Gregory S. MacBeth -

Completly Unrelated to SharePoint But Bought a New Motorcycle Today CBR1000RR Fireblade

gregmcb — Wed, 17 Jun 2009 01:02:00 GMT

This bike is awesome at 160 Horses.

http://cid-732d673e5904257f.skydrive.live.com/self.aspx/FirebladeCBR1000RR/Fireblade%2008.jpg

I am so excited....

Important TSQL, Databases & Tables for Troubleshooting

gregmcb — Fri, 05 Jun 2009 18:09:00 GMT

IMPORTANT - DIRECT MODIFICATION OF ANY SHAREPOINT DATABASE IS UNSUPPORTED. This doesn't mean you can't peak though.

Important Tables I have found usful for troubleshooting. I will publish as I define more.

Contentdb

ImmedSubscriptions / SchedSubscriptions table - Entry is made when an alert is fired. Good place to check when alerts are not working.
Eventcache - Subscriptions not processed - select * from eventcache where EventData is not null
EventLog - Subscriptions Processed - select * from eventlog where ListID = 'xxx'

Configdb

TSQL Commands

Importing Publishing Sites To Different Levels Unsupported

gregmcb — Tue, 02 Jun 2009 18:32:00 GMT

So if you move a site from a SPWeb to a SPSite and it is a publishing site you should expect an array of problems and unpredictability. In otherwords don't make a subweb a top level site.

THIS IS WHY YOU SHOULD NOT DO THIS! IT IS UNSUPPORTED! http://support.microsoft.com/default.aspx/kb/968483

Gary Lapoint does have a blog on this but it requires direct db modification which again is unsupported. Although kudos for the effort as it is a nice hack.

http://stsadm.blogspot.com/2007/09/convert-sub-site-to-site-collection.html

I am working on coming up with a supported path to do this, but my hopes are low @ this time.

MOSS SP2 Upgrade Process - Step By Step

gregmcb — Tue, 02 Jun 2009 18:29:00 GMT

MOSS SP2 Upgrade Process:

As defined and cleansed by http://technet.microsoft.com/en-us/library/cc263467.aspx#section1

Database Maintenance:

1. Run MOSS BPA and resolve any problems defined

2. Stop the www service and disconnect any users

3. Run DBCC CHECKDB on each database and repair any problems.

4. Remove Orphaned Items from databases using stsadm –o databaserepair

5. Defragment and update statistics of each database

6. Make sure that there is adequate hard drive space in your database files volumes, tempdb volumes, and Windows temporary folder on the servers running SQL Server, front-end Web servers, and application servers.

7. After you have backed up all of your databases, use the SQL Server DBCC shrinkfile command to free unused log space, making the logs as empty as possible. Shrinking the Transaction Log

Upgrade Pre-Steps:

1. Load Balance Site Collections and ensure your not exceeding site collections limits per content database and make sure content databases do not exceed 100 GB as recommended.

2. Make sure that you follow the recommendations concerning SQL Server page-fill factor and other storage planning best practices before you begin the upgrade.

3. Backup

a. Use the full backup operation from SharePoint Central Administration or Stsadm to back up search. The backup should include both the SSP and the index file.

b. Brick level backup the configuration and central admin content databases

c. Backup the SSO and every content database

d. Backup any Front End Customizations such as Templates, features, resources, webparts, etc..

4. OPTIONAL - In server farms that have a large number of sites, you will find that installing a software update with the content databases attached is not practical in terms of downtime. You may want to pursue this route for any content db > 20 GB as well. In order to minimize the downtime, we recommend that you perform the additional steps of running the Stsadm preparetomove operation and then detach the content databases. Do not run against the CONFIG database, SSP or ADMINISTRATION CONTENT DATABASE!

a. Run this command for each content database

DO NOT FOLLOW THIS STEP IN POST SP1 WITH INFRASTUCTURE UPDATE AS IT WILL DO THE OPPOSITE AND FRY THE PROFILE SYNCHRONIZATION. IF YOU DO THIS THEN THE FIX IS:

CURTESY OF TODD CARTER

PrepareToMove sets 'Moving' to True in the SiteSynch table of the SSP DB which signals to the profile sync job to ignore the DB which has the effect of not allowing profile syncs to occur for this DB. After attaching a Content DB back into the IU farm 'Moving' is still set to True. Even after the sync timer job fires and the stored procedure profilesynch_RegisterSiteToSynch is executed this content DB will still have its 'Moving' set to True.

The fix/resolution is to run PrepareToMove again (ok so I lied there is a case where you want to run it with the IU) but this time with the –undo option. The -undo command is harmless and it will set 'Moving' back to False and allow profile syncs to this content DB to start running again.

http://blogs.msdn.com/toddca/archive/2009/01/30/preparetomove-away-from-running-this-command.aspx

stsadm -o preparetomove -contentdb <db_name>

b. If you have multiple Web applications you must run.

stsadm -o deletecontentdb -url http://servername -databasename

c. In this operation, -url specifies the Web application from which the content databases will be detached and -databasename specifies the name of content database to be detached.

After you upgrade your server farm, you must attach the content databases back to the server farm. You can only attach one content database to the server farm at a time, because when you attach the databases to the upgraded server farm the content database is upgraded automatically.

5. To deploy the upgrade you must have these memberships

a. Member of the Administrators group on the Web server computer.

b. Member of the Administrators group on the server running SQL Server or be granted the fixed database role db_owner to all SharePoint Products and Technologies databases.

Install Sequence

1. Add the account for the SharePoint Central Administration v3 application pool identity to the Administrators group on each of the local Web servers and application servers and then log on by using that account. These changes are only required for installing the update and then running the SharePoint Products and Technologies Configuration Wizard to complete the upgrade.

2. The following permissions are required to run psconfig

1. Member of the Administrators group on the local computer that runs Office SharePoint Server 2007.

2. In SQL Server, the account must be:

1. Authorized to access all SharePoint Products and Technologies databases.

2. Granted the Database Creators (dbcreator) fixed server role.

3. Granted the Security Administrators (securityadmin) fixed server role.

Install the software update

This section includes all of the procedures required to install a software update successfully in any size server farm. If you are in a large server farm you should read the Large farm optimization section later in this document.

The following procedure provides the steps to:

· Make all software update files available on all servers in your server farm.

· Complete the update from one of the servers hosting the Central Administration site.

· Finish updating the remaining servers in the server farm.

Note

You must perform steps 1 though 8 from the following procedure on every Office SharePoint Server 2007 server in the server farm before you complete the installation on any one Office SharePoint Server 2007 server.

To install a software update

1. Disconnect users from the server farm by stopping the World Wide Web Publishing service (w3svc) on all Web servers.

2. Download and install the appropriate Windows SharePoint Services 3.0 software update for all servers in your server farm.

Important

At the end of the Windows SharePoint Services 3.0 software update, do not run the SharePoint Products and Technologies Configuration Wizard.

3. Download and install the appropriate Office SharePoint Server 2007 software update for all servers in your server farm.

4. At the end of the software update installation, the SharePoint Products and Technologies Configuration Wizard starts.

Note

If the wizard does not start automatically, click Start, point to All Programs, point to Administrative Tools, and then click SharePoint Products and Technologies Configuration Wizard.

5. On the SharePoint Products and Technologies Configuration Wizard Welcome page, click Next.

6. In the dialog box that notifies you that some services might need to be restarted during configuration, click Yes.

7. On the Completing the SharePoint Products and Technologies Configuration Wizard page, click Next.

8. When the dialog box about installation in a server farm appears, do not click OK. Instead, leave each server with the following dialog box displayed:

You must run Setup to install new binary files for every server in your server farm. If you have multiple servers in your server farm, run Setup and the configuration wizard on the other servers now, and then return to this server and click OK to continue.

9. When the dialog box from the previous step is displayed on all the application servers and Web servers in the server farm, use one Web server that hosts the Central Administration Web site to finalize the installation.

Notes

We recommend that you install the software update on an application server that is hosting the Central Administration Web site, then the other application servers, and finally the front-end Web servers.

