Finding The Filegroups Associatted With A Table In T-SQL
Jul 20, 2005
Hi,
I want to find out in my SP the filegroup associated with a particular
table? I was not able to establish the relationship between sysobjects
and sysfilegroups table. Is there any other way to find filegroup of a
table in T_SQL?
Also lets say there are 3 dtatfiles in a filegroup 'FG1' and table
'xyz' is created on filegroup 'FG1' - Is there any way to find out the
datafile's names on which table xyz's data is scattered?
I want to change the filegroup of a table. While we can do this in enterprise mgr, I would like to accomplish this in transact sql. We have a new configuration where the filegroups are located on different disks and to take advantage of that, I want to move the heavily accessed tables onto different groups. Example: table emp needs to be moved from filegroup Primary to file group Secondary. please help...
CREATE DATABASE Dummy ON -- Primary file contains Startup information of the database PRIMARY ( NAME = PrimaryLog, FILENAME = 'D:primary.mdf', SIZE = 5MB, MAXSIZE = 500MB, FILEGROWTH = 20MB ), -- Holds The Data of LookUPTables,TPProfile,CRM. ( NAME = Data, FILENAME = 'D:Data.ndf', SIZE = 5MB, MAXSIZE = 500MB, FILEGROWTH = 20MB ), LOG ON -- Stores The Log Information used To Recover The Database ( NAME = Log, FILENAME = 'D:Log.ldf', SIZE = 5MB, MAXSIZE = 500MB, FILEGROWTH = 20MB ) Go
After this I want to create table on Data .
CREATE TABLE Sample (
No INT ,Name VARCHAR(30) , Department VARCHAR(4000) NULL
CREATE DATABASE Dummy ON -- Primary file contains Startup information of the database PRIMARY ( NAME = PrimaryLog, FILENAME = 'D:primary.mdf', SIZE = 5MB, MAXSIZE = 500MB, FILEGROWTH = 20MB ), -- Holds The Data of LookUPTables,TPProfile,CRM. ( NAME = Data, FILENAME = 'D:Data.ndf', SIZE = 5MB, MAXSIZE = 500MB, FILEGROWTH = 20MB ), LOG ON -- Stores The Log Information used To Recover The Database ( NAME = Log, FILENAME = 'D:Log.ldf', SIZE = 5MB, MAXSIZE = 500MB, FILEGROWTH = 20MB ) Go
After this I want to create table on Data .
CREATE TABLE Sample (
No INT ,Name VARCHAR(30) , Department VARCHAR(4000) NULL
Recently, we had some tables inadvertantly dropped from a 20+ GB data warehouse in the middle of one of our load cycles. In order to avoid restarting the load, we had to make restore a copy of the database from a full database backup and then manually move in the dropped tables using DTS.
This ended up being a painful process, and I am looking to avoid it by restructuring the database into filegroups so I can mimic the table restore function from 6.5.
I am looking for pros, cons, gotchas, and best practices regarding splitting a MSSQL 7 database into multiple filegroups. We would be including 1 or more tables and all of thier indexes in each filegroup. Specifically, I have the following questions:
1. All the literature I read mentions the need for doing transaction log backups and restores if I utilize filegroup backups. Is this really necessary? We have turned off transaction logging for performance reasons, and since we completely control the addition of new data into the db.
2. Are there implications for disk read performance involved with filegroups? Our data resides on an EMC disk array. I feel that we'll most likely have between 20 and 30 filegroups.
3. If we decide to move a table (and its indexes) between filegroups, is the process as simple a dropping the clustered index and recreating it on the new filegroup (then dropping and recreating the n-c indexes)?
Any information y'all can post would be greatly appreciated.
I have approximately 400 - 600 tables that need to have their filegroups changed on a new SQL 2005 installation. In 2000, I could go the table properties in Enterprise Manager and change the filegroup, but on 2005 Management Studio, I can view (but not change) the table properties.
I realize that I can create the table on the other filegroup using a new name, move my constraints and indexes to this new table, copy the data over, drop the original table, then rename the new table to the name my application is expecting. But that could take me weeks (especially with such a time-consuming process)! I don't mind the process being so slow, but I do mind it being labor intensive.
Does anyone have an idea how I can automate this process (or at least make it as easy as it was with Enterprise Manager?)
I have a script that creates these spin off tables (used for lightning fast access) on a near daily basis. But it is possible the table does not exist for that day - in which the script need to check one day back - so on and so forth until finding the most recently created table.
The script I have written fails with overflow error. I was hoping someone could tell me the correct syntax to accomplish this and also why i recieve this overflow error...
