiencing this issue. If you install new child sites after applying this fix to parent site AND specify to transfer files over network you will not need to apply this fix. However if you install from RTM media - you will still need to apply the fix:

http://support.microsoft.com/kb/945898/en-us

2. Manually exchanging the site key between sites when AD schema is not extended

The SCCM RTM documentation and the website: http://technet.microsoft.com/en-us/library/bb693690.aspx have a procedure for manual key exchange between sites. However for my simple parent (primary) - child (secondary) setup I found the procedure to be a bit misleading.

It reads:

To manually transfer the child site public key to the parent site

1. While logged on to the child site, open a command prompt and navigate to the location of Preinst.exe.

2. Run the following command to export the child site’s public key: Preinst /keyforparent

3. The Preinst /keyforparent command places the public key of the child site in the <site code>.CT4 file located at the root of the system drive.

4. Move the <site code>.CT4 file to the parent site's <install directory>\inboxes\hman.box directory.

To manually transfer the parent site public key to the child site

1. While logged on to the parent site, open a command prompt and navigate to the location of Preinst.exe.

2. Run the following command to export the parent site’s public key: Preinst /keyforchild.

3. The Preinst /keyforchild command places the public key of the parent site in the <site code>.CT5 file located at the root of the system drive.

4. Move the <site code>.CT5 file to the <install directory>\inboxes\hman.box directory on the child site.

However I found this procedure to work in my setup (see inverted child<->parent wording in point 4):

To manually transfer the child site public key to the parent site

1. While logged on to the child site, open a command prompt and navigate to the location of Preinst.exe.
2. Run the following command to export the child site’s public key: Preinst /keyforparent
3. The Preinst /keyforparent command places the public key of the child site in the <site code>.CT4 file located at the root of the system drive.
4. Move the <site code>.CT4 file to this child site's <install directory>\inboxes\hman.box directory.

To manually transfer the parent site public key to the child site

1. While logged on to the parent site, open a command prompt and navigate to the location of Preinst.exe.
2. Run the following command to export the parent site’s public key: Preinst /keyforchild.
3. The Preinst /keyforchild command places the public key of the parent site in the <site code>.CT5 file located at the root of the system drive.
4. Move the <site code>.CT5 file to this parent site’s <install directory>\inboxes\hman.box directory.

Excerpt from hman.log on my parent site

Attempt to copy the CT4 file as per original documentation fails:

Processing C:\Program Files\SCCM2007\inboxes\hman.box\S01.CT4 file, containing 1 keys.                SMS_HIERARCHY_MANAGER     1/11/2008 10:56:18 AM  4940 (0x134C)

CPublicKeyLookup::UpdateCurrentKey("S01", "0602000000A40000525341310004000001000100ED4AF40FA571B65BCB729D62A33D7B75B9B5613F224A0C39FE0B31B992924D3D36DB35626A297183B9FAFAE4CE02FF95DE3AC580995AFD07D1E746762767465FB4EF5C38168273AF9EF6F38F1F63F505062BA98618F02F99D01DDF097013E13C932DF500AEE38E415F68171ACD7F466F897DB96A5E243DBF1DC6750184FB1CB6")         SMS_HIERARCHY_MANAGER     1/11/2008 10:56:18 AM  4940 (0x134C)

CPublicKeyLookup::UpdateCurrentKey() Checking C:\Program Files\SCCM2007\inboxes\hman.box\pubkey\S01.pkc for Key0      SMS_HIERARCHY_MANAGER     1/11/2008 10:56:18 AM  4940 (0x134C)

CPublicKeyLookup::UpdateCurrentKey() Checking C:\Program Files\SCCM2007\inboxes\hman.box\pubkey\S01.pkc for Key1      SMS_HIERARCHY_MANAGER     1/11/2008 10:56:18 AM  4940 (0x134C)

CPublicKeyLookup::UpdateCurrentKey() Aging Key0       SMS_HIERARCHY_MANAGER     1/11/2008 10:56:18 AM  4940 (0x134C)