If you are hosting your Central Administration Web site on a front-end Web server, then we recommend that you install the software update on the front-end Web server hosting the Central Administration Web site, then the application servers, and finally the remaining front-end Web servers.

10. On the server you selected in the previous step, click OK.

11. On the Configuration Successful page, click Finish.

12. After you have finished updating one Web server that hosts the Central Administration Web site, you should follow the procedures in the "Verify installation" section on this one Web server to ensure that the software update installation was successful.

13. Continue updating the remaining computers in the server farm, one at a time, by clicking OK in the dialog box.

Note

It is important that the SharePoint Products and Technologies Configuration Wizard perform the configuration procedures on only one computer at a time.

14. When the software update installation and configuration is complete on all the Web servers in the server farm, make the Web servers available to users by manually starting the World Wide Web Publishing service on each server on which you manually stopped the service.

3. OPTIONAL only If you detached content dbs use the following command line to add them back to the farm

stsadm -o addcontentdb -url <http://backupservername:port> -databasename <ContentDBName>

WE HAVE CHANGED THE LOGIC AND RUNNING THIS COMMAND WILL NO LONGER TRIGGER A FULL CRAWL ON A POST SP1 WITH THE INFRASTRUCTURE UPDATE.

4. Verify the server install by running the following command against the config database

SELECT * FROM Versions

· OPTIONAL Force an upgrade by running one of the following commands if you encounter problems

1. psconfig –cmd upgrade –inplace b2b –wait –force

2. psconfig -cmd configdb -connect -server <SQLServerName> -database SharePoint_Config_<dbname> -user <domainusername> -password <password> -cmd helpcollections -installall -cmd secureresources -cmd services -install -cmd installfeatures -cmd applicationcontent –install

· Next Install The April 09 Cumulative Updates

If you have enterprise features installed, renter the PID as there is a bug that will timeout MOSS in October.

IOPS

gregmcb — Tue, 02 Jun 2009 18:16:00 GMT

Because I get asked so often about what are the best practices for IOPS. I thought I would summerize from our article:

http://technet.microsoft.com/en-us/library/cc850692.aspx

Type	RAID level	Spindles	Optimization
TempDB	[RAID 1+0] - 10% of total database size	2 IOPS/ GB	Write optimized
Transaction Logs	[RAID 1+0]	2 IOPS/GB	Write optimized
Search property store (Search database)	[RAID 1+0]	2 IOPS/GB	Read/Write optimized
Content databases	[RAID 1+0]	0.75 IOPS / GB	Read optimized

For more information, see Physical storage recommendations (Office SharePoint Server) and Planning and Monitoring SQL Server Storage for Office SharePoint Server: Performance Recommendations and Best Practices (white paper).

A healthy database server has enough headroom for databases and log files, plus enough capacity to keep up with requests. Use the recommendations in the following list to keep database servers performing optimally.

Pre-grow all databases and logs if you can. Be sure to monitor the sizes so that you do not run out of disk space.
Do not overload database servers by using too many databases or data. Use the following guidelines:
- When using SQL Server mirroring, do not store more than 50 databases on a single physical instance of SQL Server.
- Limit content databases to 100 GB.
Defragment and rebuild indices daily, if you can absorb the downtime required to rebuild.
Monitor the database server to make sure that it is responding appropriately and is not overloaded. Key performance counters to monitor include the following:
- Network Wait Queue: at 0 or 1 for good performance
- Average Disk Queue Length (latency): less than 5 ms
- Memory used: less than 70%
- Free disk space: more than 25%

Know Your Updates

gregmcb — Thu, 23 Apr 2009 22:07:00 GMT

Several people have encountered the problem of confusing cummulative updates(Infrastructure Update KB951297) where we update all of MOSS and the Oct and Dec patches where we update only certain components.

For clarity in a Cummulative Update we update the following components:

coreserver.msp - Core MOSS

dlc.msp - Workflow

ifswfe.msp - Form Server

pjsrvapp.msp - Project Server

pjsrvwfe.msp - Project Server WFE

However, in the months of October, November and December of 2008 we did not release a cummulative update. We released seperate packages that included coreserver.msp (KB957693), dlc.msp (KB958569) and Ifswfe.msp (KB957694).

So when you apply updates make sure you recognize the difference between a Cummulative update and a monthly update where we have a seperate package for each component as described above. Furthermore, when you apply some of these patches we did not document that psconfig would need to be run. This is specifically the case with KB957694.

Regression in July Cummulative update For SharePoint

gregmcb — Tue, 05 Aug 2008 23:31:00 GMT

Beware that we have come across a regression in the July patch involving reverse proxy. If you are using this functionality you should hold off until a patch is created.

Also the product team has created a blog just for the July patch. I would encourage you to read it before applying the patch.

QUESTION - I am Program Manager Charged with Implementing SharePoint - Now What?

gregmcb — Mon, 26 May 2008 08:44:00 GMT

This post is targeted towards the large customer and may not completely apply to medium or small organizations. This my list of the top ten priorities that a PM should focus and consider when given this task. This is a high level overview based upon my experience and I am sure we could detail and add more but ten is good to start.

PRIORITY ONE - As a PM you must start to develop the SharePoint Govenance Plan by using the the Microsoft Published Resource Center. This wlll allow you to do acoomplish several key items which will make your program a sucess:

Protect the enterprise, such as by helping to ensure regulatory compliance.
Greatly increase the usability of a solution. For example, the governing body in an enterprise can define metadata requirements for all content that makes it easier to categorize and search for the content across the enterprise.
Increase an organization's efficiency — for example, by requiring training as a prerequisite to becoming the administrator of a SharePoint collaboration Web site.
Build your Technical Team Focused upon the service and not individual old world silo based management virtualized teams for core skills
Establish a Service Model with Solid Service Level Agreements with the Business Unit Leaders (BULS) and a Solid Service Level Map

PRIORITY TWO - Build Your New Service Based Team.

This is going to be hard becuase you are going to have to convince your leadership that SharePoint is and should be managed with the same priority as how you manage your Exchange team and its not going to be cheep. The first problem I see organizations run into is who manages the team. The virtualized team doesn't seem to work IMHO becuase the individual managers don't have the same stake as you. My advice for whats its worth is that their should be one manager focused with direct control over the following roles: SharePoint Admins, SharePoint Developers, SQL DBA, the Architect, Windows & Cluster Admins, IIS Admins, and an Office Admin. Then their needs to be a virtualized team connected to the MOSS core team with these roles: SCOM Admin, Network Admin, SAN\Storgae Admin, Security Admin. There are ther roles that will and could be added but tis covers the core.

This is why the Program Manager needs to be a strong leader. A quote from John Fenton's Book describes exactly what you need:

"Managers think incrementally, whilst leaders think radically. "Managers do things right, while leaders do the right thing.". This means that managers do things by the book and follow company policy, while leaders follow their own intuition, which may in turn be of more benefit to the company. A leader is more emotional than a manager . "People are governed by their emotions rather than their intelligence". This quotation illustrates why teams choose to follow leaders.
"Leaders stand out by being different. They question assumption and are suspicious of tradition. They seek out the truth and make decisions based on fact, not prejudice. They have a preference for innovation."

I share this quote becuase it is the best statement I have found that describes exactly what type of person needs to be the Program Manager. As the PM for SharePoint you are going to have convince your superiors of the innvovative nature of SharePoint and why they can't think of it in the traditional fashion.

What I will provide is a solid proven example of how the United States Marine Corps has taken this strategy to fact by being agile enough to build a services based unit called the Marine Air Ground Task Force (MAGTF). MAGTFs are a balanced air-ground, combined arms task organization of Marine Corps forces under a single commander that is structured to accomplish a specific mission.which we would define in IT as a Service. If the government can pull this off then it can certainly be implemented with even greater success by the private sector.

A good way to understand this strategy of leadership would be the following book a it applies the Marines Way to business which has more in common than you may think.

PRIORITY THREE - It's not going to be cheep, think about how much it costs to runa good Messeging Environment as its the closest comparison I can provide.

Use your Governance Plan and SLAs to drive getting what you need to meet the demands of the business. If the budget is not there then go back to the governance plan and SLA's and ask the BULS what they want to cross off their list of requirements. If they say nothng, then the only thing you can say is this is what the team determined we need to provide the service you asked for. If your provide less then we have to cut the service. Where do you want to start?

