Adjust a date to the first day of the month.
"2005-01-15" will become "2005-01-01".

DATEADD("d",DAY(Your_date_here) * -1 + 1,Your_date_here)

Adjust a date to the last day of the month.
"2005-01-15" will become "2005-01-31".

DATEADD("d",-1,DATEADD("m",1,DATEADD("d",DAY(Your_date_here) * -1 + 1,Your_date_here)))

What happens is that at one point or another, the Excel file has been opened and saved again. It does not matter if anything has been edited between opening and saving. Upon opening the file, Excel will convert the numbers in any column with mixed data to scientific notation if they become too large or if they don't fit within the column. Once you save the file, the change becomes permanent in a way that Excel replaces the value that was in the cell with the scientific notation. When you import the saved file, you will get the scientific notation. This only happens when there is no explicit format defined for the column, or the column is defined as general.

To avoid this, before saving the Excel sheet, the offending column should be explicitly defined as text within Excel. Doing so will revert the conversions Excel made to your column. If you still see scientific notation, try expanding the width of the column. You can also do this for Excel files that have already been saved. Explicitly mark the column as text and enlarge the column to see the result. When importing the file into SSIS after these operations, life is peachy, again.

In the rest of the article, I will be talking about the maximum value for a field, but it’s also true of course for the minimum value.
The query type you will encounter most often contains a subquery where the current value is compared to the maximum value for that particular field. A simple example is presented below.

 
SELECT
  MAX(tb2.Field) AS MaxValueForField
FROM
  table1 tb1
    INNER join
  table2 tb2
    ON  tb2.foreign_key = tb1.primary_key
WHERE
  tb2.Field = (
          SELECT
            MAX(_tb2.field)
          FROM
            table1 _tb1
              INNER join
            table2 _tb2
              ON  _tb2.foreign_key = _tb1.primary_key
          WHERE
            _tb1.primary_key = tb1.primary_key
        )
GROUP BY
  tb1.primary_key
 

The subquery can be avoided with a Common Table Expression that not only speeds up the entire query but also offers a solution for another problem that I will be discussing below.

 
;
WITH tb2Sorted
 
AS
 
(
  SELECT
    Row_Number() OVER (partition BY tb1.primary_key ORDER BY tb2.Field DESC) AS RowId,
    Field
  FROM
    table1 tb1
      INNER join
    table2 tb2
      ON  tb2.foreign_key = tb1.primary_key
)
SELECT
  Field AS MaximaleIngangsDatum
FROM
  tb2Sorted
WHERE
  RowId = 1
 

This query performs better than the first one but also offers another advantage.

You can easily find the second or third largest value by simple changing the RowId. Bear in mind that this will not remove duplicates. If the values for Field are 1,1,2,5,5 then RowId 1 will yield 5 and RowId 2 will also yield 5. You can find the top 5 largest values for a certain field by simply changing the where clause to read RowId < 6.

 
;
WITH tb2Sorted
 
AS
 
(
  SELECT
    Row_Number() OVER (partition BY tb1.primary_key ORDER BY tb2.Field DESC) AS RowId,
    Field,
    RowId
  FROM
    table1 tb1
      INNER join
    table2 tb2
      ON  tb2.foreign_key = tb1.primary_key
)
SELECT
  Field AS MaximaleIngangsDatum
FROM
  tb2Sorted
WHERE
  RowId < 6
 

So I came up with this code. It lets you find a string in any column in any table on your database. It's a pretty length search if you have a lot of data and a lot of columns and tables, but it sure beats searching by hand.

Be warned! This may run a pretty long time on large databases.

 
DECLARE @searchString NVARCHAR(MAX)
DECLARE @sqlString NVARCHAR(MAX)
DECLARE @processedString NVARCHAR(MAX)
 
-- Insert your searchstring below
-- Use \ as escape character
-- examples:
--	to search for 'I'm waiting', use 'I\'m waiting'
--	to search for '25%' use '25\%'
 
SELECT
	@searchString = 'naar achter'
 
DECLARE @resultTable	TABLE
(
	table_name	NVARCHAR(255),
	COL_NAME	NVARCHAR(255),
	value_found	NVARCHAR(MAX),
	OBJECT_ID	INT,
	col_id	INT
)
 
SELECT
	@sqlString = '-- start',
	@processedString = ''
 