Code:
Dim str_SQL_Asset_1 daysBack = 1 tblFound = FALSE Dim xbl_Holdings_date Dim xMonth Dim xDay xbl_Holdings_date = date() If Len(Month(xbl_Holdings_date))=1 Then xMonth = "0" & Month(xbl_Holdings_date) Else xMonth = Month(xbl_Holdings_date) End If If Len(Day(xbl_Holdings_date))=1 Then xDay = "0" & Day(xbl_Holdings_date) Else xDay = Day(xbl_Holdings_date) End If
IF MM_checkCmd.EOF Then xbl_Holdings_date = (xbl_Holdings_date - daysBack) daysBack = daysBack + 1 tblFound = FALSE ELSE tblFound = TRUE MM_checkCmd.CLOSE() SET MM_checkCmd = Nothing END IF loop
The goal of this snippet is simply to identify and set the value for the existing table name. An example table name would be "xbl_Holdings_2006_12_10".
I could have posted this question in the VB forum but I would like to push as much of the work onto SQL as possible and also - this sort of operation is more likely to be required of a dba imo.
I'm trying to find the primary key on a given table in SQL Server 2000 using SQL. I'm querying the sysobjects table to find a given table, and then querying the sysindexes table. I've ALMOST found what I'm looking for. I see the indexes and columns etc. on the tables in the database, I just don't see the field that indicates that the index is the primary key. Can anyone help? Thanks, Alex
I'm fairly new to sql.I've got a database that all I want to do is view some records fromsome tables. But the thing is, I do not know the table names ortheir relationships.I've got access to the database thru query analyzer. Can someoneassist me in gettign the correct statements that will allow me to findout what tables exist in the database.
When I open a table and need to make a change within a particular record...
How do I find the particular row that I am looking for? I seem to have to scroll through 600 records to find the one I need. I've found a search and replace function but it doesn't seem to work.
Need some suggestions on what to check or look for.My stored proc is not finding the table in the sql db. This is the error...Invalid object name 'tblServRec'I use the same function to pass in an arraylist and the sp name for another sp and that one works ok.The sp is the same as another one except for the table name and that one works ok.The sp works fine on my local machine. It finds the table ok. But when I try it on the server, it doesn't work.I have checked permissions, and they are the same for all tables, even the one that the other sp works ok on the server.Here is the sp if that will help.....CREATE PROCEDURE dbo.deleteServRec @fldNum INT ASBEGIN DECLARE @errCode INT DELETE FROM tblServRec WHERE fldNum = @fldNum SET @errCode = 0 RETURN @errCode HANDLE_APPERR: SET @errCode = 1 RETURN @errCodeENDGOThanks all,Zath
Is there any way to find the time when the last DDL was happened in a table? For example: The time when the new column(s) were added into a table or changed the datatype.
Is there a way to find out when the last modification or change to data in a table occured ? How about the last change of a view ?
We'd like to be able to extract data to another table based on a view. Then, at a later date, check to see if the first table or view has changed since the last extract. If not, then we don't need to re-extract the data, we know it hasn't changed.
How the client wants me to write a query, that will find "suspected abuse of accounts" - defined by being successful logins made within one hour from at least 3 different IPs.
So I need to check for events, that are within one hour from each other (datecreated field) and have the same userID - and EventTypeId - but at least 3 different IPs.
Maybe it's just late in the day or I am overworked, but I cant seem to find an effective way to approach this.
We have an application that accesses a particular table, but it isnever updated. Without stepping through the complex application code,is there a method of logging when a particular table is accessed?
I'm trying to find out less than 10% in revenue accounts from a table. Below is a snapshot. Basically, I want to add Revenue mix column in the table using procedure.
I have a table which maps two related IDs. That table has 3 columns: ID, BHID & EPID. I need to find all of the BHIDs where the EPID is unique. It seems easy enough, but I keep going in circles..