PRIORITY FOUR - Plan for explosive growth

Once people see SharePoint, they will use it and accept it just like people did email in the 90's.

Use x64 Hardware starting from SQL, to Application Servers (Index, PerformancePoint, Excel Server, InfoPath Server, etc), to WFE
Plan for the best disk subsystem you can get for SQL in terms of IOPS weather it be DAS or SAN(which should be dedicated and not shared)
Decide if you you will Cluster or Mirror SQL
Use 8GB RAM WFE on x64
Develop of Main Farm with Multiple Child Farms for really big organizations by planning a good Taxonomy from the start

PRIORITY FIVE - Training

Train your Admins
Train your Developers
Train your End Users (Some say 75 % of your Trainign Budget Goes here)
Train your Staff in Integration Troubleshooting

PRIORITY SIX - Plan and Build a Dev, Test and Production Environment that is Exactly the Same

You need this to test custom code
You need this to test hotfixes and service packs

PRIORITY SEVEN - Build Unit Tests and Conduct a Full User Acceptance Program

Based upon your SLAs build unit tests and require BULs to signoff on acceptance
Test Your environment before, during and after you Go Live

PRIORITY EIGHT - Operate & Maintain Your Environment

Engage Microsoft Premier Support for Yearly MOSS Risk Assessments at a minImum. I would recommend a MOSS Risk Assemesnt after any major change or upgrade to hardware or Service Packs.
Do multiple Triail runs of any major upgrades such as Service Packs, Hardware Changes, as well as Storage and Network modifications. Do enough test where you have a step by step document and can complete the upgrade in test with zero flaws.
Train Your Team to Use System Center Operations Manager (SCOM) to proactively manage your environment and identify the KPIs before production.
When you do have a problem DO NOT FOCUS ON THE SYMPTOM. You should FOCUS ON FINDING THE ROOT PROBLEM!
Ensure you have a Service Level Map, this way you know exactly what and who will be impacted by any changes made.
Build a documented and tested at lest quarterly DR Plan

PRIORITY NINE - Control Growth and Set Standards

When I talk to customers, I often hear we had a lot of organic growth. What I translate this to is our Rose Garden is full of weeds!

Define what strategies you will use to customize SharePoint. Each has a set of pros and cons that go along with them that will

SharePoint Designer
Site Definitions
Template Modifications\Additions
Features
WebParts
Event Handlers
Workflows
etc...

PRIORITY TEN - Have Fun Becuase You Are Leading a Project That Will Impact Your Company For Years To Come With The Same End Result as Email Had in the 90s. No Pressure:)

SSL and Authenticode Causes CRL lookups if Your Machine Cannot Access the CRL for Verification

gregmcb — Wed, 07 May 2008 01:52:00 GMT

Special Thanks to Tim Walton for Researching And Contributing This.

PROBLEM:

Slow response from custom SharePoint Application components may be happening because we are seeing a timeout going to the Certificate Revocation List. As I understand the behavior is that we may wait for up to 45 seconds before .NET concludes that the CRL is unavailable. This conclusion may then be cached for up to a minute before a fresh attempt is made to contact the CRL again.

THEORY:

To verify that the certificate used to sign the .Net package hasn’t been revoked, .Net will attempt to download the Certificate Revocation Lists from both of these URL’s:
http://crl.microsoft.com/pki/crl/products/CodeSignPCA.crl
http://crl.microsoft.com/pki/crl/products/CodeSignPCA2.crl
If the .Net application cannot reach those end destinations then the .Net package assumes that an attempt to block verification has occurred and that the .Net packages signed by the certificates are not to be trusted. End result, the .Net packages may run slowly.

PREFERRED RESOLUTION:
Allow the server and the services accounts to access the crl.microsoft.com domain.

RESOLUTION 2:
Turn off CRL checking by creating a turnoffCRL.reg file with this as the content:

Windows Registry Editor Version 5.00
[HKEY_USERS\S-1-5-20\Software\Microsoft\Windows\CurrentVersion\WinTrust\Tru¬st
Providers\Software Publishing
"State"=dword:00023e00

Note that this would be considered to be compromising the integrity of the .Net signed packages stuff as we would have prevented any compromised certificates from being disabled automatically.

RESOLUTION 3:

Manually add the CRLs from the above Urls to the server manually (assumes you’ve downloaded Them locally first):
certutil -addstore CA CodeSignPCA.crl
certutil -addstore CA CodeSignPCA2.crl

QUESTION: I Need a Good SharePoint Admin. What should I be looking for?

gregmcb — Sat, 26 Apr 2008 22:35:00 GMT

I thought I might spend of a few minute with my answer to this question that seems to be coming up with managers more often with each passing day. My response is complicated to explain to becuase the answer is always you are asking the wrong question and exposing yourself to risk by limiting your attention to the traditional focus.

First, lets start with the my opinion. SharePoint is a platform similar to ASP.NET or Windows that you build applications and services on. SharePoint respresents the first true class of products by Microsoft that completely blurs the line of Admin and Developer. Thinks of it in these terms. VBScript was the first technology Microsoft developed beyond bacth files that automated the administrators job. Since then we have made the gigantic leap to .NET, C# and Powershell. These technologies have changed the game and the sooner Engineers, Developers and Management understand this the better. This is not to say that the traditional role of application developer or adminstrator is dead. There will always be the need for specialist that focus at a deep level upon building applications and administratering them. However, the future Star IT employee is the person that can engineer and write code across all parts of the solution.

So when asked, by my customers how and what should I be looking for in an finding an administrator for SharePoint, my response is always STOP looking for an adminstrator and start searching for the NEXT GENERATION IT STAR that can code and adminster your solution. The simple fact is that when looking for that star you need someone who understands common adminstration skills (installing windows, patching, monitoring and configuring the server as well as managing the hardware, adminstering SQL, Networking and IIS) and then they need developer skills (.NET Coding, Powershell, WindowsServices Distributed Computing, XML,CAML, SQL Coding, AJAX, JavaScript, HTML, etc...). This is the perfect solution but it will take time for the world to get to this point and not everybody in IT can accomplish this task.

The second point I make is that the distributed engineering strategy of managing an IT environment by organizing along traditional product responsabilities such as having a Core Windows Team, Client Team, SQL Team and SharePoint Team is not going to result in a cost effective succesful solution. To be clear I am not saying that having a Core Windows Team doing tasks such as Building Servers and Installing the OS is wrong. Nor is having a core Client, SQL or SharePoint Team is wrong. What I see in the real world is the creation of the virtual team made up of members of each team. In some environment this works fine, but in larger environments this approach will only result in less than agile and fragmented team that creates more problems and costs to the overall operation. What needs to happen is the opposite approach and adoption of a service based approach.

My suggestion upon this matter is to look at the technologies from a service provided approach. For example, SharePoint is a collaborative platform and demands in a larger environment that the entire scope of technologies be managed as a distinct unit with a solidified plan under a unified direct management structure. Large organizations are agoing to bulk at this statement, BUT THEY ARE THE ONES THAT SHOULD BE LISTENING THE MOST. This trend started with Exchange and has grown into far more by the evolution of the technology. This is a pretty radical idea for many organizations and push back is always expected BUT it can be over come with strong leadrehship by the Project Manager. Regardless, this approach has to happen at some point. So an appropriate team would include an organization model based upon service rather than product. Even more important is that finding that IT STAR will be hard at least for a while until the workforce adopts the approach that being an integration expert is where the real money and possibaility of sucess exists. So until that occurs and even after that occurs organizations need to move from the virtual team based upon products to services where the priorities and goals are unfied.

The answer to these problems is simple and is determined by the politics, size and scope of your organization. In the large organization there should be a collaboration team which includes Windows, SQL, SharePoint, and Developer Resources under a single manager combined with a traditional virtualied team of Network, SAN, and security engineers with a unfied business strategy and goals driven by a strong SLA. This means the organization needs to start doing the opposite of the product based virtual team but rather reorganize into the service based team where for example the Collaboration SQL expert is a vrtual member of the companies core SQL team. This doesn't mean that the organization core SQL team drives the collaboration implementation of SQL but rather the Collaborative SQL team member communicates with the core team but the final decision is made by the collaboration SQL expert and project manager. There might not be even a core product team in the organization, as a true service based model would allow for the SQL team made up of members from all thekey services model based team.