WHILE
	@sqlString > ''
BEGIN
 
	SELECT
		@sqlString = '',
		@processedString = ''
 
	SELECT
		@processedString = @processedString +
		CASE
			WHEN DATALENGTH(@sqlString) < 7500 -- assume a sql statement will never go beyond 500 characters
			THEN
		'
			select
				''' + QUOTENAME(SCHEMA_NAME(tbl.schema_id)) + '.' + QUOTENAME(OBJECT_NAME(tbl.OBJECT_ID)) + ''' as table_name,
				''' + col.name + ''',
				NULL as value_found,
				' + CONVERT(NVARCHAR,tbl.OBJECT_ID) + ' as object_id,
				' + CONVERT(NVARCHAR,col.column_id) + ' as col_id
		'
			ELSE
				''
		END,
		@sqlString = @sqlString +
		CASE
			WHEN DATALENGTH(@sqlString) < 7500 -- assume a sql statement will never go beyond 500 characters
			THEN
		'
			select
				top 1
				''' + QUOTENAME(SCHEMA_NAME(tbl.schema_id)) + '.' + QUOTENAME(OBJECT_NAME(tbl.OBJECT_ID)) + ''' as table_name,
				''' + col.name + ''',
				' + col.name + ' as value_found,
				' + CONVERT(NVARCHAR,tbl.OBJECT_ID) + ' as object_id,
				' + CONVERT(NVARCHAR,col.column_id) + ' as col_id
			from
				' + QUOTENAME(SCHEMA_NAME(tbl.schema_id)) + '.' + QUOTENAME(OBJECT_NAME(tbl.OBJECT_ID)) + '
			where
				' + col.name + ' like ''%' + @searchString + '%'' escape ''\''
		'
			else
				''
		end
	from
		sys.columns col
			inner join
		sys.tables tbl
			on	col.object_id = tbl.object_id
			and	col.user_type_id in (231, 167, 175, 239)
			 --	nvarchar, varchar, char, nchar
			 --	no text
			left join
		@resultTable rst
			on	col.column_id = rst.col_id
			and	col.object_id = rst.object_id
	where
		rst.table_name is null
 
	print	@sqlString
	print	@processedString	
 
	insert into
		@resultTable
	exec
		(@sqlString)
 
	insert into
		@resultTable
	exec
		(@processedString)
 
	print	'---8<-----------------------------------------------------------------------------------------------'
 
END
 
SELECT
	table_name,
	COL_NAME,
	value_found
FROM
	@resultTable
WHERE
	value_found IS not null
ORDER BY
	1, 2, 3
 

 
DECLARE @SQL NVARCHAR(MAX)	
 
SELECT
	@SQL = '-- Start '
 
WHILE @SQL > ''
BEGIN
	SELECT
		@SQL = ''
 
	SELECT
		@SQL = @SQL +
			CASE
				WHEN DATALENGTH(@SQL) < 7500 THEN
					N'alter table ' + QUOTENAME(schema_name(schema_id))
					+ N'.'
					+ QUOTENAME(OBJECT_NAME(parent_object_id))
					+ N' drop constraint '
					+ name
					+ CHAR(13) + CHAR(10) --+ 'GO' + CHAR(13) + CHAR(10)
				ELSE
					''
			END
	FROM
		sys.foreign_keys
 
	PRINT @SQL
	EXEC SP_EXECUTESQL @SQL
	PRINT '---8<------------------------------------------------------------------------------------------'
END
 
GO
 
---8<-------------------------------------------------------------
-- Uncomment the line below to drop all tables too
 
--exec sp_msforeachtable 'drop table ?'
 
PRINT 'All done'
 

This little code will list all of your tables, columns, datatypes, nullable option, identity option and will also indicate primary and foreign keys and default constraints.

