Business Data Connectivity Services (BCS) in Office 365 – SharePoint Online

The following video walks you through using SharePoint Online’s Business Connectivity Services (BCS) to connect your line of business (LoB) systems via a Windows Communication Foundation (WCF) service. This allows data from the LoB system on-premises to be viewed and edited within your SharePoint Online environment.   (Please visit the site to view this video)   For more info about BCS, see Notes on Business Connectivity Services (BCS) in SharePoint Online .   Thanks to Manas and Alex from Microsoft Support for putting together this video.   Let us know what you think!   The SharePoint Online Community Team  

Business Data Connectivity Services (BCS) in Office 365 – SharePoint Online

The following video walks you through using SharePoint Online’s Business Connectivity Services (BCS) to connect your line of business (LoB) systems via a Windows Communication Foundation (WCF) service. This allows data from the LoB system on-premises to be viewed and edited within your SharePoint Online environment.   (Please visit the site to view this video)   For more info about BCS, see Notes on Business Connectivity Services (BCS) in SharePoint Online .   Thanks to Manas and Alex from Microsoft Support for putting together this video.   Let us know what you think!   The SharePoint Online Community Team  

New Policy for Inactive Migration Batches

Audience: Exchange Online/Office 365 administrators Author: Ram Poornalingam, Senior Program Manager, Exchange   We’re implementing a new policy for migration batches in the cloud-based service. Active migration batches with a status of “Synced” that have no administrator-initiated activity for the last 90 days will be stopped, and then deleted 30 days later if no further activity is taken by the administrator.   What constitutes activity by the administrator?   Stopping and restarting a migration batch Editing a migration batch by using the Exchange Control Panel or the Set-MigrationBatch cmdlet in Windows PowerShell.   How do we determine if a migration batch is inactive? There is a “Last modified time” property for migration batches, and if the value for this property is more than 90 days, the batch is stopped. If the batch remains inactive for another 30 days (in other words, when the value of this property reaches 120 days), the migration batch is deleted.   This new policy applies to all migration types: Cutover Exchange migrations, staged Exchange migrations, and IMAP migrations.   Why are we doing this? Inactive migration batches consume resources in the Microsoft datacenter. By stopping and then deleting these batches, we can free up resources and help improve the migration experience for customers who are actively migrating mailboxes to Office 365.   For more information about migration, see the following topics on   Manage Migration Batches in Exchange Online Cutover Exchange migration Staged Exchange migration IMAP migration

Project Online, Project Server 2013, the new Project TechEd Slides

Getting access to the new Project slides is in hot demand these days and voila! (see expect a lot more in the future…)   Microsoft Project Online Overview   Project Server 2013, a Landmark Release!

Project Server 2007 Mainstream Support Ends on October 9, 2012 UPGRADE TODAY!

Dear community, important reminder that the Project Server 2007 mainstream support ends in a few days on October 9th 2012 to be precise! See for yourself on the official Microsoft Product Lifecycle site: We have lots of content to help you upgrade to Project Server 2010 on TechNet: Upgrade and migration for Project Server 2010 as well as a rich partner ecosystem to help you move forward: Want to upgrade to Project Server 2013 (note it’s still in Preview and not final), see this: Plan for upgrade to Project Server 2013 Start your upgrade today!

Outlook Web App and Mobile Phone Videos

Did you know that we have a collection of how-to videos on Outlook Web App, desktop clients, and mobile phones? Did you know you can access these videos all in one convenient place?   For a list of all of the Outlook Web App, desktop client, and mobile phone videos, see Outlook Web App Help Videos .   The help videos available include: Creating Inbox Rules Sending Automatic Replies Finding items in your mailbox Viewing messages by conversation Sending and receiving email from other email systems Redirecting incoming messages to another email account Setting up a POP or IMAP email program to send and receive messages Setting up email in Outlook 2007 Setting up email in Outlook 2010 Learning to use contacts, chat, and groups Learning to use the calendar Setting up your phone for Exchange ActiveSync Setting up your phone for POP and IMAP Do you have an idea for a video you’d like to see? Let us know!

Office 365 Usability Research Wants You!