The answer is complicated and the solution is even more complicated requiring organizations to reorganize teams and develop IT STARS. In short the traditional old school organization is going to flip with the initial idea and work required to implement what I am suggestion, but the end justifies the means and where this approach has been adopted the success is clearly obvious. The strategy is also well documented and promoted as being the future of IT.

The proof of this approach is not only well documented by IT Management organizations but also by history in general. A simple look at hostory shows that the the most agile organization is the always the most sucessful. For the sake of brevity, I will spare reader the countless examples I could provide from history proving this advice, but if you really want it send me an email and I will provide you the evidence.

Some good links supporting this advice are:

Workforce Management by Microsoft

Gartner IT Strategy

HP IT Strategy Management

HP Services Mnagement Framework

Microsoft Services Management Framework

ITIL Service Level Management

Decisions About When To Use a Farm, Web Application, Content Database or Site Collection

gregmcb — Wed, 23 Apr 2008 18:14:00 GMT

I wanted to say a few things that may help when deciding upon when a Farm, Web Application, Content Database and Site Collection are necessary. For the purposes of this discussion I am defining a Large organization as any organization implementing a medium to large server farm. A small organization would be one using a single server farm.

First think of these items in terms of how we suggest implementing Active Directory:

Web Farm is Like a AD Forrest

Web Application is Like a AD Domain

Content Database is Like an AD Partition

Site Collection is like an AD Organizational Unit

Web Farm:

You need to consider separate Web Farms for Very Large Organizations where political or Very Large Implementations are needed such as separate Shared Service Providers(Max 8 per farm) are required.

Web Application:

Consider Web Applications when you need to define Zones and Policies that are unique.

Content Database:

The recommended limit is 100 GB. The main thought here is Disaster Recovery time and SQL Performance.

Site Collections:

In smaller organizations, separate site collections organized by business units is appropriate. In a large organization, I would recommend a separate site collection for each high level team. Focus your efforts upon building a wide rather than deep hierarchy. Site collections will allow the IT department freedom to maintain just application itself without the worry of security or content hierarchy maintenance. The following is a list of what an individual site collection offers.

For the Users:

Dedicated Recycle bins
Dedicated usage Reports
Distributed administration (site collection administrators)
Dedicated search scopes, keywords, and best-bets
Custom feature deployments
Dedicated language translation maintenance
Dedicated galleries for web parts, master pages, content types, site columns, site templates, and list templates
Dedicated shared libraries, such as site collection images and site collection styles
Dedicated real estate (Self Containment)

For the IT Administrators:

Site quota templates
Distributed administration
Site locking
Database maintenance options
Backup / Restore abilities
Content Deployments
InfoPath forms services global template targeting

Some good links to read that may help:

· http://www.sharepointblogs.com/llowevad/archive/2007/11/09/when-to-use-site-collections-vs-sub-sites.aspx

· http://www.sharepointblogs.com/llowevad/archive/2007/06/25/site-collection-logical-architecture.aspx

MOSS SP1 Upgrade Process Step By Step

gregmcb — Mon, 21 Apr 2008 17:02:00 GMT

MOSS SP1 Upgrade Process:

As defined and cleansed by http://technet.microsoft.com/en-us/library/cc263467.aspx#section1

Database Maintenance:

1. Run MOSS BPA and resolve any problems defined

2. Stop the www service and disconnect any users

3. Run DBCC CHECKDB on each database and repair any problems.

4. Remove Orphaned Items from databases using stsadm –o databaserepair

5. Defragment and update statistics of each database

7. After you have backed up all of your databases, use the SQL Server DBCC shrinkfile command to free unused log space, making the logs as empty as possible. Shrinking the Transaction Log

Upgrade Pre-Steps:

1. Load Balance Site Collections and ensure your not exceeding site collections limits per content database and make sure content databases do not exceed 100 GB as recommended.

2. Make sure that you follow the recommendations concerning SQL Server page-fill factor and other storage planning best practices before you begin the upgrade.

3. Backup

a. Use the full backup operation from SharePoint Central Administration or Stsadm to back up search. The backup should include both the SSP and the index file.

b. Brick level backup the configuration and central admin content databases

c. Backup the SSO and every content database

d. Backup any Front End Customizations such as Templates, features, resources, webparts, etc..

a. Run this command for each content database

stsadm -o preparetomove -contentdb <db_name>

If you do not run this operation before you detach the content database, then the membership and profile information in the content database is static and will not be synchronized after upgrade. BUT IF YOU INSTALL THE INFRASTRURE UPDATE TOO THEN DO NOT RUN THIS COMMAND. THE FIX IS TO FOLLOW THE STEPS BELOW.

CURTESY OF TODD CARTER

b. If you have multiple Web applications you must run.

stsadm -o deletecontentdb -url http://servername -databasename

c. In this operation, -url specifies the Web application from which the content databases will be detached and -databasename specifies the name of content database to be detached.

5. To deploy the upgrade you must have these memberships

a. Member of the Administrators group on the Web server computer.

b. Member of the Administrators group on the server running SQL Server or be granted the fixed database role db_owner to all SharePoint Products and Technologies databases.

Install Sequence

2. The following permissions are required to run psconfig

1. Member of the Administrators group on the local computer that runs Office SharePoint Server 2007.

2. In SQL Server, the account must be:

1. Authorized to access all SharePoint Products and Technologies databases.

2. Granted the Database Creators (dbcreator) fixed server role.

3. Granted the Security Administrators (securityadmin) fixed server role.

Install the software update

The following procedure provides the steps to:

· Make all software update files available on all servers in your server farm.

· Complete the update from one of the servers hosting the Central Administration site.

· Finish updating the remaining servers in the server farm.

Note

To install a software update

1. Disconnect users from the server farm by stopping the World Wide Web Publishing service (w3svc) on all Web servers.

2. Download and install the appropriate Windows SharePoint Services 3.0 software update for all servers in your server farm.

Important

At the end of the Windows SharePoint Services 3.0 software update, do not run the SharePoint Products and Technologies Configuration Wizard.

3. Download and install the appropriate Office SharePoint Server 2007 software update for all servers in your server farm.

4. At the end of the software update installation, the SharePoint Products and Technologies Configuration Wizard starts.

Note

5. On the SharePoint Products and Technologies Configuration Wizard Welcome page, click Next.

6. In the dialog box that notifies you that some services might need to be restarted during configuration, click Yes.

7. On the Completing the SharePoint Products and Technologies Configuration Wizard page, click Next.

8. When the dialog box about installation in a server farm appears, do not click OK. Instead, leave each server with the following dialog box displayed:

Notes

10. On the server you selected in the previous step, click OK.

11. On the Configuration Successful page, click Finish.

13. Continue updating the remaining computers in the server farm, one at a time, by clicking OK in the dialog box.

Note

It is important that the SharePoint Products and Technologies Configuration Wizard perform the configuration procedures on only one computer at a time.

3. OPTIONAL only If you detached content dbs use the following command line to add them back to the farm

stsadm -o addcontentdb -url <http://backupservername:port> -databasename <ContentDBName>

If you detach and reattach a content database, be aware that the next time the content within that content database is crawled a full crawl will occur even if an incremental crawl is requested. Because a full crawl re-crawls all content that the crawler encounters, regardless of whether that content has been previously crawled, full crawls can take significantly more time to complete than incremental crawls. CPU will be maxed! BUT IF YOU INSTALL THE INFRASTURE UPDATE THEN A FULL INDEX WILL NOT HAPPEN. THE LOGIC HAS BEEN CHANGED TO WHERE THIS IS NO LONGER AN ISSUE.

