Pfeuw, this thread has finally ended 🙂

When I delevop my packages, the OLE DB Destination usually gives an error or warning during design when I map a dataflow column to...

I don't know what your mappings are in the OLE DB Destination, but if you map the data viewer column Time to the column Time in your destination table, it...

Datetime is the correct datatype for SQL Server and (DT_DBTIMESTAMP) is the correct data type for SSIS. Since you're working with SQL Server 2005, don't convert the values to (DT_DBTIME),...

In a previous post you said that the destination is a database timestamp field.

Do you mean that you use the datatype timestamp in SQL Server? Because that should be the...

So basically a combination of wschampheleer's solution and my solution...

Or, to give you another method, you can use a script component the compose the e-mail message, store it in a variable and then use the Send Mail Task.

You can use the RAW File Destination to write your update lines to a RAW file.

But I think it won't give monstreous performance gains above a (global) temp table.

But I'm...

Paul_Harvey (2/24/2010)

---

For Field 3 - SUBSTRING([RefNumber],FINDSTRING("/",[RefNumber],3),(FINDSTRING("$",[RefNumber],3)) - FINDSTRING("/",[RefNumber],3))

This is accepted by the Derived COlumn transformation, but when I debug the package it fails with the following:

[Derived Column [4862]] Error: SSIS...

I've read all four parts of the article and I've found it very interesting and useful.

Thx

I thought this was only necessary in...

True, you can do almost everything with a script component/task, but there are solutions that don't require scripting.

Some people aren't comfortable with VB.NET or C# and what if the guy...

Well, that's genious. Why didn't I think of that? :hehe:

Griffster (2/24/2010)

---

For...

steveb. (2/24/2010)

---

One of the quickest way to do updates in bulk manner is to delete the rows from the destination table first then add the updates in the same...

You can use SQL Server Profiler to monitor all the queries against your cube.

With the right options specified, you can filter out most of the events that you don't need.

But,...

Forget about my previous post. I've tried to convert the number using the time_t specification, but it gives an overflow. Which actually makes sense. There are about 31536000 seconds in...