Knowledge

Project Objective

The objective of this project is to modernize and future-proof the enterprise collaboration and workflow platforms by upgrading K2, SharePoint, and underlying infrastructure across all environments (QA, DEV, PROD, and DR). This initiative will:

• Transition K2 Five 5.8.1 → 5.9 on Windows Server 2022 with SQL Server 2022 (AlwaysOn) for improved scalability, security, and supportability.

• Upgrade SharePoint 2019 → SharePoint Server Subscription Edition (SPSE), adopting Microsoft’s evergreen release model to ensure continuous feature and security updates.

• Migrate the standalone SharePoint 2016 Document Center farm → SPSE using the supported Database Attach Method, enabling consolidation and consistent user experience across farms.

• Establish full production and disaster recovery parity, with SQL AlwaysOn and VIP-based failover to strengthen resilience and business continuity.

• Standardize QA and DEV environments as simplified standalone setups to enable safe validation and pre-production testing.

The upgrade will follow a parallel side-by-side migration strategy, minimizing downtime, enabling rollback if needed, and ensuring clean configurations in line with Microsoft and Nintex best practices. This program will deliver a secure, stable, and supportable platform foundation for business processes, content management, and workflow automation.

Powershell

Updated Stored Procedure

sql
CREATE PROCEDURE SP_GetAvailableTasksForTimesheet
    @TimeSheetID INT
AS
BEGIN
    -- Declare variables to hold the WeekStartDate and WeekEndDate
    DECLARE @WeekStartDate DATE;
    DECLARE @WeekEndDate DATE;
    DECLARE @UserName NVARCHAR(100);

    -- Fetch the WeekStartDate, WeekEndDate, and UserName from the TimeSheet table
    SELECT 
        @WeekStartDate = TimeSheetStartDate,
        @WeekEndDate = TimeSheetEndDate,
        @UserName = EmployeeNWName
    FROM 
        TimeSheet
    WHERE 
        TimeSheetID = @TimeSheetID;

    -- Select tasks that overlap with the given WeekStartDate and WeekEndDate
    -- and are not already present in the TimeEntry table
    SELECT DISTINCT
        T.TaskID,
        T.TaskName,
        T.TaskStartDate,
        T.TaskEndDate,
        PR.ProjectID,
        P.ProjectName
    FROM Task T
    JOIN ProjectResource PR ON PR.ProjectID = T.ProjectID
    JOIN Project P ON PR.ProjectID = P.ProjectID
    WHERE PR.EmployeeNWName = @UserName
      AND P.Progress = 'Open'
      AND T.Active = 1
      AND (
            -- Check overlap between WeekStartDate-WeekEndDate and TaskStartDate-TaskEndDate
            @WeekStartDate <= ISNULL(T.TaskEndDate, '9999-12-31') AND @WeekEndDate >= T.TaskStartDate
            AND
            -- Check overlap between WeekStartDate-WeekEndDate and ResourceStartDate-ResourceEndDate
            @WeekStartDate <= ISNULL(PR.ResourceEndDate, '9999-12-31') AND @WeekEndDate >= PR.ResourceStartDate
            AND
            -- Check overlap between TaskStartDate-TaskEndDate and ResourceStartDate-ResourceEndDate
            T.TaskStartDate <= ISNULL(PR.ResourceEndDate, '9999-12-31') AND ISNULL(T.TaskEndDate, '9999-12-31') >= PR.ResourceStartDate
          )
      AND NOT EXISTS (
            SELECT 1
            FROM TimeEntry TE
            WHERE TE.TaskID = T.TaskID
              AND TE.TimeSheetID = @TimeSheetID
        );
END;

SharePoint Server Subscription Edition (SE) supports the following versions of Windows Server:

1. Windows Server 2019
2. Windows Server 2022

Sharepoint server upgrade

SharePoint upgrades from SharePoint 2019 to SharePoint Server Subscription Edition does not support In place upgrade.

We will need Install and configure SharePoint Server Subscription Edition on the new farm

SQL Server

SharePoint Server Subscription Edition (SE) supports the following versions of SQL Server:

1. SQL Server 2019 (Standard or Enterprise edition)
2. SQL Server 2022 (Standard or Enterprise edition)

We will need to create new SQL server instance for the 

SharePoint Server Subscription Edition (SE) farm and then backup and restore DBs from old to new SQL instance

If we already have SQL Server 2019 in place it will streamline the upgrade process, as it's supported by both SharePoint versions.