4. Verify the server install by running the following command against the config database

SELECT * FROM Versions

1. Release 12.0.4518.1016

2. October public update 12.0.0.6036

3. Service Pack 1 12.0.0.6219

4. Post Service Pack 1 rollup 12.0.6300.5000

· OPTIONAL Force an upgrade by running one of the following commands if you encounter problems

1. psconfig –cmd upgrade –inplace b2b –wait –force

· Next Install Rollups 1 & 2 for both WSS and MOSS and reboot the server

IMPORTANT NOTICE FOR FUTURE INSTALLS:

From this point forward when you add a new server to the farm you should make sure you do the following:

1. Install MOSS with the SP1 Slipstream Version

2. Install The Latest Rollup which are presently WSSv3 Blackout rollup & MOSS Blackout rollup

3. Run PSConfig to attach to the farm.

Controlling PeoplePicker Content

gregmcb — Mon, 14 Apr 2008 23:14:00 GMT

You have two choices, you can use either of the following stsadm cmds: searchadcustomfilter or setsiteuseraccountdirectorypath.

setsiteuseraccountdirectorypath:

When people picker resolves the user, it will check whether the user exists in the site collection or not. If it exists, return the user. Otherwise, search the AD or membership provider. For the AD case, if there is “siteuseraccountdirectorypath”, it will only search under that directory path, otherwise, the whole AD is searched.

Suppose the site is empty. After the administrator use

-o setsiteuseraccountdirectorypath

Only users under that path could be added to the site collection and no one else could be added to the site collection. In such case, the user returned will always be under the “siteuseraccountdirectorypath”.

Suppose the site is not empty and there are already some users exists. After the administrator use

-o setsiteuseraccountdiretorypath

To add a new user, the new user must be under the directory path. The people picker will return existing users in the site collection and users under the directory path.

searchadcustomfilter:

This path allows you to control the users shown in the PeoplePicker control by LDAP, BUT a user can still type in a valid alias and click the check name button to be given access.

Virtualization Known Impacts

gregmcb — Thu, 27 Mar 2008 22:26:00 GMT

VMWare could impact Windows services that depend upon the Windows Time Service for maintaining time synchronization across systems. Apparently the proper way to keep the guest time synced is to use the VMWare TimerSync and not the default Windows Timer Sync. If the system was to become out of sync by more than 5 minutes let’s say then Kerberos could be a problem. This could impact user access or Timer jobs performing tasks. So if a customer has not configured this correctly then there could be problems. The solution however is an easy workaround described below.

* VMware Time Sync and Windows Time Service

http://kb.vmware.com/selfservice/viewContent.do?language=en_US&externalId=1318

“The most accurate way to keep guest operating system time synchronized with real time is to use the VMware Tools time synchronization function. You should not use the Windows Time service or other form of clock synchronization meant for physical machines to set the time in the guest operating system.”

The risk of having the USN Rollback effect.

If the virtual machine is switched off with “Don’t Save Changes”, or the host is turned off in an uncontrolled manner (power failure) …

This DC doesn’t save its latest changes.

But the replication USN numbers were incremented, and all other DCs have already learnt about this on their High Watermark vectors.

The DC has jumped backwards in time (lost the latest USN changes), and Replication Dampening will avoid to re-replicate these lost changes again. (All DCs think it has already got them)

Likely to happen (as “Don’t Save Changes” can be the default), difficult to detect, and more difficult to fix …

How to detect and recover from a USN rollback in Windows Server 2003 http://support.microsoft.com/?id=875495

Top 10 Things You Need to Consider When Implementing MOSS

gregmcb — Mon, 17 Mar 2008 23:41:00 GMT

Over the years I have seen a lot of SharePoint deployments from many customers going back to 2003. This has allowed me to develop a unique perspective and observation on how to deploy this product best. Here is what I have learned are the top ten things you need to do to make your deployment the success that will provide you with the best ROI and take full advantage of the product. MOSS and the surrounding products such as Office, Exchange, PerformancePoint, Office Communication Server, Project Server, Excel Server, Forms Server, SQL Analysis and Reporting Server have the potential to provide the same productivity increases seen when email was deployed universally in the 90’s. Make full use of these technologies AND make Your Company succeed. That being said, it is important to remember that MOSS is about people and collaboration not products.

1. “MOSS is not your father’s corporate web!” When you deploy MOSS you need to think future. You need to be thinking about not only deploying a portal but also a collaborative integration solution that will dramatically increase your companies productivity potential. MOSS is not just about document management and lists. Collaboration/portals has to be about people. If you don’t plan on training your end users about new ways to think and new ways to share information, then the value of the tools will be minimized.

2. Train your Employees on how they can use MOSS functionality and provide them with the insight they need to see how they can increase their productivity and use all the features of the product. Don’t allow this opportunity to become a means of using this new technology the same way you would in the past. Think beyond document libraries and lists. Think communication, business intelligence such as Dashboards, forms , workflow and collaboration to the fullest potential. Develop Power Users that allows for your portal to be driven by the users. They know what they need and if you provide the opportunity and knowledge to allow their ingenuity to drive the process it will allow for greater success and allow your IT staff to focus on IT and not driving the collaboration. Everyone is equal when collaborating online and senior management of a company needs to buy in to the freedom and flexibility that will come from allowing people to work like this. Consider spending 25% of your budget on training the power users and 75 percent on training all users. The Power Users should be those stakeholders in the business and should be instructed upon the possibilities of the technology so they can chose which to implement effectively for their teams. This is the key to success.

3. Plan your overall design, Taxonomy, site structure and navigation, Search, Document Management, Records Management, Security, Server Farms, logical design with great care, knowing that your deployment will grow far beyond what your initial intentions were. Think about how you deployed AD and Organizational Units so that they would sustain your deployment lifetime and future upgrades. Use the Planning Worksheets provided. Adopt a SharePoint Governance model. Here is a sample Governance plan. By implementing a distributed governance model which involves creating power users in the various lines of business for day to day governance tasks like site creation and security you not only mold champions to drive awareness of the capabilities of the platform, you offload a significant portion of the non-strategic IT tasks to end users. This frees up your IT folks to do the activities that really interest them. Those power users now represent a great foundation for a portal-steering committee that can determine standards, branding, navigation, etc.

4. Plan for performance by knowing sizing constraints, SQL Performance and how each functionality will influence your environments stability. Design and deploy an effective policy with regards to the site creation and maintenance, recycle bin, retention, and archival. Do not allow your data to become stale and out of date.

5. Go to KB Alertz and register for MOSS and WSS KBs as well as view and register an RSS feed to SharePoint Team Blog before you plan and deploy.

6. Actually, take the time to build a test, QA and production environment that are exactly the same as far as general design. Then use content management to deploy between environments.

7. Define your growth strategy before you even install the product.

8. Have a documented and tested DR plan with SLAs with each top level organization that will be involved in your project before even installing. This should contribute greatly to your design especially with regards to Content Databases and Site Collections. With SP1 you can move site collections between content databases and webs.

9. Most environment can accommodate 80-90 percent of their needs without writing a single line of code. When defining how you will meet business needs ask if a built in component will suffice and evaluate the delta between the exact business needs and the cost of designing, developing, deploying and maintaining custom code. While it’s fun to write WebParts, it is not fun to manage and debug them when something happens. That being said, if you do use the built in WP make sure they are consistent with Microsoft best practices so they will upgrade to the next version or not break on a Service Pack.

10. Think, plan and put in place your Operations Guide before you deploy. What will you monitor in MOSS and what tools will you use to do this? What are the daily, weekly, monthly and quarterly tasks that you need to accomplish to keep your master piece running at optimum levels as defined by your SLAs.

PS: Start using those mail enabled lists types to replace your public folders now. The best choice is the email enabled discussion lists as well as email enabled calendars.

User Profiles Out of Sync with AD

gregmcb — Sat, 15 Mar 2008 08:09:00 GMT

The most common cause of this problem is not running stsadm -o preparetomove prior to moving or upgrading the database.

In case if you don't know there is a hidden list named "USER INFORMATION LIST" which displays all the users of the site (collection) {MOSS / Sharepoint 2007}. And there is a page "Simple.aspx" which show this list, the url for the page would be http://URSITENAME/_catalogs/users/simple.aspx normally no one has access to this list (except for the admin) and as the list is hidden it is not viewable in the "view all site contents" too. You will have to open you site in SPD and then click on _catalogs + and then you should see the "USER INFORMATION LIST" under it.

