Trying to group by a value in SQL - sql

Question

I have a table called TESTTABLE
The table script and some sample date
CREATE TABLE Test_Table(
NODE VARCHAR(10) NOT NULL PRIMARY KEY
,EVENTID CHAR(255) NOT NULL
,TYPE INTEGER NOT NULL
,FIRSTOCCURRENCE VARCHAR(16) NOT NULL
,LASTOCCURRENCE VARCHAR(16) NOT NULL
,TALLY INTEGER NOT NULL
,TICKETNUMBER VARCHAR(20)
,TIME_DELTA VARCHAR(5)
);
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('Washington','ReachabilityProblem',2,'12/13/2017 23:24','12/13/2017 23:24',1,NULL,'1 sec');
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('San Diego','ReachabilityProblem',1,'12/13/2017 23:23','12/13/2017 23:23',1,NULL,NULL);
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('Richmond','ReachabilityProblem',1,'12/13/2017 14:23','12/13/2017 14:23',1,NULL,NULL);
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('Richmond','ReachabilityProblem',1,'12/13/2017 23:23','12/13/2017 23:23',1,NULL,NULL);
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('New York','ReachabilityProblem',2,'12/13/2017 23:24','12/13/2017 23:24',1,NULL,'1 sec');
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('New York','ReachabilityProblem',2,'12/13/2017 11:32','12/13/2017 11:33',2,NULL,'1 sec');
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('New York','ReachabilityProblem',1,'12/13/2017 16:35','12/13/2017 16:35',1,NULL,NULL);
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('Landsdown','ReachabilityProblem',2,'12/13/2017 23:24','12/13/2017 23:24',1,NULL,'1 sec');
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('Houston','ReachabilityProblem',2,'12/13/2017 14:24','12/13/2017 14:24',1,NULL,'1 sec');
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('Houston','ReachabilityProblem',1,'12/13/2017 11:31','12/13/2017 11:32',2,NULL,NULL);
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('Dallas','ReachabilityProblem',1,'12/13/2017 23:23','12/13/2017 23:23',1,NULL,NULL);
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('Dallas','ReachabilityProblem',2,'12/13/2017 23:24','12/13/2017 23:24',1,NULL,'1 sec');
INSERT INTO Test_Table(NODE,EVENTID,TYPE,FIRSTOCCURRENCE,LASTOCCURRENCE,TALLY,TICKETNUMBER,TIME_DELTA) VALUES ('Coco Beach','ReachabilityProblem',1,'12/13/2017 23:23','12/13/2017 23:23',1,NULL,NULL);
I'm trying to obtain this
I have tried this
Select DATEDIFF(Day, GETDATE(), DATEADD(HOUR, 15, GETDATE()))
Select
[NODE]
,[EVENTID]
,[TYPE]
,[FIRSTOCCURRENCE]
,LASTOCCURRENCE]
,DATEDIFF(Minute, FIrst OCCURENCE, LAST OCCURENCE) as [Outage in MIN]
,[TicketNumber]
,[Severity]
,Tally]
From
[XYZ].[XYZ].[XYZ_STATUS]
Where
[FIRST OCCURRENCE] >= DATEADD(hh, -24, GETDATE())
Group by node;
Please help a rookie

Answer 1

Group by returns a relation/table with a row for each group, if you are going to use the GROUP BY clause, so in your SELECT statement you can only select the column that you are grouping by and use aggregate functions on that column because the other columns will not appear in the resulting table.
Maybe this is what you want...
Select
DATEDIFF (DAY, GETDATE(), DATEADD(Hour, 15, GETDATE())),
,Node
,EventID
,Type
,Severity
,Tally
FROM xyz.xyz.xyz_status
GROUP BY Node,EventID,Type,Severity,Tally
When we group by two or more columns, it is saying "Group them so that all of those with the same col1 and col2 are in the same group, and then calculate all the aggregate functions (Count, Sum, Average, etc.) for each of those groups"
Maybe you want this...
SELECT DATEDIFF(minute,(SELECT TOP(1) FIRSTOCCURRENCE FROM
xyz.xyz.xyz_status),(SELECT TOP(1) LASTOCCURRENCE FROM
xyz.xyz.xyz_status))
FROM xyz.xyz.xyz_status
WHERE node = 'Houston';
Here you can take a look at more examples of DATEDIFF function.

Answer 2

This should put you on track although Writing reports in SQL is probably a bad idea. What I believe you're wanting to do it output. You can also look at the ROLLUP options some of which are deprecated.
with data as (
select
NODE, EVENTID, TYPE, FIRSTOCCURRENCE, LASTOCCURRENCE,
DATEDIFF(Minute, FIRSTOCCURRENCE, LASTOCCURRENCE) as OutageInMin,
TicketNumber, Tally,
ROW_NUMBER() OVER (PARTITION BY NODE ORDER BY FIRSTOCCURRENCE) as rn
from Test_Table
--WHERE FIRSTOCCURRENCE >= DATEADD(hh, -24, GETDATE())
)
select
case when grouping(rn) = 1 then 'SITE TOTAL' else NODE end as NODE,
case when grouping(rn) = 1 then null else min(EVENTID) end as EVENTID,
case when grouping(rn) = 1 then null else min(TYPE) end as TYPE,
case when grouping(rn) = 1 then null else min(FIRSTOCCURRENCE) end as FIRSTOCCURRENCE,
case when grouping(rn) = 1 then null else min(LASTOCCURRENCE) end as LASTOCCURRENCE,
case when grouping(rn) = 1 then null else min(Tally) end as Tally,
case when grouping(rn) = 1 then null else min(TicketNumber) end as TicketNumber,
case when grouping(node) = 1
then min(OutageInMin) else sum(OutageInMin) end as "Outage In MIN"
from
data
group by grouping sets ( (NODE, rn), (NODE) )
order by data.NODE, grouping(rn), rn;
http://rextester.com/DZIHJ81264

Answer 3

GROUP BY is only authorized in SQL when you are aggregating something. The easiest exemple is a count.
Example : you want to know how much EventID are linked to a given Node :
SELECT Count(EventId), node FROM xyz.xyz.xyz_status GROUP BY node;
Here is a site that present the Group By function. If you clarify what you are searching for, we'll give you a more concrete example.