Updated Stored Procedure

Here is how the updated SP_GetMissingTimesheets stored procedure might look, incorporating the overlap logic:

sql
ALTER PROCEDURE [TT].[SP_GetMissingTimesheets]
AS
BEGIN
    DECLARE @Today DATE = GETDATE();
    DECLARE @WeekStartDate DATE;
    DECLARE @WeekEndDate DATE;

    -- Clear previous entries in the MissingTimesheets table (optional)
    -- DELETE FROM MissingTimesheets WHERE Created < DATEADD(WEEK, -5, GETDATE());

    -- Loop through the past 5 weeks
    DECLARE @WeekOffset INT = 1;

    -- Truncate the MissingTimesheets table to clear any existing data
    TRUNCATE TABLE MissingTimesheets;

    WHILE @WeekOffset < 5
    BEGIN
        -- Calculate the start and end dates of the week using the function
        SELECT @WeekStartDate = WeekStartDate, @WeekEndDate = WeekEndDate
        FROM Fun_GetWeekStartEndDate(@Today, -@WeekOffset);

        -- Insert missing timesheets into the permanent table, excluding records where date ranges do not overlap
        INSERT INTO MissingTimesheets (EmployeeEmail, WeekStartDate, WeekEndDate)
        SELECT DISTINCT E.EmployeeEmail, @WeekStartDate, @WeekEndDate
        FROM Resource E
        LEFT JOIN TimeSheet TS ON E.EmployeeNWName = TS.EmployeeNWName
            AND TS.TimeSheetStartDate = @WeekStartDate
            AND TS.TimeSheetEndDate = @WeekEndDate
        WHERE (TS.TimeSheetID IS NULL OR TS.TimeSheetStatus IN ('Not Started', 'Saved', 'Returned'))
          AND EXISTS (
              SELECT 1
              FROM Task T
              JOIN ProjectResource PR ON PR.ProjectID = T.ProjectID
              WHERE PR.EmployeeNWName = E.EmployeeNWName
                AND (
                    -- Check overlap between WeekStartDate-WeekEndDate and TaskStartDate-TaskEndDate
                    (@WeekStartDate <= ISNULL(T.TaskEndDate, '9999-12-31') AND @WeekEndDate >= T.TaskStartDate)
                    AND
                    -- Check overlap between WeekStartDate-WeekEndDate and ResourceStartDate-ResourceEndDate
                    (@WeekStartDate <= ISNULL(PR.ResourceEndDate, '9999-12-31') AND @WeekEndDate >= PR.ResourceStartDate)
                    AND
                    -- Check overlap between TaskStartDate-TaskEndDate and ResourceStartDate-ResourceEndDate
                    (T.TaskStartDate <= ISNULL(PR.ResourceEndDate, '9999-12-31') AND ISNULL(T.TaskEndDate, '9999-12-31') >= PR.ResourceStartDate)
                )
          );

        SET @WeekOffset = @WeekOffset + 1;
    END

    -- Select and format the missing timesheets
    SELECT 
        ROW_NUMBER() OVER (ORDER BY E.EmployeeEmail) AS MissingTimesheetID,
        E.EmployeeNWName,
        E.EmployeeEmail,
        E.EmployeeDisplayName,
        '<ul>' + STRING_AGG('<li>' + CONVERT(VARCHAR, M.WeekStartDate, 101) + ' - ' + CONVERT(VARCHAR, M.WeekEndDate, 101) + '</li>', '') + '</ul>' AS MissingWeeks
    FROM 
        MissingTimesheets M
    JOIN Resource E ON M.EmployeeEmail = E.EmployeeEmail
    WHERE 
        M.Created >= DATEADD(WEEK, -5, @Today)
    GROUP BY 
        E.EmployeeNWName,
        E.EmployeeEmail,
        E.EmployeeDisplayName;

END;

Stored Procedure

sql
CREATE PROCEDURE SP_AddEmployeeToGeneralProjects
    @EmployeeNWName NVARCHAR(100)
AS
BEGIN
    -- Declare a variable to hold the ProjectID
    DECLARE @ProjectID INT;

    -- Cursor to iterate through projects with Portfolio 'General'
    DECLARE ProjectCursor CURSOR FOR
    SELECT ProjectID
    FROM Project
    WHERE PortfolioID = (SELECT PortfolioID FROM Portfolio WHERE PortfolioName = 'General');

    OPEN ProjectCursor;
    FETCH NEXT FROM ProjectCursor INTO @ProjectID;