WSS sites (including those within the sites collection of a portal) have something called a User Information list that keeps a limited amount of info about it’s users (login name, email, display name). When a user is granted access to a site, their info is put in a User Information list. When a user creates/edits an item in a list SharePoint creates a link to the user’s record in the User Information list so when the user is shown in “Created By” and “Modified By” indicators, the Display Name from the User Information list is shown

SharePoint Portal Server has something called a Profile Database. When you first visit a portal, SPS checks to see if it has a profile for you and if not, it creates one on the fly with your info from AD. You can also configure SPS to automatically populate the Profile Database with a full import, an incremental update, and optionally set it to pull objects with a custom LDAP query.

Here is the summary. When a user accesses a site collection the first time, they are added to the profile. Howevere, a property named tp_IsActive in the USER INFORMATION LIST is set to false until the user interacts with the site by adding a document or clicking a link. When this property is false the user information will not be synced by the jobs WSSProfileSynch and WSSSweepSynch . Thus the inconsistency begins to develop.

A lot of use information can be obtained from http://msdn2.microsoft.com/en-us/magazine/cc163287.aspx or http://www.21apps.com/2008/02/user-profiles-why-do-my-changes-not.html.

Planning a SharePoint Implementation From a DBA Perspective

gregmcb — Fri, 14 Mar 2008 19:50:00 GMT

Content database IO utilization depends upon how the system is used. Generally, read access is more of a prominent need, but if your plans include lots of updates and additions of documents or list items then write IO will be important to your decision. Furthermore, different components of SharePoint will have different needs such as the tempdb and Search db. So to really answer the question you need to ask how will my company be using MOSS. The search and temp databases are high IO resources as you stated and explained in the in greater detail in the documentation below.

I would recommend the following resources

Database Administration Tasks

Estimating SharePoint Performance

Estimate performance and capacity requirements (Office SharePoint Server)

The most important DBA doc which just became available in Dec though is White paper: Performance recommendations for storage planning and monitoring. I have attached that file as well as the link. So what this really means to you is that to effectively plan your DB environment for performance and scalability, you first need the business to provide you with at a minimum what they plan on doing.

· Planning and architecture for Office SharePoint Server 2007 (http://go.microsoft.com/fwlink/?LinkID=105576&clcid=0x409)

· Planning and architecture for Windows SharePoint Services 3.0 technology (http://go.microsoft.com/fwlink/?LinkID=105577&clcid=0x409)

All this will lead you to Windows SharePoint Services collaboration environments and eventually this Physical Database Storage Design or Estimate performance and capacity requirements for Windows SharePoint Services collaboration environments (Office SharePoint Server) and eventually to Physical Database Storage Design depending upon the farm being MOSS or WSS.

This will then eventually lead you to Predeployment I/O Best Practices.

So what this all boils down to is the business has to tell you as the DBA what they are going to do and how they plan to use MOSS. To answer your other questions about planning for the tempdb see below from the attached White paper: Performance recommendations for storage planning and monitoring.

Ideally, the tempdb, content databases and SQL Server transaction logs should all be on separate physical hard disks.

· When prioritizing data among faster disks, use the following ranking:

1. tempdb data and transaction logs

2. Database transaction log files.

3. Search database.

4. Database data files.

Note: In a heavily read-oriented portal, prioritize data over logs.

· Testing and customer data have shown that Office SharePoint Server farm performance can be significantly impeded by insufficient disk I/O for the tempdb. To avoid this issue, allocate dedicated disks for the tempdb file. If a high workload is projected or monitored (Avg. Disk Sec/Read or Avg. Disk Sec/Write is greater than 20 ms), you may need to ease the bottleneck by either separating the files across disks, or replacing your disks with faster disks.

· For best performance, place the tempdb file on a RAID 10 array. The number of data files should be equal to the number of core CPUs, and of equal size.

· Separate database data and transaction log file across different disks. If files must share disks because the files are too small to warrant a whole disk or stripe or you have a shortage of disk space, put files on the same disk that have different usage patterns, so that simultaneous access requests will be unlikely.

· Consult your storage hardware vendor for how to configure all logs and the Search databases for write optimization for your particular storage solution.

· Pre-size your data and log files.

· Do not rely on AUTOGROW, instead manage the growth of data and log files manually. You may leave AUTOGROW ON for safety reasons, but you should proactively manage the growth of the data files.

· Allocate dedicated spindles for the Search database.

Depending upon the specific hardware and configuration, you are going to see different results. The WP and articles we publish usually display specific hardware examples. IO of course is going to be the your biggest focus, so I would strongly encourage you to engage your hardware provider as in many cases they will have specific recommendations for each component.

Creating Effective Enterprise Portals by Using SharePoint Server 2007

MSIT - Deploying and Supporting Enterprise Search

Microsoft Office SharePoint Server 2007 on HP ProLiant Servers - Performance Summary

Microsoft Office SharePoint Server 2007 Sizing and Configuration Tool

How Dell Does IT - SharePoint Portal Server 2007

Dell MOSS Portal

Dell MOSS Sizing Guide

Domain Local Users Group Can't Access SharePoint

gregmcb — Sat, 08 Mar 2008 10:48:00 GMT

932378 Members of the domain local group cannot view a Microsoft Office SharePoint Server 2007 Web site
http://support.microsoft.com/default.aspx?scid=kb;EN-US;932378

Kerberos Fails When Using CNAME

gregmcb — Sat, 08 Mar 2008 10:44:00 GMT

Users see the following error which can be fixed by a hotfix for IE or by using HOST records rather than CNAME in DNS

Error message in Internet Explorer when you try to access a Web site that requires Kerberos authentication on a Windows XP-based computer: "HTTP Error 401 - Unauthorized: Access is denied due to invalid credentials"

http://support.microsoft.com/kb/911149

This problem may occur if the Web site uses a CNAME resource record in the Domain Name System (DNS) to contact the server that initiates Kerberos authentication. (This server also issues the Kerberos ticket.) When you use Internet Explorer to access the Web site, Internet Explorer uses the host name of the server instead of the CNAME resource record to contact the server. The authentication process does not work because only the Service Principal Name (SPN) for the CNAME resource record is registered on the account that is used for the authentication.

Howto use Powershell on Sharepoint from wbaer

gregmcb — Wed, 05 Mar 2008 22:40:00 GMT

Reference the Microsoft.SharePoint.Administration Namespace and call the SPWebApplicationBuilder class to create a new SPWebApplication object.

[system.reflection.assembly]::LoadWithPartialName("Microsoft.Sharepoint")
$webAppBuilder=new-object Microsoft.SharePoint.Administration.SPWebApplicationBuilder( [Microsoft.SharePoint.Administration.SPFarm]::Local)
$webAppBuilder.Port=80
$webApp=$webAppBuilder.Create()
$webApp.Provision()
$webApp.Sites.Add("/","contoso\wbaer",wbaer@contoso.com)

How to Move a SharePoint Database To a New Server

gregmcb — Wed, 05 Mar 2008 22:11:00 GMT

1) Find the content Database

These are listed under Central Admin->Application Management->Site Collection List

2) Backup the content database

You could alternatively detach it, and copy it. Just doing a backup in SQL Server 2005 Management studio is easier.

3) Restore content database to new server

Copy the BAK file to new server. Create an empty DB in Management Studio, restore from backup, you may need to change an option in the "options" tab of the restore dialog to get it to work. (Overwrite db).

4) Create Web App on new Server

Central Admin->Application Management->Create or extend Web App->Create New Web App.

5) Associate restored DB with new Web App

Central Admin->Application Management->

SharePoint Web Application Management->Content Databases->

Remove Content Database from your new web app.

Now use STSADM to add restored DB to this web app

c:\program files\common files\microsoft shared\web server extentions\12\bin on new server is where you can find the STSADM.

run this command from there.

stsadm -o addcontentdb -url http://yourwebapp:port -databasename yourcontentdb -databaseserver yoursqlserver

6) Run ISSRESET from command prompt.