 
SELECT
--	distinct
--	quotename(schema_name(tbl.schema_id)) + N'.' + quotename(tbl.name) as TableName,
--	col.column_id,
	col.name AS ColumnName,
	type_name(col.user_type_id)+
	CASE
		WHEN COLUMNPROPERTY(tbl.OBJECT_ID, col.name, 'scale') IS null THEN
			' (' + CONVERT(VARCHAR,COLUMNPROPERTY(tbl.OBJECT_ID, col.name, 'precision')) + ')' --is null --convert(varchar, col.max_length)
		WHEN col.user_type_id in (60, 106, 108, 122) THEN
			' (' + CONVERT(VARCHAR,col.PRECISION) + ',' + CONVERT(VARCHAR, col.scale) + ')'
		ELSE ''
	END	AS DataType,
	CASE col.is_nullable
		WHEN 0 THEN 'NOT '
		ELSE ''
	END
		+	 'NULL' AS Nullable,
	CASE col.is_identity
		WHEN 0 THEN ''
		ELSE 'IDENTITY (' + CONVERT(VARCHAR, idc.seed_value) + ',' + CONVERT(VARCHAR, idc.increment_value) + ')'
	END	AS [IDENTITY],
	CASE
		WHEN ixc.column_id IS not null THEN
			'PK'
		WHEN fkc.constraint_object_id IS not null THEN
			QUOTENAME(schema_name(tb2.schema_id)) + N'.' + QUOTENAME(tb2.name) + N'.' + fcl.name
		ELSE ''
	END AS Key_Constraint,
	CASE
		WHEN dfc.OBJECT_ID IS not null THEN
			dfc.definition
		ELSE ''
	END AS [DEFAULT CONSTRAINT]
FROM
	sys.tables tbl
		INNER join
	(
		SELECT
			QUOTENAME(schema_name(tbl.schema_id)) + N'.' + QUOTENAME(tbl.name) AS TableName,
			col.name,
			col.column_id,
			col.OBJECT_ID,
			col.is_identity,
			col.is_nullable,
			col.PRECISION,
			col.scale,
			col.user_type_id,
			1 AS SortKey
		FROM
			sys.columns col
				INNER join
			sys.tables tbl
				ON	tbl.OBJECT_ID = col.OBJECT_ID
 
		UNION 
 
		SELECT
			QUOTENAME(schema_name(tbl.schema_id)) + N'.' + QUOTENAME(tbl.name) AS TableName,
			QUOTENAME(schema_name(tbl.schema_id)) + N'.' + QUOTENAME(tbl.name) AS TableName,
			0,
			tbl.OBJECT_ID,
			0,
			0,
			0,
			0,
			0,
			0
		FROM
			sys.tables tbl
 
	)	col
		ON	tbl.OBJECT_ID = col.OBJECT_ID
		LEFT join
	sys.key_constraints pkc
		ON	pkc.parent_object_id = tbl.OBJECT_ID
		and	pkc.type = 'PK'
		LEFT join
	sys.index_columns ixc
		ON	ixc.OBJECT_ID = tbl.OBJECT_ID
		and	ixc.index_id = pkc.unique_index_id
		and	ixc.column_id = col.column_id
		LEFT join
	sys.foreign_key_columns fkc
		ON	fkc.parent_object_id = tbl.OBJECT_ID
		and	fkc.parent_column_id = col.column_id
		LEFT join
	sys.columns fcl
		ON	fkc.referenced_column_id = fcl.column_id
		and	fkc.referenced_object_id = fcl.OBJECT_ID
		LEFT join
	sys.tables tb2
		ON	fcl.OBJECT_ID = tb2.OBJECT_ID
		LEFT join
	sys.default_constraints dfc
		ON	dfc.parent_object_id = tbl.OBJECT_ID
		and	dfc.parent_column_id = col.column_id
		LEFT join
	sys.identity_columns idc
		ON	idc.OBJECT_ID = tbl.OBJECT_ID
		and	idc.column_id = col.column_id
WHERE
	tbl.type = 'U'
ORDER BY
	QUOTENAME(schema_name(tbl.schema_id)) + N'.' + QUOTENAME(tbl.name),
	SortKey,
	ColumnName
 

But to my dismay, not this time. I powercycled the LaCie Internet Space and it booted allright, the blue light came on and then ... nothing.

I dismantled the complete thing, took out the drive (Hitachi Deskstar 1TB 7200 RPM) and plugged it into my usb/sata converter and again, nothing happened.

Cue slight panic as my entire music collection was stored on this, as well as my wife's photo portfolio. Since the drive did seem to spin up, I suspected the logic board. I scoured the internet for an exact hard disk and found one. In the meantime, I had also bought a second hand LaCie Internet Space, thinking there would be a Hitachi Deskstar inside but it turned out to be a Samsung HD103SI. I discarded the second LaCie Internet Space and put all my hopes on the spare disc, which, by the way, was rather expensive due to the floods in Thailand.

