Different Behavior In 2000 And 2005 For Same Query
Jul 18, 2007
Hi there,
We are in the process of upgrading from SQL Server 2000 to 2005. During testing we came across the following situation.
To reproduce the issue you can do the following
Create this structure in a 2000 and 2005 server instances
CREATE TABLE test
(a int,
b varchar(30))
INSERT INTO test
(a, b)
VALUES (1, '2')
INSERT INTO test
(a, b)
VALUES (3, '3a')
INSERT INTO test
(a, b)
VALUES (4, '4')
Then run the following statement in both servers:
UPDATE test
SET a = ltrim(rtrim(b))
WHERE b NOT LIKE '%a%'
AND ltrim(rtrim(b)) <> a
In 2000 this last statement will execute with no problem and it will update one row, whereas in 2005 the following error message is given:
Msg 245, Level 16, State 1, Line 20
Conversion failed when converting the varchar value '3a' to data type int.
By looking at the execution plan it seems that 2005 first tries to evaluate ltrim(rtrim(b)) <> a and then excludes those rows containing a whereas 2000 first excludes those rows containing a and then evaluates the different than condition.
I know fixing this instance itself is easy but I€™m more concerned about having to rewrite many more stored procedures where we find this same scenario; is there any setting that can be changed to avoid this?
A stored procedure takes an IN parameter, an INOUT parameter, and returns an OUT parameter. When this stored procedure is defined in SQL Server 2000, the JDBC DatabaseMetadata method getProcedureColumns() returns three rows in the resultset:
one for the IN parameter (COLUMN_TYPE=1) one for the INOUT parameter (COLUMN_TYPE=2), and one for OUT parameter (COLUMN_TYPE=5).However, when the same stored procedure is defined in SQL Server 2005 (SP2), the getProcedureColumns() method returns only two rows in the resultset:
one for the IN parameter (COLUMN_TYPE=1) one for the INOUT parameter (COLUMN_TYPE=2).No row for the OUT parameter is returned.
jTDS JDBC driver was used for both of the above tests. When the Microsoft JDBC Driver for 2005 was used against SQL Server 2005, the same behavior (only two rows in the resultset) was observed.
Has someone else run into such a problem? Is this a bug in SQL Server 2005 because the same jTDS driver works as expected against SQL Server 2000 but not against SQL Server 2005? Any feedback will be appreciated.
Hi,anyone could explain this : I have installed the same asp.net application on 2 distinct hosts, each with its own SQL Server 2000 database.However, in some SQL queries in Stored Procedures, the results do not come out the same.For example, on one server the query "select @value = (select ....)-(select ...)" works and on the other one it returns a null valueAny hint ?Thanks for your helpJohann
I was able to successfully create a database maintenance plan for SQL Server 2000 Transaction Log Shipping for a few databases a few weeks ago. Yesterday, I created a few more but to my surprise, I can no longer do it. I can create a maintenance plan but the job it creates does not start even if I force the job to start. I did exactly the same thing as what I did (as I document everything I do) before but no luck.
Hi I've got a sql server 2000 database that when running is runnign fine. About 9 months ago I altered one of the stored procedures and ever since then when the machine is rebooted the stored procedure is "reverted" back to the old sproc... ??? is there any way I can recrete a sproc in a job that runs every day?? why would it be doing this?
db_TBOdb_TBT-------------------------------------------------------------------------------------TBOID | Date TBTID| TBOID| Date-------------------------------------------------------------------------------------rp01 | 01/08/2006 ap01 |rp01|02/08/2006many rows ap02 | rp01 |05/08/2006ap03 |rp03|04/08/2006I want to find TBTTD field of table db_TBTwho have of db_TBO date db_TBT table dateand TBOID should be 'rp01' of both tableswhen I give a query as it works as I neededSelect TBT.Datefrom db_TBO TBO , db_TBT TBTwhere TBT.TBOID = 'rp01'and TBO.TBOID ='rp01'and TBO.Date TBT.DateMy doubt is when I run the following query" Select TBT.Date from db_TBO TBO , db_TBT TBT where TBT.TBOID = 'rp01'"it gives me more than 7 recordswhen I run the query using some change" Select TBT.Date , TBT.TBID from db_TBO TBO , db_TBT TBT whereTBT.TBOID = 'rp01' "it gives more rows than previous rowCan u give me any explations.ThanksPai
We have an interesting problem. We are attempting to migrate from sql 2000 to sql 2005. the schema we have is exactly the same. the new 2005 box is more powerful than our 2000 box.