Caveats:

Make sure your running the same service pack level on both source and destination sharepoint if possible.
Make sure you install all webparts, solutions, features, etc on new server before you restore the content database, so that it can find all the features it's looking for.
Make sure you copy any files that may be living in the file system that you need, sometimes people have css/javascript here. (This refers to files in the hive)
Leave your old site and contentDB intact until you get the backup running on the new server, this way your ok if there is a problem.
DON'T try this with your config database! It won't work!

Think About Your SharePoint Taxonomy Before Deployment

gregmcb — Wed, 05 Mar 2008 15:31:00 GMT

The quickest way to fail from a user perspective is to not train your users how to build a good site taxonomy in collaboration with IT. Remember that you want to build your taxonomy wide not deep. Use the 6 P's and you will have a true collaboration site that will last. Failure to do this up front will cause you to have to spend a lot of time exporting & importing your sites to get it right.

Planning worksheets for Office SharePoint Server 2007
URL: http://technet2.microsoft.com/Office/en-us/library/49cf7edd-14ee-445b-8ab0-1d1339f2435f1033.mspx

Plan Web site structure and publishing (Office SharePoint Server)
URL: http://technet2.microsoft.com/Office/en-us/library/6a4b0ec4-8802-40f9-87ac-6a6691b544a41033.mspx

Plan Web site structure and publishing (Windows SharePoint Services)
URL: http://technet2.microsoft.com/windowsserver/WSS/en/library/aa456629-8de8-4328-873b-2e2db96714011033.mspx

Plan site navigation (Windows SharePoint Services)
URL: http://technet2.microsoft.com/windowsserver/WSS/en/library/a7e52934-30ff-455d-bfcd-52a3e0ff170a1033.mspx

Logical architecture components
URL: http://technet2.microsoft.com/Office/en-us/library/aaed3a01-f4dc-4353-abda-0beced2080b61033.mspx

Design Records Center architecture
URL: http://technet2.microsoft.com/Office/en-us/library/f82c58ba-6d25-406d-835e-aa00acd5b76d1033.mspx

Once you have done this you need to plan site navigation

http://technet2.microsoft.com/windowsserver/WSS/en/library/a7e52934-30ff-455d-bfcd-52a3e0ff170a1033.mspx

Determine individual site content needs and structure
URL: http://technet2.microsoft.com/windowsserver/WSS/en/library/1c1ff8b1-a114-4feb-8373-e2e829f9ad871033.mspx

Plan document management
URL: http://technet2.microsoft.com/Office/en-us/library/cb57d4f7-8353-420b-8a10-d05c9bdf1b561033.mspx

UNDERSTANDING SITE COLLECTIONS AND SITES

By default, all content for a Web application is stored in one content database. You can separate content into multiple content databases at the site collection level. A content database can include one or more site collections. A single site collection cannot span multiple databases. Backing up and restoring sites takes place at the content database level.

Site collections will allow the IT department freedom to maintain just application itself without the worry of security or content hierarchy maintenance. The following is a list of what an individual site collection offers.

For the Users:

Dedicated Recycle bins
Dedicated usage Reports
Distributed administration (site collection administrators)
Dedicated search scopes, keywords, and best-bets
Custom feature deployments
Dedicated language translation maintenance
Dedicated galleries for web parts, master pages, content types, site columns, site templates, and list templates
Dedicated shared libraries, such as site collection images and site collection styles
Dedicated real estate (Self Containment)

For the IT Administrators:

Site quota templates
Distributed administration
Site locking
Database maintenance options
Backup / Restore abilities
Content Deployments
InfoPath forms services global template targeting

Top Level Sites - Usually created by department, project, etc...

Resources that will help you plan your taxonomy

Decide whether to use individual site collections or subsites within one site collection

You must decide whether to create your sites as top-level Web sites in separate site collections, or as subsites within the same site collection. This decision is based on how much the sites have in common with each other, whether you want to be able to manage them individually, and whether you want them to share elements, such as navigation or search.

Within a site collection, all sites can use the same:

•	Navigation bars (top link bar and breadcrumb navigation)
•	Content types
•	Workflows
•	Security groups
•	Lookup fields across lists
•	Search scope
•	Feature set

Choose top-level Web sites in separate site collections when you:

•

Need separate security for different sites.

	Note:
	Although you can have unique permissions for a subsite, at times you might want to be sure that there are no users and permissions in common between two sites. In those cases, you should use separate site collections.

•

Might need to move the site collection to a different database in the future.

•

Want to be able to back up or restore that site only.

•

Want to be able to scope a workflow to that site only.

•

Want to have a separate search scope for that site only.

•

Want to use quotas to separately manage the amount of space that each site takes up.

•

Want to decentralize your administration and have site collection administrators perform tasks, such as approving requests for access or confirming site use.

Choose subsites within the same site collection when you:

•	Want to share navigation between sites.
•	Want to have subsites inherit permissions from parent sites.
•	Want to share lists between sites.
•	Want to share design elements (such as themes or styles) between sites.

Zones represent different logical paths (URLs) of gaining access to the same Web application. Within each Web application, you can create up to five zones using one of the available zone names: Default, Intranet, Internet, Custom, or Extranet. Each zone is represented by a different Web site in IIS.

The Default zone is the zone that is first created when a Web application is created.

Capacity

You can create up to five zones within a Web application. A farm that hosts more than one Web application can support user requests from more than five different network zones. Typically, zones are coordinated across Web applications so that zones of the same name are configured for the same users.

Sharing and isolation

Zones provide a method of partitioning users by:

•	Authentication type Each zone can be configured to use a different authentication provider, enabling you to share the same content across partner companies.
•	Network zone Each zone can be configured to accommodate users entering from a different network zone, such as an extranet or the Internet.
•	Policy permissions You can explicitly allow or deny read or write access to content per zone based on a user account or a group account.

Service Account Descriptions & Permissions

gregmcb — Wed, 05 Mar 2008 14:44:00 GMT

Search service account

- It is a db_owner in ALL SSP databases.

- It is a db_owner in ALL Search databases.

- It has READ ONLY access to all the content in ALL web applications via a policy.

- It has read/write access to the propagation share on Query servers.

- It has read/write access to the Search registry hive.

- It has read/write access to the Search index location.

SSP administration site application pool identity

- This account is determined by the web application that you select when you create the SSP.

- It has read/write access to the SSP database and the Search database.

- This account has full control over the Search service via its COM interfaces.

- It has read/write access to the Search registry hive.

Global web service account

- This is the GLOBAL application pool account of the Office Server Web Services, i.e. an application pool that does not belong to any SSP.

- It is always set to NetworkService.

SSP (Application) web service

- The application pool account of an SSP web service (the credentials entered in the SSP creation/details page).

- This account has read/write access to the SSP database and to the Search database of an SSP.

- This account has full control over the Search service via its COM interfaces.

- It has read/write access to the Search registry hive.

Default content access account

- The default account used within the SSP to crawl content.

- If a specific account is not specified, the search service account is used.

Content access account

- A specific account that is configured to particular content.

- This account is optional and is specified when you create a new crawl rule. For example, content that is external to Office SharePoint Server 2007 for Search (such as a file share) might require a different access account.

SSP and OWSTimer Service - Job Descriptions Used for Synchronization

gregmcb — Wed, 05 Mar 2008 14:41:00 GMT

The configuration of a SharePoint farm could be broken into two parts:

- Initial deployment, e.g. adding a server to the farm and provisioning an indexer.

- Synchronization, i.e. continuous automatic activities to change and self heal the configuration of the Search feature accross the SharePoint farm based on farm level configuration changes and SSP configuration changes. Some of the changes are a replay of an administrative action in the Central Administration site.

The synchronization mechanism enhances the robustness of the system because it allows the farm to successfully manage servers with a certain amount of redundancy (e.g. a failure on one server will usually not interfere with the synchronization on another server).

SharePoint infrastructure

The WSS timer Service infrastructure executes jobs across the farm in the context of the farm administrator. Synchronization in MOSS Search and in WSS Search is coded into a set of jobs that periodically adapt the configuration of the Search service according to the configuration of other system components.