    WHILE @@FETCH_STATUS = 0
    BEGIN
        -- Check if EmployeeNWName is already in the ProjectResource table for the current ProjectID
        IF NOT EXISTS (
            SELECT 1
            FROM ProjectResource
            WHERE ProjectID = @ProjectID
            AND EmployeeNWName = @EmployeeNWName
        )
        BEGIN
            -- Insert the EmployeeNWName into the ProjectResource table
            INSERT INTO ProjectResource (ProjectID, EmployeeNWName, ResourceStartDate, Active)
            VALUES (@ProjectID, @EmployeeNWName, GETDATE(), 1);  -- Assuming Active and StartDate
        END

        FETCH NEXT FROM ProjectCursor INTO @ProjectID;
    END

    CLOSE ProjectCursor;
    DEALLOCATE ProjectCursor;
END;

Updated Stored Procedure

sql
ALTER PROCEDURE [TT].[SP_GetMissingTimesheets]
AS
BEGIN
    DECLARE @Today DATE = GETDATE();
    DECLARE @WeekStartDate DATE;
    DECLARE @WeekEndDate DATE;

    -- Clear previous entries in the MissingTimesheets table (optional)
    -- DELETE FROM MissingTimesheets WHERE Created < DATEADD(WEEK, -5, GETDATE());

    -- Loop through the past 5 weeks
    DECLARE @WeekOffset INT = 1;

    -- Truncate the MissingTimesheets table to clear any existing data
    TRUNCATE TABLE MissingTimesheets;

    WHILE @WeekOffset < 5
    BEGIN
        -- Calculate the start and end dates of the week using the function
        SELECT @WeekStartDate = WeekStartDate, @WeekEndDate = WeekEndDate
        FROM Fun_GetWeekStartEndDate(@Today, -@WeekOffset);

        -- Insert missing timesheets into the permanent table
        INSERT INTO MissingTimesheets (EmployeeEmail, WeekStartDate, WeekEndDate)
        SELECT E.EmployeeEmail, @WeekStartDate, @WeekEndDate
        FROM Resource E
        LEFT JOIN TimeSheet TS
        ON E.EmployeeNWName = TS.EmployeeNWName
        AND TS.TimeSheetStartDate = @WeekStartDate
        AND TS.TimeSheetEndDate = @WeekEndDate
        WHERE TS.TimeSheetID IS NULL
           OR TS.TimeSheetStatus IN ('Not Started', 'Saved', 'Returned');

        SET @WeekOffset = @WeekOffset + 1;
    END

    -- Select and format the missing timesheets
    SELECT 
        ROW_NUMBER() OVER (ORDER BY EmployeeEmail) AS MissingTimesheetID,
        E.EmployeeNWName,
        E.EmployeeEmail,
        E.EmployeeDisplayName,
        '<ul>' + STRING_AGG('<li>' + CONVERT(VARCHAR, M.WeekStartDate, 101) + ' - ' + CONVERT(VARCHAR, M.WeekEndDate, 101) + '</li>', '') + '</ul>' AS MissingWeeks
    FROM 
        MissingTimesheets M
    JOIN Resource E ON M.EmployeeEmail = E.EmployeeEmail
    WHERE 
        M.Created >= DATEADD(WEEK, -5, @Today)
    GROUP BY 
        E.EmployeeNWName,
        E.EmployeeEmail,
        E.EmployeeDisplayName;

END;

sql
CREATE PROCEDURE SP_FindMissingTimesheets
AS
BEGIN
    DECLARE @Today DATE = GETDATE();
    DECLARE @WeekStartDate DATE;
    DECLARE @WeekEndDate DATE;

    -- Clear previous entries in the MissingTimesheets table (optional)
    -- DELETE FROM MissingTimesheets WHERE Created < DATEADD(WEEK, -5, GETDATE());

    -- Loop through the past 5 weeks
    DECLARE @WeekOffset INT = 0;

    WHILE @WeekOffset < 5
    BEGIN
        -- Calculate the start and end dates of the week using the function
        SELECT @WeekStartDate = WeekStartDate, @WeekEndDate = WeekEndDate
        FROM dbo.Fun_GetWeekStartEndDate(@Today, -@WeekOffset);

        -- Insert missing timesheets into the permanent table
        INSERT INTO MissingTimesheets (EmployeeEmail, WeekStartDate, WeekEndDate)
        SELECT E.EmployeeEmail, @WeekStartDate, @WeekEndDate
        FROM Resource E
        LEFT JOIN TimeSheet TS
        ON E.EmployeeNWName = TS.EmployeeNWName
        AND TS.TimeSheetStartDate = @WeekStartDate
        AND TS.TimeSheetEndDate = @WeekEndDate
        WHERE TS.TimeSheetID IS NULL
           OR TS.Status IN ('Not Started', 'Saved');