here is our schema:
tbl_Items ItemID int pk ReferenceID int sessionid varchar(255) StatusID int
tbl_ItemsStatus statusid int pk isinternalstatus bit
there is an index on (ReferenceID, SessionID, StatusID) and (SessionID, StatusID)
this is the query:
DECLARE @referenceid INTEGER SET @referenceid = 1019
SELECT MAX(i2.itemid) FROM tbl_Items i2 (NOLOCK) JOIN tbl_ItemsStatus s (NOLOCK) ON i2.StatusID = s.StatusID WHERE s.IsInternalStatus = 0 AND i2.referenceid = @referenceid AND i2.sessionid IN ( SELECT i3.sessionid FROM tbl_Items i3 (NOLOCK) WHERE i3.referenceid = @referenceid AND i3.status <> 7 AND i3.status <> 8 AND i3.status <> 10 AND i3.itemid IN ( SELECT max(i4.itemid) FROM tbl_Items i4 (NOLOCK) WHERE i4.referenceid = @referenceid GROUP BY i4.sessionid ) AND i3.itemid NOT IN ( SELECT MAX(i7.itemid ) FROM tbl_Items i7 (NOLOCK) WHERE i7.referenceid = @referenceid AND i7.SessionID IN ( SELECT i5.SessionID FROM tbl_Items i5 (NOLOCK) WHERE i5.status <> 11 AND i5.referenceid = @referenceid AND i5.itemid IN ( SELECT MAX(i6.itemid) FROM tbl_Items i6 (NOLOCK) WHERE i6.referenceid = @referenceid AND i6.status IN (7,11,8) GROUP BY i6.sessionid ) ) GROUP BY i7.SessionID ) )
GROUP BY i2.sessionid
we know this query is pretty bad and can be optimized. however, if we run this query as is on 2005 it takes about 2 hours to run...if we run the exact same query on 2000 it takes 9 seconds.
so this query on 2005 if run takes 2 hours..however, if we omit the s.IsInternalStatus = 0 or the i2.referenceid = @referenceid line it takes about 9 seconds.
why would this be? it makes no sense why omitting one of those where clauses would increase the performance of the query by 2 hours? we know its a bad query...but this doesnt make sense.
We are having some problems with the SQL Server 2005 Keep Alive mechanism causing dropped connections. The Keep Alive time and interval appear to work as documented. However the number of retries made before dropping the connection seems to be variable. When running Network Monitor I have seen this vary between 3 and 9 retries. The value for TcpMaxDataRetransmissions in the registry is set to 5. Does anyone know what the correct number of retries should be for SQL Server 2005 Keep Alive? Is this dynamically determined or is this a bug?
We moved a very tiny db (< 100 mb) from 2000 to 2005 (via attach/ detach) recently and now there are a couple sprocs which take forever to run from the app. Yes, the stats have since been updated as I reorg the indexes weekly. Now here is where it gets really strange, I can run this sproc in 0 seconds via Query Analyzer (QA). In other words I:
Have a developer run it from the app while monitoring in Profiler. There is a setting in the app for a 30 timeout, so after 30 seconds the app blows up and I see the exact code that the developer tried to run.
I can grab that code from Profiler, plop it into QA, and run it. Results are returned instantly. We have done this test a number of times, so I know it's not just a one time thing. How can this be? The only thing that got changed in the app was the connection string to the new 2005 server.
On a SQL 2005 Cluster, when you move the SQL resources you generally have to kind of "reauthenticate". For example, in SSMS and after you have moved the SQL cluster resources, you run a query in SSMS and you get a connection failure. Run it again and it works. Does anyone else notice this? I am assuming it is because the resources are now on a differen't server. But then what's the purpose of the SQL virtual server name and instance? I thought when you authenticate to SQL it's to the virtual SQL server like so... VirtualServerSQLInstance. Can someone straighten out my poor mangled mind? I'd like to avoid this reauthentication thing.
BEGIN declare @datefin_flag datetime, @strip datetime SELECT @strip = dateadd(d,datediff(d,0,getdate()),0) SELECT @datefin_flag = DATE_FIN_PERIODE_FISCALE FROM DM_LKP_CALENDRIER_PERIODE_F WHERE DATE_DEBUT_PERIODE_FISCALE < @strip AND DATE_FIN_PERIODE_FISCALE = @strip --select @datefin_flag --select @strip IF(@datefin_flag != @strip) RAISERROR('You cant run this',16,1) END
Well this Query should return the raiserror it returns completes successfuly since todays date is not the same as the date in the database. if you select @datefin_flag it returns NULL and if you select @strip it brings back todays date how can NULL be equal to to todays date assuming that todays date is equal to NULL. ?
