If I understood your requirements, the following is what you'll need. Offhand about your test data, everything had an id of 1 before I modified it.

Short form of what this does: Pivot the test data, check each value against the validation set, and return outliers. Only return each ID once in case of multiple failures.

declare @Table1 table

(id int,

StartDt datetime,

EndDt datetime)

insert into @Table1

select 1,'2009-11-28 00:00:00.000',''

union all

select 2,'2009-11-07 00:00:00.000','2009-11-27 00:00:00.000'

union all

select 3,'2009-07-01 00:00:00.000','2009-11-06 00:00:00.000'

union all

select 4,'2009-06-19 00:00:00.000','2009-06-30 00:00:00.000'

select * from @Table1

declare @Table2 table

(id int,

StartDt datetime,

EndDt datetime)

insert into @Table2

select 1,'2008-11-05 00:00:00.000','2009-11-06 00:00:00.000'

union all

select 2,'2007-11-29 00:00:00.000','2008-11-04 00:00:00.000'

union all

select 3,'2007-09-04 00:00:00.000','2007-09-04 00:00:00.000'

union all

select 4,'2007-07-02 00:00:00.000','2007-07-12 00:00:00.000'

union all

select 5,'2006-09-05 00:00:00.000','2007-06-21 00:00:00.000'

select * from @Table2

;WITH pivotedData AS

(SELECT id, StartDt as TestDT

FROM@Table1

UNION ALL

SELECT id, EndDt AS TestDT

FROM@Table1

)

SELECT DISTINCT

pd.ID

FROM

pivotedData AS pd

LEFT JOIN

@table2 AS t2

ONpd.TestDT >= t2.StartDt

AND pd.TestDt <= t2.EndDt

WHERE

t2.id IS NULL