CPublicKeyLookup::UpdateCurrentKey() Updating Key0               SMS_HIERARCHY_MANAGER     1/11/2008 10:56:18 AM  4940 (0x134C)

No parent site to forward CT4 file C:\Program Files\SCCM2007\inboxes\hman.box\S01.CT4 to.  Deleting.                SMS_HIERARCHY_MANAGER     1/11/2008 10:56:18 AM  4940 (0x134C)

Using P01.CT5 file succeeds:

Processing C:\Program Files\SCCM2007\inboxes\hman.box\P01.CT5 file, containing 1 keys.    SMS_HIERARCHY_MANAGER    1/11/2008 11:23:01 AM    4940 (0x134C)
CPublicKeyLookup::UpdateCurrentKey("P01", "0602000000A400005253413100040000010001000BEA74D6492BA114831648F56EBD42EF4542FFE358FDD4CEC6FA9473089641A5628DAB9395F767C9F1418A5BFCC8FF94E4B63B004FE65AC5BA3FEDA606363B3BFE539C2D2B2AD41A2B6C3681D96425F07B396E358DB3C89BA898F0B7A1194005D20C00E2B2D414658E78AA36C3C972276C4D885F3C366FF571FFB3E2142BF7CD")    SMS_HIERARCHY_MANAGER    1/11/2008 11:23:01 AM    4940 (0x134C)
CPublicKeyLookup::UpdateCurrentKey() Checking C:\Program Files\SCCM2007\inboxes\hman.box\pubkey\P01.pkp for Key0    SMS_HIERARCHY_MANAGER    1/11/2008 11:23:01 AM    4940 (0x134C)
CPublicKeyLookup::UpdateCurrentKey() Matching Key found in iteration 0.  Updating Date.    SMS_HIERARCHY_MANAGER    1/11/2008 11:23:01 AM    4940 (0x134C)
Successfully forwarded CT5 file to child site S01.    SMS_HIERARCHY_MANAGER    1/11/2008 11:23:01 AM    4940 (0x134C)

Manual exchange of public keys is needed only if you don't exchange the AD schema. I would really encourage to extend AD schema, large deployment without this, if you don't have a solid WINS(who uses that these days?!)+DNS infrastructure, will give you big administrative overhead. Read more at: Decide If You Should Extend the Active Directory Schema.

3. Software distribution mandatory assignment does not seem to work (if you are in a +GMT zone).

You may find that you created a package, program and a mandatory "as soon as possible" advertisement, but your clients are not receiving this package. This is because default setting on the advertisement is to use UTC time and not client time. Change to client time and it should go much quicker:

4. Wake On Lan does not seem to work (if you are not using Daylight Savings Time on the server)

See hotfix http://support.microsoft.com/kb/944542 for resolution.

5. Extending the SCCM console with basic tools

Remember back in the SMS 2003 days we had nice tools that enabled right-click on computers/collections to ping/force policy refresh etc. You will find that now extending the SCCM console is easier. It all comes down to an XML file placed in a proper folder. We can soon expect many extensions (also those from SMS) to be ported to SCCM console. As an example a ping extension for computers from SCCM console.

Create a file called e.g. PingWorkstation.xml:

<ActionDescription Class="Group" DisplayName="SMS Tools" MnemonicDisplayName="SMS Tools" Description="SMS tools for agent" SqmDataPoint="53">
  <ActionGroups>
      <ActionDescription Class="Executable" DisplayName="Ping Workstation" MnemonicDisplayName="Ping Workstation" Description="Will ping a workstation">
        <Executable>
          <FilePath>cmd.exe</FilePath>
          <Parameters> /k Ping ##SUB:NetBIOSName##</Parameters>
        </Executable>
      </ActionDescription>
  </ActionGroups>
</ActionDescription>

Place the file in c:\Program Files\SCCM2007\AdminUI\XmlStorage\Extensions\Actions\7ba8bf44-2344-4035-bdb4-16630291dcf6\ and restart the console. The GUID is reference to the resource object in collections. A collection would have these GUIDs:

dbb315c3-1d8b-4e6a-a7b1-db8246890f59 - top level collection
fa922e1a-6add-477f-b70e-9a164f3b11a2 - subcollections

The GUID referring to the SCCM resource object. Additional extensions can be made with the use of SDK documentation in a similar fashion. SDK is now in December CTP available on https://connect.microsoft.com/availableconnections.aspx site (look for SCCM SDK beta - you need to sign in with a live account). RTM should appear somewhere this quater.