Note the code below, running on the version noted.
I just found this today, figured I'd share. Not sure if it's a known bug or a "special" feature. The only difference between the two queries is the 3rd line, everything else is the same. Notice that the sort order changes, yet no errors or warnings are given. I assume that the table aliases are ignored for the order by, unless there are duplicate column names in the results.
I abstracted this from a bug I discovered in one of my apps today, where I have sales reps assigned to their clients with start and end dates. A while back, a developer asked me to format the dates without the time portion, and when I did so, I introduced the problem. I resolved it temporarily by ordering by convert(datetime, startdate) but I found it strange that the column alias match overrides the table alias attempted match in the order by. Another way to get around this would be to change the column aliases, then the sort order would be as desired, but I didn't want to have to change the app code for something so trivial.
My apologies if this is a duplicate.
@@version = Microsoft SQL Server 2005 - 9.00.3054.00 (X64) Mar 23 2007 18:41:50 Copyright (c) 1988-2005 Microsoft Corporation Standard Edition (64-bit) on Windows NT 5.2 (Build 3790: Service Pack 2)
select convert(varchar(10), table_alias.startdate, 101) as startdate,
convert(varchar(10), table_alias.enddate, 101) as enddate
from
(
select convert(datetime, dateadd(mm, -4, getdate()-1)) as startdate, convert(datetime, dateadd(mm, -4, getdate())) as enddate
union
select convert(datetime, dateadd(mm, -3, getdate()-1)) as startdate, convert(datetime, dateadd(mm, -3, getdate())) as enddate
union
select convert(datetime, dateadd(mm, -2, getdate()-1)) as startdate, convert(datetime, dateadd(mm, -2, getdate())) as enddate
union
select convert(datetime, dateadd(mm, -1, getdate()-1)) as startdate, convert(datetime, dateadd(mm, -1, getdate())) as enddate
union
select convert(datetime, getdate()-1) as startdate, convert(datetime, getdate()) as enddate
) as table_alias
order by table_alias.startdate
select convert(varchar(10), table_alias.startdate, 101) as startdate,
convert(varchar(10), table_alias.enddate, 101) as enddate,
table_alias.startdate, table_alias.enddate
from
(
select convert(datetime, dateadd(mm, -4, getdate()-1)) as startdate, convert(datetime, dateadd(mm, -4, getdate())) as enddate
union
select convert(datetime, dateadd(mm, -3, getdate()-1)) as startdate, convert(datetime, dateadd(mm, -3, getdate())) as enddate
union
select convert(datetime, dateadd(mm, -2, getdate()-1)) as startdate, convert(datetime, dateadd(mm, -2, getdate())) as enddate
union
select convert(datetime, dateadd(mm, -1, getdate()-1)) as startdate, convert(datetime, dateadd(mm, -1, getdate())) as enddate
union
select convert(datetime, getdate()-1) as startdate, convert(datetime, getdate()) as enddate
We are into a phase of retiring SQL 2000 database and replacing with SQL 2005. Could you please guide me to get a list of all the MUST DO changes on syntax, statements/query those have been changed in 2005 when compared to 2000.
I know there is a list of new features list on site, but it doesn't tells me precisely what all syntaxes will result differently then expected in 2000. For Example - if we open help for SET ANSI_NULLS ON, we can read that MS is suggesting to avoid its use as it will be absolete in later version. So do we have a list of all such things in one place that we can read and analyse our code for changes to be done now instead of finding it later.
We are currently trying to get to grips with SQL Server 2005, and have a basic question. Is it possible to user Query Analyser from SQL 2000 with 2005? All our support staff use Query Analyser on a day to ay basis and all our clients are currently on SQL Server 2000. It would be usefull if when new customers come on line with 2005 our staff can use one tool to perform their tasks
We've tried to run Query Anayser and it's comming up with a connection error, this may be down to us not setting some access up or not connecting correctly.
The following query works in SQL Server 2000 but gives follwoing error in SQL Server 2005
Msg 8114, Level 16, State 5, Line 1
Error converting data type varchar to float.