Audience: Office 365 for professionals and small businesses Office 365 for enterprises   Written by Brendan Reeves, User Researcher II at Microsoft Our user-centered engineering process needs your input to make sure Office 365 matches your needs. The Office 365 User Experience team would like to learn more about how your organization uses Office 365. We are seeking feedback and information on how Office 365 helps your company function. This is your opportunity to share feedback with the product team and help us improve and evolve the Office 365 service. Please go to to submit your name, phone number, email address, and company name. The information you provide will only be used to recruit participants for user research projects. Your level of participation in research projects is completely up to you. There will be projects that only ask for a very low-level of engagement (e.g. surveys) and those that require high level engagement (e.g. on-site visits). As a thank you for your participation, we will provide a gratuity ranging from software to gift cards depending on study type. Study types include: Complete a 5-15 minute online survey Attend online webinar Usability lab studies On-site visit to your organization by 3-4 members of the Office 365 team for approximately 3 hours We appreciate your time and understand that it is valuable. If you are unable to participate in any research projects in the near future, you are still encouraged to sign up. User experience research is an ongoing process throughout the product development cycle and you will have the opportunity to participate in lighter weight research projects like surveys, webinars, or conference calls down the road. You can participate in as many or as few research studies as you like and remove yourself from the list at any time. Sincerely, Brendan Reeves If you would like to be removed from the Microsoft User Research program email with “Remove” in the subject line.  

Office 365 Expert Discussion Series #4: Exchange Online Q&A w/ Support Team

Updated 9/27: Thank you for attending this Office 365 Expert Discussions Series. If you missed it, don’t worry! You can view the  Lync Recording here  and below.  Microsoft Support will be presenting a series of webcasts called the  Office 365 Expert Discussion Series ,  to showcase our tools and content that can be used to help make your Office 365 experience better and help solve common issues more quickly.  (Please visit the site to view this video) Team Members from Exchange Online Escalation Team will be on hand to answer a few selected technical questions followed by open Q&A. Bring any technical questions you may have regarding Exchange Online such as tips for resolving directory sync issues, managing users, improving Outlook connectivity. Join us  on Thursday, September 27th at 8:00 AM Pacific Time  for this exciting webcast. The duration of the meeting is planned for 1 hour. After the meeting concludes a recording of the session will be published to this blog post and the  Office 365 YouTube channel . Microsoft Support will be monitoring the Office 365 Expert Discussion  Forum  and  Wiki  as well as the Community closely for this topic for a week after the event to answer any questions about the live session. Be sure to view our past Expert Discussions here .  To sign up for this webcast please  download and save the calendar invite below  where you will find the Lync invite and all other Office 365 Expert Discussion information.  We are excited to see you there!  —————————————————————————————————————————————————————————— Presenter :  Jeremy Kelly , Principle Escalation Engineer from Microsoft Commercial Technical Support and other Microsoft Support Team members. Date/Time:  Thursday September 27th , at 8:00 AM Pacific Time . (1 Hour presentation) Online Meeting Information: ……………………………………………………………………………………………………………………….. à Join Lync meeting            Join by phone +14257063500 (USA – Redmond Campus)                     English (United States) +18883203585 (USA – Redmond Campus)                     English (United States)   Find a local number   Conference ID: 35047018     Forgot your dial-in PIN? | Help      [! ……………………………………………………………………………………………………………………….. Audience: Office 365 MVP and  Grid Members  and open to anyone who is interested in Office 365 Expert Discussions. 