        SET @WeekOffset = @WeekOffset + 1;
    END

    -- Select and format the missing timesheets
    SELECT 
        ROW_NUMBER() OVER (ORDER BY EmployeeEmail) AS ID,
        EmployeeEmail,
        '<ul>' + STRING_AGG('<li>' + CONVERT(VARCHAR, WeekStartDate, 101) + ' - ' + CONVERT(VARCHAR, WeekEndDate, 101) + '</li>', '') + '</ul>' AS MissingWeeks
    FROM 
        MissingTimesheets
    WHERE 
        Created >= DATEADD(WEEK, -5, @Today)
    GROUP BY 
        EmployeeEmail;
END;

__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT__PRESENT

Delete k2 workflow instances in bulk

K2 workflow instances bloats the database over the time period and at some point you would want to delete the old workflow instance which are completed or in error status.

Unfortunately, there is no way to delete specific instances of workflow, in other words, you could not select specific instance and delete them. However, you can delete then in a bulk.

For instance, you would want to delete old workflow instances, which are completed and occupying the significant space in database. Instances where you are not going to report anything. Or, you just want to delete then as a part of clean up activity.

You can follow below steps to do so.

1. If you have 4.7 vision installed you can go to management workspace. For older version, you can follow similar steps
2. Expand workflow server node.
3. Go to your category to find workflow for which you want to delete workflow instances.
4. Click on versions.
5. Select version you want to delete.

6. Check “Delete all historical (log) data for selected versions”

Be careful though, “Delete all historical (log) data for selected versions” option deletes all workflow instances including active as well. Basically, it wipes out all the instances for that version from k2 database.

Different ways to create smartobjects from SQL service instance

I asked the question to k2 support, What is the difference between Create and generate smartobject? and they provided below answer, thought would be helpful to share.

There are three different options available to create/manage SmartObject from a SQL Service Instance via SmartObject Service Tester tool:

a.  From service instance > Create SmartObjects
- manages the creation and updates of all SQL SmartObjects of the service instance as well as SmartObject's category as a set (no control over displayname and guid)
- allows for the appending of a string to all of the SmartObject system name (no control over the overall displayname, systemname, guid)
- can be used to update all SmartObject as a set if there are changes to the database objects schema(tables, views, sprocs)
- displayname, systemname and guid will remain the same (if targeting the same set of SmartObject)

b.  From service instance > Generate SmartObjects
- manages the creation and updates of all SQL SmartObjects of the service instance as a set (no control over displayname, systemname, category, guid)
- can be used to update all SmartObject as a set if there are changes to the database objects schema (tables, views, sprocs)
- displayname, systemname and guid will remain the same

c.  From Service Object > Create SmartObject (singular)
- allows for managing of the SystemName/DisplayName, GUID, and category SmartObject is created in
- can be used to regenerate/update the existing SmartObject by targeting the 'SystemName' of the SmartObject (with the 'Check Name' button) and returning the current GUID for this SmartObject (with the 'Get Existing Guid' button) before publishing it again
** if not returning the existing name and GUID, a new SmartObject with different value will be created

* all three options mainly used if there are NO customization with the SmartObject and the default generated functionality is desired; as regenerating the SmartObject with the same SystemName and GUID will remove any customization previously done; if SmartObject was customized, manually changes to these objects in the designer that they were designed in (K2 Designer, K2 Studio, K2 for Visual Studio) will preserve the customization.

K2 Interview Questions: Workspace, Part 1 (Intermidiate)

All K2 Interview Questions

What are Worklist Item/task statuses?

0 = Available	The item has not been opened yet
1 = Open	The item was previously opened by the current user
2 = Allocated	The item was opened by another user
3 = Sleep	The item was set to sleep for a specified amount of time
4 = Completed	The item was already completed by this or another user

In a majority of cases, once a user has opened the task, the task status is Open for that user and Allocated for everyone else.

K2 Interview Questions: Workflows, Part 2 (Advance)

All K2 Interview Questions

How do you fix the process in errors?
1. Sometimes the error is because of network issues, some connection problem or no proper inputs. If you just go to the error profile in the workspace and retry the error it fixes the problem.
2. If above approach does not resolve the error you can use the K2 Process Management tool in Visual Studio.
 As shown in the figure below. you can select process clicking View->K2 Process Management tool in visual studio. Select the process in error and open the process and fix the error.

After fixing the error, go back to the k2 process management and click redeploy the process.