=======================
SELECT DISTINCT VWCONTACT.[ID] ID FROM VWCONTACT WHERE VWCONTACT.ACCOUNTID = 2615 AND VWCONTACT.VIRTUALDELETIONDATE IS NULL AND VWCONTACT.EMAIL LIKE '%@%' AND NOT EXISTS (SELECT VWCONTACT.ID FROM (SELECT DISTINCT VWCONTACT.[ID] ID FROM VWCONTACT JOIN (SELECT CONTACTID FROM VWCONTACTATTRIBUTEVALUE WHERE CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) >= 77002) TMP0 ON VWCONTACT.ID = TMP0.CONTACTID JOIN (SELECT CONTACTID FROM VWCONTACTATTRIBUTEVALUE WHERE CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) <= 77099) TMP1 ON VWCONTACT.ID = TMP1.CONTACTID WHERE VWCONTACT.ACCOUNTID = 2615 AND VWCONTACT.VIRTUALDELETIONDATE IS NULL ) NESTEDQUERY1 WHERE VWCONTACT.ID = NESTEDQUERY1.ID) AND NOT EXISTS (SELECT VWCONTACT.ID FROM (SELECT DISTINCT VWCONTACT.[ID] ID FROM VWCONTACT JOIN (SELECT CONTACTID FROM VWCONTACTATTRIBUTEVALUE WHERE (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77336) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77338) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77345) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77365) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77396) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77489) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77504) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77598)) ATTR ON ATTR.CONTACTID = VWCONTACT.ID WHERE VWCONTACT.ACCOUNTID = 2615 AND VWCONTACT.VIRTUALDELETIONDATE IS NULL ) NESTEDQUERY1 WHERE VWCONTACT.ID = NESTEDQUERY1.ID) AND NOT EXISTS (SELECT VWCONTACT.ID FROM (SELECT DISTINCT VWCONTACT.[ID] ID FROM VWCONTACT JOIN (SELECT CONTACTID FROM VWCONTACTATTRIBUTEVALUE WHERE CONTACTATTRIBUTEID = 23102 AND ATTRIBUTEVALUE = 'Houston') TMP0 ON VWCONTACT.ID = TMP0.CONTACTID JOIN (SELECT CONTACTID FROM VWCONTACTATTRIBUTEVALUE WHERE CONTACTATTRIBUTEID = 22944 AND ATTRIBUTEVALUE = 'TX') TMP1 ON VWCONTACT.ID = TMP1.CONTACTID WHERE VWCONTACT.ACCOUNTID = 2615 AND VWCONTACT.VIRTUALDELETIONDATE IS NULL ) NESTEDQUERY1 WHERE VWCONTACT.ID = NESTEDQUERY1.ID)
when I modify the query like the following in SQL Server 2005 it works. Now the problem is since it is adynamically generated query from our application based on users selection of criteria, it means a lot to us to change the code.
SELECT DISTINCT VWCONTACT.[ID] ID FROM VWCONTACT WHERE VWCONTACT.ACCOUNTID = 2615 AND VWCONTACT.VIRTUALDELETIONDATE IS NULL AND VWCONTACT.EMAIL LIKE '%@%'
EXCEPT
((SELECT DISTINCT VWCONTACT.[ID] ID FROM VWCONTACT JOIN (SELECT CONTACTID FROM VWCONTACTATTRIBUTEVALUE WHERE CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) >= 77002) TMP0 ON VWCONTACT.ID = TMP0.CONTACTID JOIN (SELECT CONTACTID FROM VWCONTACTATTRIBUTEVALUE WHERE CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) <= 77099) TMP1 ON VWCONTACT.ID = TMP1.CONTACTID WHERE VWCONTACT.ACCOUNTID = 2615 AND VWCONTACT.VIRTUALDELETIONDATE IS NULL ) UNION (SELECT DISTINCT VWCONTACT.[ID] ID FROM VWCONTACT JOIN (SELECT CONTACTID FROM VWCONTACTATTRIBUTEVALUE WHERE (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77336) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77338) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77345) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77365) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77396) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77489) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77504) OR (CONTACTATTRIBUTEID = 22943 AND ISNUMERIC(ATTRIBUTEVALUE) = 1 AND CONVERT(float,ATTRIBUTEVALUE) = 77598)) ATTR ON ATTR.CONTACTID = VWCONTACT.ID WHERE VWCONTACT.ACCOUNTID = 2615 AND VWCONTACT.VIRTUALDELETIONDATE IS NULL ) UNION (SELECT DISTINCT VWCONTACT.[ID] ID FROM VWCONTACT JOIN (SELECT CONTACTID FROM VWCONTACTATTRIBUTEVALUE WHERE CONTACTATTRIBUTEID = 23102 AND ATTRIBUTEVALUE = 'Houston') TMP0 ON VWCONTACT.ID = TMP0.CONTACTID JOIN (SELECT CONTACTID FROM VWCONTACTATTRIBUTEVALUE WHERE CONTACTATTRIBUTEID = 22944 AND ATTRIBUTEVALUE = 'TX') TMP1 ON VWCONTACT.ID = TMP1.CONTACTID WHERE VWCONTACT.ACCOUNTID = 2615 AND VWCONTACT.VIRTUALDELETIONDATE IS NULL ))