If you have any other first deployment tips I encourage to share.

6. You cannot edit a task sequence, getting an error from SCCM UI (only if your registered organization or username of computer where SCCM console is installed is null or " ")

When a computer on which the admin console is running has null or “ “ (space) in the registered for user or organization field (specified during computer setup) sequence tasks generated from the UI cannot be edited. To resolve export the task to XML and modify the registeredUserTo (or company) field to assign it some value (for the editor it cannot be null). To assure problem does not happen on workstation again change the registered field in registry: http://gettoknowyourpc.blogspot.com/2006/07/change-registered-windows-info.html

I am definitely not a "sales" type of guy ;) and usually know nothing about licensing, but people do ask a lot about SCCM pricing and licensing so I needed to find out.

Basically this is the link you want to read: http://www.microsoft.com/systemcenter/configmgr/howtobuy/default.mspx.

However there are a few FAQs not answered by that URL:

Q: Is there an SCCM server license needed for secondary site server?

A: No, "The advantages of using secondary sites are that they require no additional Configuration Manager 2007 server license and do not require the overhead of maintaining an additional database.". See: http://technet.microsoft.com/en-us/library/bb632547.aspx

Q: OK, do I need a client license then on the secondary site server?

A: No, you do not need to install an SCCM client on the SCCM secondary site (and hence don't need an SCCM CAL)

Q: What is the difference in Standard and Enterprise ML for servers?

A: Standard includes SCCM functionality except for DCM for workloads other then Base OS, file&print, networking. And it can only be used on a single installation per-physical machine. That does not apply to Enterprise ML: full workload DCM and can be used to manage any number of OS installations on a single physical server.

For System Center Server suite bundles see: http://www.microsoft.com/systemcenter/svrmgmtsuites/howtobuy/default.mspx

Q: Can I use the Client ML on server OSes?

A: No.

Q: Do I need the Client ML for each of my client OSes?

A: Only if that client OS does not have a Core CAL, which is rare for large customers: http://www.microsoft.com/calsuites/core.mspx.

My friend Jaro (our Xbox Bike Team Director ;)) started the new website for our XBOX bike team: http://www.xbox360biketeam.pl. Currently there are only photos there, but we hope to add some content in the future, you are welcome to visit. Meanwhile some photos still back from the summer (oh God, I am missing that weather).

Here I am on one of the marathons (still on my good, old Trek bike):

And here I am at the finish line (this time on the new, shiny... well not after a race;)) Mongoose):

Scary huh? I don't normally look like that - I usually try to look more serious then I really am ;) This photo is just after going "all out" on an 80km marathon that's why my face has a weird expression;)

But now winter has come to Poland (its subzero but no snow except mountains...) and its beginning to be increasingly hard to keep biking. I really don't know nor understand how I can still enjoy biking nor motivate myself to go out - its cold, wet, dark (in Warsaw its now dark already at 4 PM) and the only place at such times where you can ride is the asphalts of the city:( However somehow I still find it as a pleasure - this is one of the things that amounts to my weirdness I guess;)

ANYWAY: I wish everyone a big, happy, joyful Christmas with your computers stored AWAY. (At least that is a definition of a good holiday for me - but I know there are some definitely geekier ppl out there - cheers:))

Also all the best for the new 2008 year - lots of love, happiness and success. Many thanks for this 2007 year, also for reading my blog. I notice some people actually do read it and this is the real motivation for me to keep it running amongst many other duties, responsibilities and hobbies.