SQL Server-查询以返回唯一的ID，该ID在另一列中每个ID具有不同的值

Question

SQL Server 2016 Enterprise - Let's say I run a query on our main table to bring back all records from ID's 1 through 5. I would get something like this:

ID  Status  BatchID
1   Active  493fd8bc
1   Active  493fd8bc
1   Active  493fd8bc
1   Active  493fd8bc
1   Active  493fd8bc
2   Active  2d91c2f2
2   Active  2d91c2f2
3   Active  2a1338ac
3   Active  2a1338ac
3   Active  2a1338ac
4   Active  0c469bcc
4   Active  0c469bcc
4   Active  0c469bcc
4   Active  0c469bcc
4   Active  0c469bcc
4   Active  0c469bcc
4   Active  0c469bcc
5   Active  ca8896bd
5   Active  ca8896bd
5   Active  ca8896bd

The above is what we would expect to see, such that each ID has the same BatchID associated with it (additional columns that differentiate each record are irrelevant to this question).

We've run into some issues where we are seeing this is not the case for all ID's . For example let's say we ran the same query and it returned this instead:

ID  Status  BatchID
1   Active  493fd8bc
1   Active  493fd8bc
1   Active  752e8d5d
1   Active  752e8d5d
1   Active  493fd8bc
2   Active  2d91c2f2
2   Active  2d91c2f2
3   Active  f2191595
3   Active  2a1338ac
3   Active  2a1338ac
4   Active  feaf9567
4   Active  f5d64c8c
4   Active  e5eff3a2
4   Active  0c469bcc
4   Active  49e5e2f2
4   Active  0c469bcc
4   Active  e5eff3a2
5   Active  ca8896bd
5   Active  ca8896bd
5   Active  ca8896bd

In this case we see that ID's 1, 3, and 4's BatchID do NOT all match.

If I wanted to run a query on all ID's and have it return to me a list of ID's whose BatchID's do NOT match for all records of that ID , how would I accomplish this?

The results would look like this if my table contained only the data in the 2nd example above.

ID
1
3
4

Alternatively, we'd be just as well-suited if the results were returned as entire records as long as the record contains the ID , and there is only one record per ID . The BatchID returned is of no consequence.

Answer 1

Use aggregation and count the number of distinct batches for each ID :

SELECT ID
FROM yourTable
GROUP BY ID
HAVING COUNT(DISTINCT BatchID) > 1;

Answer 2

If you happen to want the entire row, you can do:

select t.*
from t
where exists (select 1
              from t t2
              where t2.id = t.id and t2.batchid <> t.batchid
             );

If you just want the ids, then Tim's answer is the better approach.