I found an unusual problem between 2000 and 2005 I haven't been ableto decipher from any documentation.The query structure is as follows:select *fromtableA ajointableB b ON a.somekey = b.somekeywherea.type = 'A'and datediff(yyyy, b.someDateField, getdate()) betweena.lowboundary and a.highboundarySome basic facts about the elements and data. The low and high-boundary fields are varchar datatypes. In 2005 (regardless ofcompatibility type I run the database under), the query evaluates theBETWEEN and errors out due to the fact that it is evaluating theDATEDIFF as an integer and finds a non-integer entry in eitherlowboundary or highboundary. I understand and expect this behavior.Obviously, changing the result of the DATEDIFF function to varcharallows the operation to go forth.The odd thing is that there is no "a.type = 'A' " entry, thus thequery wouldn't return anything. In 2000, it seems as though theengine is evaluating the type = 'A' and short-circuiting and in 2005,it is trying to evaluate the entire query OR is there an implicitconversion occuring in 2000 and not in 2005?As I mentioned, the compatibility mode doesn't change how this reacts,but running this on a native 2000 server allows this to happen. Thisparticular code isn't the problem, it's what we might have to contendwith when we migrate this through. Sure, we're going to performregression testing, but I'm concerned about what we would miss.Thanks for any replies.
We have an interesting problem. We are attempting to migrate from sql 2000 to sql 2005. the schema we have is exactly the same. the new 2005 box is more powerful than our 2000 box.
here is our schema:
tbl_Items ItemID int pk ReferenceID int sessionid varchar(255) StatusID int
tbl_ItemsStatus statusid int pk isinternalstatus bit
there is an index on (ReferenceID, SessionID, StatusID) and (SessionID, StatusID)
this is the query:
DECLARE @referenceid INTEGER SET @referenceid = 1019
SELECT MAX(i2.itemid) FROM tbl_Items i2 (NOLOCK) JOIN tbl_ItemsStatus s (NOLOCK) ON i2.StatusID = s.StatusID WHERE s.IsInternalStatus = 0 AND i2.referenceid = @referenceid AND i2.sessionid IN ( SELECT i3.sessionid FROM tbl_Items i3 (NOLOCK) WHERE i3.referenceid = @referenceid AND i3.status <> 7 AND i3.status <> 8 AND i3.status <> 10 AND i3.itemid IN ( SELECT max(i4.itemid) FROM tbl_Items i4 (NOLOCK) WHERE i4.referenceid = @referenceid GROUP BY i4.sessionid ) AND i3.itemid NOT IN ( SELECT MAX(i7.itemid ) FROM tbl_Items i7 (NOLOCK) WHERE i7.referenceid = @referenceid AND i7.SessionID IN ( SELECT i5.SessionID FROM tbl_Items i5 (NOLOCK) WHERE i5.status <> 11 AND i5.referenceid = @referenceid AND i5.itemid IN ( SELECT MAX(i6.itemid) FROM tbl_Items i6 (NOLOCK) WHERE i6.referenceid = @referenceid AND i6.status IN (7,11,8) GROUP BY i6.sessionid ) ) GROUP BY i7.SessionID ) )
GROUP BY i2.sessionid
we know this query is pretty bad and can be optimized. however, if we run this query as is on 2005 it takes about 2 hours to run...if we run the exact same query on 2000 it takes 9 seconds.
so this query on 2005 if run takes 2 hours..however, if we omit the s.IsInternalStatus = 0 or the i2.referenceid = @referenceid line it takes about 9 seconds.
why would this be? it makes no sense why omitting one of those where clauses would increase the performance of the query by 2 hours? we know its a bad query...but this doesnt make sense.
if have a problem with a SELECT query that works very fine on SQL Server 2000 but not on 2005. I've transfered my db by creating a full backup and restoring the db on 2005. The db is working except this problem.