Replaying administrative actions

The synchronization process can also replay highly privileged administrative actions such as changing the Search NT service account. If the farm administrator is not a local administrator on the Search server, then the timer service asks the SharePoint Administration service ("SPAdmin") to perform this task on its behalf (SPAdmin runs as LocalSystem) via a .NET remoting channel.

WSS Search

When an administrator deploys (selects "Start" in the Central Administration Services on Server page server.aspx) the WSS Search service on a server, a "SharePoint Services Search Refresh" (more info below) job is created and immediately deployed on ALL the servers in the farm. This job runs every 5 minutes (by default). It automatically changes and self heals the configuration of the WSS Search service.

WSS: SPSearchServiceInstance

The "SharePoint Services Search Refresh" job calls into the Synchronize method of this instance on Online Search servers. For example, when a web application is created or a content database is added to an existing web application, it adds a start address to the Search content source and instructs the Search engine to index the new content on a regular basis. When a content database is removed, the start address is automatically removed.

WSS: SPSearchDataAccessServiceInstance

The "SharePoint Services Search Refresh" job calls into the Provision method of this instance on ALL servers. It configures the local server to allow communications with the Search SQL database. Specifically, it seeds the SQL connection information in the registry under the Search Applications hive (HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Shared Tools\Web Server Extensions\12.0\Search\Applications). This info is then utilized by the Search engine and by the Query provider.

MOSS

The synchronization work in MOSS Search is managed by a generic MOSS Shared Services infrastructure via Application Server and Shared Services jobs that execute every minute:

- The system uses the ISharedComponent interface to manage the lifecycle of Shared Applications like SearchSharedApplication. If the status is Provisioning then the Install() method is called and the status changes to Online. If the status is Online then the Synchronize() method is called.

- The system uses the ISharedServiceInstance interface to manage the lifecycle of Shared service instances like SearchServiceInstance and SearchDataAccessServiceInstance. If the status is Online then the Synchronize() method is called.

SearchServiceInstance

Adapts the local server to farm level configuration changes and to SSPs configuration changes. For example:

- Creates an indexing gatherer application for an SSP that has been assigned to an indexer.

- Removes a gatherer application from the old indexer when the SSP is not associated anymore with it.

- Creates a query gatherer application for an SSP on all Query servers.

- Removes a gatherer application from all Query servers when the SSP has been unprovisioned or left unassigned to an indexer.

- Creates a start address in the default content source.

- Deletes a start address from the default content source (only when the content source is idle, i.e. the gatherer is not indexing its start addresses).

SearchDataAccessServiceInstance

- Configures the local server to allow communications with the SSP and Search SQL databases. On a child farm, it allows the query provider to communicate with the databases of the central farm's SSP. Specifically, it seeds the SQL connection information in the registry under the Search Applications hive (HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Office Server\12.0\Search\Applications). This info is then utilized by the Search engine and by the Query provider.

- Remove obsolete gatherer applications, i.e. applications that do not belong to a live SSP.

- Persists the IP address of the local server to help the functionality of the Dedicated WFE indexing feature.

SearchSharedApplication

Persists application settings and instructions for the synchronizing code:

installGathererApplication - Create a new gatherer application on the indexer or import an existing one from the SSP database.

indexer - index server.

indexLocation - index location on index server.

newIndexLocation - An optional setting when the index is moved on the same indexer.

needsReset - Reset the index.

needsResume - Resume indexing (after backup/restore).

needsRestore - SSP Sql only restore.

Synchronization jobs

Application Server Timer Job

- Type: Microsoft.Office.Server.Administration.ApplicationServerJob

- Schedule: Once a minute.

- This job does not require the SharePoint Administration ("SPAdmin") service.

- It is designed to manage Shared service instances that do not perform highly privileged operations (e.g. configuring an NT service).

- The Search service instance (SearchServiceInstance) is managed by this job on Basic deployments.

Application Server Administration Service Timer Job

- Type: Microsoft.Office.Server.Administration.ApplicationServerAdministrationServiceJob

- Schedule: Once a minute.

- This job requires that the SharePoint Administration ("SPAdmin") service is running. Otherwise, it does not manage Shared service instances and an error is recorded in the event log.

- It is designed to manage Shared service instances that may perform highly privileged operations (e.g. configuring an NT service).

- The Search service instance (SearchServiceInstance) is managed by this job on non-Basic deployments.

Shared Services Provider Synchronizing Job

- Type:Microsoft.Office.Server.Administration.SharedResourceProviderJob

- Schedule: Once a minute.

- Automatically configures an SSP.

- If the status of the SSP is Provisioning, then call the Provision method of the SSP and change its status to Online.

- If the status is Online, then call the Synchronize method.

Shared Services Timer Job

- Type: Microsoft.Office.Server.Administration.JobScheduler

- Schedule: Once a minute.

- Schedules SSP jobs.

- A shared application can register a job to run with this scheduler. The job will run in the security context of the SSP web service account.

Indexing Schedule Manager on <index server name>

- Type: Microsoft.Office.Server.Search.Administration.IndexingScheduleJobDefinition

- Schedule: Once a minute.

- A MOSS Search SharePoint timer job that kicks off scheduled crawls.

- It utilizes the mstask.dll Windows Task Scheduler client DLL to find the next execution time.

Child farm content source discovery

- Type: Microsoft.Office.Server.Search.Administration.ChildFarmContentSourcesJobDefinition

- Schedule: Once every 15 minutes.

- A MOSS Search SharePoint timer job that runs on child farms, discovers web applications and registers them in the SSP database. Later on, the synchronizer on the central farm configures the default content source with a start address per registered child web application.

SharePoint Services Search Refresh

- Type: Microsoft.SharePoint.Search.Administration.SPSearchJobDefinition

- Schedule: Once every 5 minutes.

- It automatically changes and self heals the configuration of the WSS Search service.

Config Refresh

- Type: Microsoft.SharePoint.Administration.SPTimerJobConfig

- Schedule: Once a minute.

- A SharePoint job that runs on all servers. It refreshes the server cache, i.e. updates it with changes from the SharePoint Configuration database.

- The server cache resides here: "%ALLUSERSPROFILE%\Application Data\Microsoft\SharePoint\Config"

Why Does My SSP and Index Service Keep Becoming Unstable

gregmcb — Tue, 04 Mar 2008 16:42:00 GMT

In short, a lot of what you do in the SSP admin console is executed through OWSTimer jobs. There is a known issue on 64 bit systems that can result in an exception being thown and the service will shutdown. From there what can happen is that whatever the job or jobs that are being executed may not finish and introduce an inconsistent state. For example, changing the password for service accounts causes access denied or 403 errors.

If you apply the following hotfix it should prevent future problems. You should definitly apply this Windows fix before applying Service Pack 1 or making any changes to the SSP. If you do not do this you place your SSP at risk.

http://support.microsoft.com/kb/946517 - Fix to correct 6482 Timer job Heap Corruption Issue on 64 bit system

Common Jobs Completed by the OWSTIMER service that contribute to SSP Services :

Application Server Timer Job - Manages the Search service instance (SearchServiceInstance) on Basic deployments.

Application Server Administration Service Timer Job - Manages the Search service instance (SearchServiceInstance) on non-Basic deployments.

Shared Services Provider Synchronizing Job - Provisions and synchronizes SSPs.

Shared Services Timer Job - Executes shared application jobs in the security context of the SSP web service account. Might result in User Profile processing issues.

Indexing Schedule Manager on <index server name> - A MOSS Search SharePoint timer job that kicks off scheduled crawls.

Child farm content source discovery - A MOSS Search SharePoint timer job that runs on child farms, discovers web applications and registers them in the SSP database. Later on, the synchronizer on the central farm configures the default content source with a start address per registered child web application.

SharePoint Services Search Refresh - Automatically changes and self heals the configuration of the WSS Search service.

Config Refresh - A SharePoint job that runs on all servers. It refreshes the server cache, i.e. updates it with changes from the SharePoint Configuration

You might also see this problem.

Event ID's 7076, 6398 and 6432
In SharePoint Server 2007, tasks that are scheduled do not run.
When you try to manage IIS 6.0 by using Server Manager, you receive a blank page, or you receive the following error message: "the path specified cannot be used at this time".