I switched the logic boards, connected the Hitachi to the usb/sata converter and nothing happened. I replaced the logic boards and in a final act of lucidity decided to put the hard drive into the replacement LaCie Internet Space casing. All of a sudden, the disc sprang back to live, started rattling merrily and I could see the disc once again in my network places.

It seemed that the hard disc wasn't at fault but the original LaCie Internet Sace casing. Go figure. Why the drive wasn't recognised by the usb/sata converter is beyond me. I'm copying 1 TB of data to my other NAS right now and I'm never buying anything from LaCie again.

1. Internal error: Invalid enumeration value. Please call customer support! is not a valid value for this element.
2. An error occurred while parsing the 'StorageMode' element at line 1, column 6751 ('http://schemas.microsoft.com/analysisservices/2003/engine' namespace) under Load/ObjectDefinition/Dimension/StorageMode.
3. Errors in the metadata manager. An error occurred when instantiating a metadata object from the file, '\\?\N:\SSAS\OLAP\Data\RMT_CUBE\RP DIM TIME.12.dim.xml'

Additionally, building the project and then deploying using the wizard results in a similar string of error messages

The column in the second error and the file referenced in the third error may vary. The filename in the third error however is worth noticing because it's the dimension that is stopping you from deploying. In this case, my dimension was named "RMT DIM TIME".

On the internet, you will find many solutions advising you to drop the database and redeploy. I have found a way that is less time consuming and a lot less worrying than simply dropping your entire database.

I was able to solve the problem by simply re-processing the offending dimension in SSMS (Sql Server Management Studio). After re-processing the offending dimension, life was peachy again.

Hope this helps you. Apparently, installing the latest SSAS service pack would also solve the problem but that's not always a straightforward solution in a managed environment.

I installed Microsoft's Powertoy "Microsoft RAW Image Thumbnailer and Viewer for Windows XP" but that failed with a "cannot load image" message. I was at wit's end. Until ...

Finally, I discovered that Microsoft's Live Photo Gallery that is part of Vista and Windows 7 and that does suppot CR2 RAW files was available for Windows XP. Download it here: http://download.live.com/photogallery and then uncheck all the vile software you don't need. Et voila, CR2 RAW support from your explorer in Windows XP.

Whew. That only took me about 2 hours to fix.

I consulted the Event viewer, but that only yielded the following message.

Log Name: Application
Source: MSSQL$SE2008R2
Date: 1-6-2010 12:38:43
Event ID: 17120
Task Category: Server
Level: Error
Keywords: Classic
User: N/A
Computer: [censored]
Description:
SQL Server could not spawn lazy writer thread. Check the SQL Server error log and the Windows event logs for information about possible related problems.

Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="MSSQL$SE2008R2" />
<EventID Qualifiers="49152">17120</EventID>
<Level>2</Level>
<Task>2</Task>
<Keywords>0x80000000000000</Keywords>
<TimeCreated SystemTime="2010-06-01T10:38:43.000Z" />
<EventRecordID>7660</EventRecordID>
<Channel>Application</Channel>
<Computer>censored</Computer>
<Security />
</System>
<EventData>
<Data>lazy writer</Data>
<Binary>E042000010000000130000004400540041003000300035003400360030005C0053004500320030003000380052003200000000000000</Binary>
</EventData>
</Event>

After checking the internet for a solution, I came up with none. I decided to check the SQL Server Log file and it contained the following line.

I/O affinity turned on, processor mask 0x00000002. Disk I/Os will execute on CPUs per affinity I/O mask/affinity64 mask config option. This is an informational message only; no user action is required.

I them remembered that I had tried to assign CPU2 to handle I/O requests. Since the only way to remedy something that goes wrong and which you haven't got a clue about is to retrace your steps, my next step would be to undo that setting. But the SQL Service wouldn't start and the -I command line option didn't work.

I started SQL Server manually as described in this MSDN article: How to: Start an Instance of SQL Server (sqlservr.exe). I specifically used the -f option to make sure SQL Server would start with as little configuration as possible.

SQL Server started and I launched my Enterprise Manager and checked the "Automatically set I/O affinity mask for all processors".

I then closed the Enterprise Manager, quit the command prompt and started the service. Life was peachy again.