When i start executing it doesn't finish. I waited a couple of minutes. On 2000 it only takes about 6 seconds to run.
Here it is:
Code Snippet
SELECT * FROM PPS_TERMbesttmpwhere PPS_TermBestTmp.BestNr + CONVERT(varchar(30),CAST(REPLACE(PPS_TermBestTmp.Pos1, ',', '.') AS float),2)
NOT IN (SELECT PPS_TermBest.BestNr + CONVERT(varchar(30), PPS_TermBest.Pos1,2) FROM PPS_TermBest) Any ideas?
I'm having an issue with a query I'm running on Sql Server 2005. It's a semi-complex query involving an in-line table function and several left outer joins which are joined on to the results of the function call. Two of the left outer joins are then qualified in a where clause of the form where table.Col is not null; the idea is that the final result set contains data that has no match in those two tables.
The problem revolves around a where clause in the function and the last left outer join (ie, one of the ones qualified with where not null). When I alter the where clause of the function to further restrict the result set the function returns, the query times shoots up from 1 second to roughly 2-3 minutes. Note that the time the function takes to complete is not affected. The difference in time is purely down to what the query does with the results the function provides. Also note that the change to the where clause provides a subset of the original data; it does not add any more data (it actually restricts the original resultset by roughly 1000 rows).
I can bring the query speed back down again by removing the last left outer join - this join takes one of the columns from the function, and joins it to a small table - 924 rows. So it appears that this particular join is the cause of the issue, but only when using the resultset generated from the modified function query.
Now, as the thread title alludes, Sql Server 2000 and 2005 handle this differently, or appear to. When I execute this same query on a Sql 2000 machine, there's no apparent time differences, and the data that is returned is as expected. Does anyone have any suggestions as to what might be causing this and how I can fix it? I could simply return the larger resultset and use managed code to filter out the rows I don't want; however, I would like to get to the bottom of this, especially if it's going to effect future queries.
I have a simple update statement that is running forever in SQL 2005 but works fine in SQL 2000. We have a new server we put SQL 2005, restored db. The table in question WEEKLYSALESHISTORY I even re-indexed all the indexes and rebuilt the stats as well. But still no luck, still running extremely long. 1 hour 20 minutes.
I'll try to give you some background on these table. Weeklysalehistory has approx 30 fields. I have 11 indesxes set up weekending date being one of them. And replication control has index on lasttrandatetime as well. So I think my indexes are fine.
/* Update WeekEnding Date for current weeks WeeklySales Records */ Update WeeklySalesHistory set weekendingdate = (SELECT LastTransDateTime from ReplicationControl where TableName = 'WEEKHST') where weekendingdate is null
Weekly sales has approx 100,000,000 rows Replication control has 631,000 (Ithink I can delete some from here to bring it down to 100 or 200 records) Although I don't think this is issue since on 2000 has same thing and works fine.
I was trying to do this within SSIS and thought that was issue. I am new so SSIS but it runs long even if I just run it as a job with this simple Update statement so I think its something with tables, etc that is wrong.
One thing on noticed if I look at the statistics in SQL Server Management studio there is a ton of stats. some being statistics on indexes which makes sense then I have a ton of hind_113_9_6 and simiiar one like this. I must have 90 or so named like this. Not sure how to check on SQL 2000 all the stats to see if they moved over from there or what. I checked a few other tables and don't have all these extra stats. Could this be causing the issue do I need to delete all these extras? Any help would be greatly appreciated.
We recently upgrade to 2005(sp). We have one report that ran fine in 2000 but leaves out data from certain columns (date related) in the results, so we chalked it up to being a non compatiable issue. So, I decided to try and switch the DB back to 2000 compatibility (in our test env) and then back to 2005. After that the report started returning the proper data. We can€™t really explain why it worked but it did. So we thought we would try it in prod (we knew it was a long shot) and it didn€™t work. So the business needs this report so we thought we would refresh the test system from prod, but now we are back to square one. I was wondering if anyone else has heard or seen anything like this. I am open to any idea€™s, no matter how crazy. J The systems are configured identically. Let me know if you need more information.
I am trying to write some admin only procedures which will collect information to one of my development server from other production and development servers.
