Hi, we discovered a strange phenomenon with MaxDB 7.5.00.30 on Linux (RHEE 4). Consider these two queries:
SELECT TA1.abortcode AS COL1, TA1.description AS COL2, SUM(TA2.refcount) AS COL3 FROM ag_abortcode TA1, v_ag_hourlymaster TA2 WHERE TA1.abortcodeid=TA2.abortcodeid AND ( TA2.applianceid=1 ) GROUP BY TA1.abortcode, TA1.description SELECT count(*) AS COL3, SUM(TA2.refcount) FROM ag_abortcode TA1, v_ag_hourlymaster TA2 WHERE TA1.abortcodeid=TA2.abortcodeid AND ( TA2.applianceid=1 ) Same tables, same join condition, same filter criteria and since the grouping just returns one row in the first case comparing of results is easy. Now here's the strange thing: the second query always returns the same (correct) result for SUM(): 4502345. The first query returns 4436366 as well as 2921189 and probably other values, too, if we try longer - without any change to the data! Earlier 7.5 versions on Windows do not exhibit this behavior. Our tests have shown that as soon as we add a group by clause with two (!) columns the results are, um, unpredictable. Did anybody see something like this before? I didn't find anything similar in the bug database... We'll try to come up with a smaller test case (several million rows are involved) but I thought I might ask in the meantime whether this is a known issue. Kind regards robert -- MaxDB Discussion Mailing List For list archives: http://lists.mysql.com/maxdb To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