USE [CGB] GO /****** Object: Table [dbo].[ePID_BHID] Script Date: 04/15/2015 15:48:14 ******/ SET ANSI_NULLS ON GO SET QUOTED_IDENTIFIER ON
This function will generate all DELETE statements in correct order to perform a CASCADING delete. For self-joined tables, it will generate the T-SQL code to "unwind" the table, also in correct order!CREATE FUNCTION dbo.fnCascadingDelete ( @Schema NVARCHAR(128) = NULL, @Table NVARCHAR(128) = NULL ) RETURNS@Return TABLE ( RowID INT PRIMARY KEY CLUSTERED, IsSelfJoin TINYINT NOT NULL, HasPk TINYINT NOT NULL, [SQL] NVARCHAR(4000) NOT NULL ) AS BEGIN DECLARE@Constraints TABLE ( RowID INT NOT NULL, Indent SMALLINT NOT NULL, [Catalog] NVARCHAR(128) NOT NULL, [Schema] NVARCHAR(128) NOT NULL, [Table] NVARCHAR(128) NOT NULL, [Column] NVARCHAR(128), pkCatalog NVARCHAR(128), pkSchema NVARCHAR(128), pkTable NVARCHAR(128), pkColumn NVARCHAR(128), pkType NVARCHAR(128), pkSize INT, IsSelfJoin TINYINT NOT NULL, HasPk TINYINT NOT NULL )
INSERT@Constraints ( RowID, Indent, [Catalog], [Schema], [Table], [Column], pkCatalog, pkSchema, pkTable, pkColumn, pkType, pkSize, IsSelfJoin, HasPk ) SELECTRowID, Indent, [Catalog], [Schema], [Table], [Column], pkCatalog, pkSchema, pkTable, pkColumn, pkType, pkSize, SelfJoin, CASE WHEN [Column] IS NULL THEN 0 ELSE 1 END FROMdbo.fnTableTree(@Schema, @Table)
IF @@ROWCOUNT = 0 RETURN
DECLARE@SQL TABLE ( ID INT IDENTITY(1, 1), RowID INT PRIMARY KEY CLUSTERED, IsSelfJoin TINYINT NOT NULL, HasPk TINYINT NOT NULL, [SQL] NVARCHAR(4000) NOT NULL )
DECLARE@Unwind TABLE ( RowID INT NOT NULL, StepID INT IDENTITY(0, 1) PRIMARY KEY NONCLUSTERED, [SQL] NVARCHAR(4000) )
WHILE NOT EXISTS (SELECT * FROM @SQL WHERE RowID = 1) BEGIN SELECT TOP 1@RowID = c.RowID, @ID = c.RowID, @Indent = c.Indent, @TSQL = N'', @EndSQL = N'', @IsSelfJoin = c.IsSelfjoin, @HasPk = c.HasPk FROM@Constraints AS c LEFT JOIN@SQL AS s ON s.RowID = c.RowID WHEREs.RowID IS NULL ORDER BYc.Indent DESC, c.RowID DESC
WHILE @ID > 0 BEGIN IF @Indent = 0 SELECT@RowSQL = N'DELETE t' + CAST(@RowID AS NVARCHAR(12)), @RowSQL = @RowSQL + N' FROM ' + QUOTENAME(c.[Catalog]) + N'.' + QUOTENAME(c.[Schema]) + N'.' + QUOTENAME(c.[Table]) + N' AS t' + CAST(@ID AS NVARCHAR(12)), @EndSQL = N' WHERE t' + CAST(@ID AS NVARCHAR(12)) + '.' + QUOTENAME(COALESCE(c.[Column], '%0')) + N' = ''%1''', @IsSelfJoin = @IsSelfJoin | c.IsSelfJoin FROM@Constraints AS c WHEREc.RowID = @ID ELSE SELECT@RowSQL = N' INNER JOIN ' + QUOTENAME(c.[Catalog]) + N'.' + QUOTENAME(c.[Schema]) + N'.' + QUOTENAME(c.[Table]), @RowSQL = @RowSQL + N' AS t' + CAST(@ID AS NVARCHAR(12)) + N' ON t' + CAST(@ID AS NVARCHAR(12)) + N'.' + QUOTENAME(c.[Column]), @pkColumn = QUOTENAME(c.pkColumn), @IsSelfJoin = @IsSelfJoin | c.IsSelfJoin FROM@Constraints AS c WHEREc.RowID = @ID
SELECT TOP 1@ID = c.RowID, @Indent = c.Indent, @RowSQL = @RowSQL + N' = t' + CAST(c.RowID AS NVARCHAR(12)) + N'.' + @pkColumn, @IsSelfJoin = @IsSelfJoin | c.IsSelfJoin FROM@Constraints AS c WHEREc.RowID < @ID AND c.Indent < @Indent ORDER BYc.Indent DESC, c.RowID DESC
INSERT@Stage (Lvl, RowKey) SELECT@Lvl, t.' + QUOTENAME(@pkColumn) + ' FROM' + QUOTENAME(@Catalog) + '.' + QUOTENAME(@Schema) + '.' + QUOTENAME(@Table) + ' AS t INNER JOIN@Stage AS s ON s.RowKey = t.' + QUOTENAME(@Column) + ' AND s.Lvl = @Lvl - 1 LEFT JOIN@Stage AS cr ON cr.RowKey = t.' + QUOTENAME(@pkColumn) + ' WHEREcr.RowKey IS NULL END SELECT ''DELETE FROM ' + QUOTENAME(@Catalog) + '.' + QUOTENAME(@Schema) + '.' + QUOTENAME(@Table) + ' WHERE ' + QUOTENAME(@pkColumn) + ' = '' + QUOTENAME(RowKey, '''''''') FROM @Stage WHERE RowID > 0 ORDER BY RowID DESC'
INSERT@Unwind ( RowID, [SQL] ) VALUES( @RowID, @RowSQL ) END END
INSERT@Return ( RowID, IsSelfJoin, HasPk, [SQL] ) SELECTs.ID, s.IsSelfJoin, s.HasPk, CASE WHEN u.RowID IS NULL THEN s.[SQL] ELSE u.[SQL] END FROM@SQL AS s LEFT JOIN@Unwind AS u ON u.RowID = s.RowID ORDER BYs.ID, u.StepID