I have created linked servers to access these other servers on the development server. This development server is SQL Server 2000 EE. Other servers which I want to access are 2000 and 2005 (vaious editions)
E.g I have another development server called PRODTEST which is SQL Server 2005 and on the development server I have created a linked server pointing to PRODTEST called TESTLINKSRV. I want to access new object catalog view (as I do not want to use sysobjects)
When I run the following query
SELECT * FROM [TESTLINKSRV].[DBNAME].[sys].[objects]
I get following error,
OLE DB error trace [Non-interface error: OLE DB provider does not contain the table: ProviderName=' TESTLINKSRV ', TableName='" DBNAME "."sys"."objects"'].
Msg 7314, Level 16, State 1, Line 1
OLE DB provider ' TESTLINKSRV ' does not contain table '"DBNAME"."sys"."objects"'. The table either does not exist or the current user does not have permissions on that table.
So I try this query
SELECT * FROM [TESTLINKSRV].[DBNAME].[sys.objects]
and I get following error
Msg 208, Level 16, State 1, Line 1
Invalid object name TESTLINKSRV.DBNAME.sys.objects'.
So bottom line is how do I access catalog views on a 2005 server from a 2000 server using linked server?
I hope someone understands what I am trying to achieve. Please let me know what is it that I am doing wrong. Thank you
We would like to install Sql 2005 Enterprise Edition (including database engine, reporting service, integration service and analysis service) as a sepearte instance on a server which already has Sql 2000 with reporting services and analysis services. We do not want to disturb the existing sql 2000 setup.
If we do that then what will happen to my earlier sql 2000 reporting service? Will it be upgraded to sql 2005 reporting service? I heard that reporting services are instance unaware application. Where will be the default reporting service database available?
if it is possible to run a distributed query against 2000 from 2005, what would the OPENDATASOURCE parameters look like? I'd like to be able to pivot without copying my older 2000 db to 2005 or using linked servers..
For reference, here's an example of a distrib query that reads excel...
ie SELECT * INTO XLImport3 FROM OPENDATASOURCE('Microsoft.Jet.OLEDB.4.0', 'Data Source=C: estxltest.xls;Extended Properties=Excel 8.0')...[Customers$]
SQL 2000: 8.00.2187 x86, 8 way 700mhz, 6GB Ram SQL 2005: 9.00.3042 IA64 2 Way Dual-Core 1.66Mhz 16 GB ram
Symptoms
Querys to the SQL 2005 box from SQL 2000 work but when the query is parameterised with non-literals (@variables) then the query run on the SQL 2005 box excludes any where clause causing the entire table to be returned to the SQL 200 box. When the query is parameterised using literal values the query is executed on SQL 2005 including the where clause.
At first I thought that the "Collation Compatible" setting was the culprit but setting this to 1 made no difference. Other SQL 2000 boxes work as expected and any queries from these using literal and non-literal parameters.
Please, any ideas?
Working
SELECT A.Column FROM linkedServer.IA.dbo.Table Where A.Column = 'value'
Not Working (correctly anyway!)
DECALRE @Value tinyint SET @Value = 22 SELECT A.Column FROM linkedServer.IA.dbo.Table Where A.Column = @value
I have a query slower in SQL Server 2005 than in SQL Server 2000. I have a database in SQL 2000, I put it on the same server, but with SQL 2005 and the query take 5 seconds instead of 0 seconds. The DB compatibility is SQL Server 2000 (I tried with 2005 and result is the same). Execution plan seems right and I tried to change some DB options without results. It is weird, when I remove left join on MandatsEx, it take 2 seconds. The view currentEmployeeLevelHistoric returns 45 000 rows and mandatsex has 0 rows.