You can provide the label and deploy the process. It will fix that instance of the process. Then you can fix the processes with similar error hitting the retry button and selecting the fixed version of the process.

How do you debug the process if  the Visual studio is not installed on the K2 box?

http://spandk2.blogspot.com/2012/11/how-to-debug-k2-process.html

K2 Interview Questions

K2 Interview Questions: Workflows, Part 1 (Advance)

All K2 Interview Questions

What is IPC event wizard?

• The IPC (The Inter Process Communication) event wizard is normally used to implement a Parent-Child workflow or sub-workflow requirement.
• Also it is useful when you want to split a larger workflow into sub-processes, when you have processes that are generically re-usable, or when you want to start multiple instances of a child workflow from a parent workflow.
• You can use it to set Folio, Originator, and data field values for a Child workflow as part of the wizard.
• There are two options while configuring the call to the child process
  Synchronous: Parent workflow waits for Child workflow to complete.
  Asynchronous: 'fire-and-forget' style, in which the Parent workflow continues immediately after starting the Child workflow.
• Use advanced destination rules (Plan per Slot, No Destinations) to allow one Parent workflow to start multiple Children

What are different activity planning options in Advance destination rule? and explain them.

There are three main Activity Planning options: Plan Just Once, Plan Per Destination, and Plan Per Slot.

           Plan Just Once : 

• This default option. K2 will execute the events in the Activity prior to the Client Event just one time, regardless of how many destinations or slots have been defined. The events after the Client event will be executed each time a user completes a task. This is the default behavior for Activities.
• This option is most appropriate for high-volume scenarios where a large group gets the task but only one user will ever need to open the task.
• When this option is selected, there is only one activity instance, which is shared amongst all destinations for the task.
  
  Plan per destination:
• This approach is most often used when more than one person will open the task, and you need to capture input from each user into the same activity datafields.
• K2 creates separate copies of the Activity Instance for each destination
  
  Plan per destination – All at Once :
• K2 will execute the events in the activity for each destination in the activity in parallel fashion. Each event in the activity is repeated for each destination, and then K2 moves on to the next event in the activity and repeats that event for each destination.
• This planning option specifies that all destinations will get the task at the same time (Parallel). Any one of the destination users can open the task and complete it. If any of the outcomes are true for the activity, the other copies of the task are removed from the other user’s task lists, and K2 completes the activity.

• Plan per destination – One at a Time : K2 will execute the events in the activity for each destination in the activity in serial fashion. K2 starts with the first destination and executes the Events in the activity once for that destination. If the Outcome is false, K2 moves on to the second destination and then executes the events in sequence for the second destination. This is repeated for each destination in the activity until the Outcome succeeds or no more destinations are available.
  
  Plan per slot – (no destinations) : 
• This setting is normally used for activities that only contain server events when you want to repeat the events in the activity N-number of times for each slot in the activity. 
• It is similar to the Plan Per Destination – All at Once setting except that K2 does not use the Destinations to determine how many times to execute the events but rather the number of slots specified on the next page of the Destination Rule wizard.

What is use of Update Design Templates in K2 visual studio workflows?

The purpose of this tool is to apply enhanced code templates when you have upgraded your K2 environment to a newer version of the platform.The Update Design Templates command in the K2 design tools will wipe out any customizations you may have applied because it restores the original templates for all items and will undo any customization.

What event is use to write a code in workflow?

Default server event wizard is used to write a code in workflow. It is only available in K2 for Visual studio.

How can be method be called from custom assembly or service?

There are alternative approaches to using code in your workflows: 

1. You can add reference to the assembly and call method in a class adding using namespace reference in default server code event

2. You can use the Code Reference Event to call methods for referenced assemblies and services.

When a workflow must interact with a system that is not exposed as a K2 SmartObject, but you do not wish to write code, the code reference event comes to the rescue. This event can be used to call methods on web services (asmx), WCF services and assemblies (.NET and COM dll’s).
3. If you need to call the same assembly/Service multiple times, it is recommended to expose that assembly/Service as a SmartObject.

• By using one of the available EndPoint Service Brokers,
• By writing a custom Broker that exposes the necessary functions. This will allow you (and other designers) to call the methods in the Service/Assembly easily without having to write code.

What is Asynchronous server event?

Asynchronous Server Events are used when you want the server event to wait for an external system to call back. The external system will use a serial number to complete the task and tell the K2 server event to continue.

What are the different places in workflow you can enable and handle exception?

You can handle the exception at event, activity, line, escalation rule, escalation action and at process level

All K2 Interview Questions