When your Project Server queue slows down

There can be a whole load of different reasons why a queue in Project Server might not be processing as fast as you’d like – and after listing a few of them I am going in to detail on one condition we have seen with a few different customers now, under some different scenarios.  All of these are equally as applicable to Project Server 2007 as to Project Server 2010 queue processing. 1. Poor database maintenance – see for Project Server 2010 and for the 2007 version 2. Large Shadow Tables – see 3. Incorrectly configured queue settings – too many threads for the hardware you have – the 2007 reference is still one of the best descriptions of the Project Server queue –   4. Just busy – the queue is there to spread the work out – sometimes this might take a while (long duration ‘admin’ type plans publishing…) 5. Poor execution plans for the stored procedures that control the queue processing – the main topic of this blog! Getting in to more detail on point 5 – we have seen a few scenarios where the queue is processing really slowly – although SQL Server isn’t very busy and there appears to be no good reason that the jobs aren’t going through faster.  One customer’s observation (thanks Eric!) was that the jobs appeared to also complete in groups – that related to the number of threads.  For example if you had 8 threads running your Project queue (and of course plenty of hardware to support this quite high setting) then you might see 8 jobs running and some of these might have been expected to finish quickly – but they all seem to be waiting for the longest running job – then they all finish at once.  What appeared to be happening is that certain stored procedures that govern the queue jobs that ordinarily take just a few milliseconds might slow down and take half a second – or 10 seconds – or in the extreme 30 seconds to run.  And it was these stored procedures that decided which jobs get processed next.  A good way of summarizing this was that Project Server appeared to be taking more time deciding which queue job to run and not spending any time actually running anything!  Initially a reboot was the step that got things moving – but we wanted to be a bit smarter than the ‘turn it off and turn it on again’ approach. I’ll talk a little more about the scenarios where we have seen this – some of the symptoms – some of the remedies and then some of the quick fixes that can both get things moving again and which can also confirm the problem – as well as introducing the topic of execution plans and how SQL Server decides how to get answers from the database. An execution plan will detail how SQL Server goes about finding the data it needs – and will detail if specific indexes are used – or if perhaps just a table read is going to be quicker.  The plan will depend on the indexes available, the statistics available to understand the make-up of the indexes and the accuracy of those statistics.  Some of this comes back to point 1 in the list above – database maintenance – and some may depend on certain server settings – and even patches and versions of SQL Server in use.  Most of the slow downs we have seen relate to activities that will create a large number of queue jobs in a very short time.  Active Directory Synchronization, reporting database rebuild, mass deletion of projects or very heavy timesheet usage (or even automation) are the ones that spring to mind.  My theory is that the queue tables that were running along quite nicely while relatively empty suddenly find they have thousands of rows and the execution plan that is cached is no longer optimal.  SQL Server should quickly adjust for this – and we have certainly seen that Service Pack 1 of SQL Server 2008 R2 appears less susceptible to this problem.  There is also a recommended setting for Project Server – SET AUTO_UPDATE_STATISTICS_ASYNC ON (documented at ) that could avoid this problem – as potentially it may be triggered if the database is waiting on the statistics to update rather than just getting on with the work!  As the problem is quite rare – and we (and our customers) usually want to get things working rather than examine the slow system – we haven’t had a good chance to confirm some of these theories. One typical symptom, apart from things just running slow, is a particular event in the ULS log indicating a slow running stored procedure.  These are verbose level messages from SharePoint Foundation Monitoring, so may not be traces that are normally written to your ULS logs, but if they and they occur for some of the queue stored procedures such as MSP_ProjQ_Lock_Next_Available_Group  or MSP_TimesheetQ_Lock_Next_Available_Group.  The event ID to look for is b4ly and it might look something like tis: 08/09/2012 16:10:25.10 Microsoft.Office.Project.Server (0x36D0) 0x2E58 SharePoint Foundation Monitoring b4ly Verbose Leaving Monitored Scope (FillTypedDataSet — MSP_ProjQ_Lock_Next_Available_Group). Execution Time=21102.179988219601 I’ve highlighted the Event ID and the stored procedure name.  The execution time is in ms – so this is 21 seconds – v. slow.  You may also be able to track an increase in this time across your logs. As mentioned above we usually found that a reboot got things moving again, but once we understood the likely causes of the problem we could be a little more focused and fix things with less inconvenience to the users. How to Fix 1. Always worth getting your database maintenance working well – for this and many other reasons. 2. Set the recommended value for SET AUTO_UPDATE_STATISTICS_ASYNC 3. If Project Server is the only application running on SQL Server then the sledgehammer approach would be to run the following command within SQL Server management studio:  DBCC FREEPROCCACHE.  This will empty the stored execution plans from the procedure cache and then ALL stored procedures will need to recompile and the new execution plan should get things working faster. 4.  If other applications are running you probably don’t want to run that DBCC command – so instead you could just recompile a couple of the likely stored procedures: Draft DB EXEC Sp_recompile MSP_ProjQ_Lock_Next_Available_Group EXEC Sp_recompile MSP_ProjQ_Get_Status_Of_Jobs_List Published DB EXEC Sp_recompile MSP_TimesheetQ_Get_Status_Of_Jobs_List EXEC Sp_recompile MSP_TimesheetQ_Lock_Next_Available_Group I hope this helps if you do run in to this kind of behavior – and hopefully by keeping SQL Server patched and using our recommended settings it isn’t something that you should see happen – but just in case – this detail might save a call to me and my colleagues.