Here is the sample:
Select ls.EmployeNiveau.pk_EmployeNiveauID as NoNiveau, IsNull(nullif(ls.Traduction.Description, ''), ls.TraductionDefaut.Description) + ' (' + ls.currentLevelHIstoric.NoNiveau + ')' As Nom, ls.currentEmployeeLevelHistoric.DebAssign As DateAssignationDebut, ls.EmployeNiveau.assignmentReason As AssignmentReason, ls.currentEmployeeLevelHistoric.FinAssign As DateAssignationFin, ls.MandatsEx.noDossier, ls.MandatsEx.pk_MandatID, '' As Period, ls.currentEmployeeLevelHistoric.NiveauPrincipal, ls.currentEmployeeLevelHistoric.pk_emplNiveauHisto_Id, ls.EmployeNiveau.fk_NiveauID_Niveaux, ls.EmployeNiveau.No_Ent_leg, IsNull(nullif(TradPere.Description, ''), TradDefautPere.Description) + ' (' + NiveauPere.NoNiveau + ')' As NomSup, ls.EmployeNiveau.No_Ent_leg + ls.EmployeNiveau.no_divisio + ls.EmployeNiveau.no_sec_eco + ls.EmployeNiveau.no_etabli + ls.EmployeNiveau.no_mat as employeeScope, case when ls.mandatExternalisation.fk_mandatID_MandatsEx is null then 2 else 1 end as ContractCategory From ls.currentEmployeeLevelHistoric Inner Join ls.EmployeNiveau on ls.currentEmployeeLevelHistoric.pk_EmployeNiveauID = ls.EmployeNiveau.pk_EmployeNiveauID Inner join ls.currentLevelHistoric On ls.EmployeNiveau.fk_NiveauID_Niveaux = ls.currentLevelHistoric.fk_niveauID_niveaux Left Outer Join ls.TraductionDefaut On ls.currentLevelHIstoric.fk_TraductionID_TraductionDefaut = ls.TraductionDefaut.pk_NoTraduction Left Outer Join ls.Traduction On ls.TraductionDefaut.pk_NoTraduction = ls.Traduction.No_Traduction and ls.Traduction.Langue = 1 Left Join ls.MandatsEx on ls.EmployeNiveau.fk_MandatID_MandatsEx = ls.MandatsEx.pk_MandatID Left Join ls.mandatExternalisation on ls.MandatsEx.pk_MandatID = ls.mandatExternalisation.fk_mandatID_MandatsEx left Join ls.Niveaux as NiveauPere on NiveauPere.niveauID = ls.currentLevelHIstoric.supId Left Outer Join ls.TraductionDefaut As TradDefautPere On NiveauPere.fk_TraductionID_TraductionDefaut = TradDefautPere.pk_NoTraduction Left Outer Join ls.Traduction As TradPere On TradDefautPere.pk_NoTraduction = TradPere.No_Traduction and TradPere.Langue = 1 Where ls.EmployeNiveau.fk_EmploID_Emplos = 345158 and (convert(varchar, ls.currentEmployeeLevelHistoric.DebAssign, 112) <= '20060802' and (ls.currentEmployeeLevelHistoric.FinAssign is null or convert(varchar, ls.currentEmployeeLevelHistoric.FinAssign, 112) >= '20060802'))
I need your help. I have 2 different databases. One of them is in the SQL Server 2000 and the other on is in the SQL Server 2005. I need to make a query out of these 2 databases. Is there any way that I can do this?
I am attempting to move some SQL 2000 databases to SQL 2005. My main production database does not seem to want to move. When I use the SQL 2005 GUI the .bak backup file is marked 'Incomplete'. When I attempt to restore the backup file I get a 'RESTORE detected an error on page (0:0) in database' message. I saw a thread in the SQL Express forum suggesting trying to restore from the T-SQL level to get the GUI out of the picture and I get the same 'error on page (0:0)' message. However when I take the same file and use SQL 2000 Enterprise Manager it restores with no problems.
Hi, I am trying to edit some data from a SQL2000-datasource in ASP.NET 2.0 and have a problem with a column that has bit-data and is used for selection. SQL2005 works fine when declaring <SelectParameters> <asp:Parameter DefaultValue="TRUE" Name="APL" Type="boolean" /> </SelectParameters>When running this code with SQL2000, there are no error-msgs, but after editing a record the "APL"-column looses its value of 1 and is set to 0. Looks like an issue with type-conversion, we've hit incompatibilities between SQL200 and 2005 with bit/boolean several times before. So, how is this done correctly with SQL2000? (I've tried setting the Type to "int16" -> err. Also setting Defval="1" gave an err) ThanksMichael
Hi,i have SQL 2000 and 2005 on same machine(with different intance names,of course), my laptop - XP with SP2. The 2005 works fine but i can'tconnect on SQL 2000. All the the SQL services are started.Any idea? Have i to reinstall 2000?Tks,Lourival
I have to merge the data from two databases, one is in SQL Server 2005 format, one is in 2000. The merged data will then reside on a SQL Server 2000 platform. Is there an easy way to do this through Management Studio or Enterprise Manager? Or will we have to export the data from the 2005 database to a flat file and import it into a new 2000 database. And then do the merge?
I am in the process of migrating from Sql Server 2000 to 2005. Part of my plan is to move some database's to 2005, but use the 2000 compatibility mode for the short term. My issue is this, our DR boxes are still on SQL Server 2000, would I still be able to use our log shipping processes? Or would I be better off in starting with migrating the DR boxes to 2005 first?
