Re: Build duplicate databases

Posted in 2003

Topics: Backup & Restore, Storage & Space Management, Migration, Import/Export & Data Conversion

ART KAGEL, BLOOMBERG/ 65E 55TH — Wed, 10 Dec 2003 15:14:38 — source: IIUG Forums & Mailing Lists

----- Original Message ----- From: Carol Dennis <c.dennis@nib.com.au> At: 12/ 9 23:52 > Is it possible to build 2 databases exactly the same on the one server but give > them different names and restore them with ONBAR Yes to building identical databases with different names on one server. No to restoring with onbar. IDS archive/restore utilities do not operate at the database level and even if they did would not restore one database to a different name. Use dbexport/dbimport, the HPLoader (onpload), or my dbexport/dbimport replacement package myexport (available from the IIUG Software Repository) which can be run using Jonathan Leffler's sqlcmd package for the unload/reload or using onpload. Myexport includes a feature to remap the dbspaces into which tables/fragments/indexes are loaded if they are not all loaded into the database's root. Art S. Kagel

↪ replying to ART KAGEL, BLOOMBERG/ 65E 55TH

Support Informix — Mon, 15 Dec 2003 08:07:14 — source: IIUG Forums & Mailing Lists

Yes, you can do it with onbar. Someone from IBM told me the trick for Unix. Use only relative pathnames for the chunks. Even for the ROOTDBS in onconfig. From the current directory you can use files (awfull) as chunk or symbolic links to disk partitions. When you start the engine with oninit, informix looks for the primary chunk in current directory. The same way it does with onbar backup and onbar restore. so for the restore you have to be in another directory, with the same symbolic link names pointing to other partitions. You have to change the servername, the servernum and the services in /etc/services and sqlhosts BTW : This was our method to test HDR on exactly the same machine. Yves -----Original Message----- From: nobody@ace.iiug.org [mailto:nobody@ace.iiug.org] Sent: 10 December 2003 16:15 To: ids@iiug.org Subject: Re: Build duplicate databases [2334] ----- Original Message ----- From: Carol Dennis <c.dennis@nib.com.au> At: 12/ 9 23:52 > Is it possible to build 2 databases exactly the same on the one server but give > them different names and restore them with ONBAR Yes to building identical databases with different names on one server. No to restoring with onbar. IDS archive/restore utilities do not operate at the database level and even if they did would not restore one database to a different name. Use dbexport/dbimport, the HPLoader (onpload), or my dbexport/dbimport replacement package myexport (available from the IIUG Software Repository) which can be run using Jonathan Leffler's sqlcmd package for the unload/reload or using onpload. Myexport includes a feature to remap the dbspaces into which tables/fragments/indexes are loaded if they are not all loaded into the database's root. Art S. Kagel ------_=_NextPart_001_01C3C2D8.CD27FBE0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> <HTML> <HEAD> <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; = charset=3Diso-8859-1"> <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version = 5.5.2653.12"> <TITLE>RE: Build duplicate databases [2334] </TITLE> </HEAD> <BODY> Yes, you can do it with onbar. Someone from IBM told me the trick for Unix. Use only relative pathnames for the chunks. Even for the ROOTDBS in onconfig. From the current directory you can use files = (awfull) as chunk or symbolic links to disk partitions. When you start the engine with oninit, informix = looks for the primary chunk in current directory. The same way it does with  onbar backup and = onbar restore. so for the restore you have to be in another = directory, with the same symbolic link names pointing to other = partitions. You have to change the servername, the servernum and = the services in /etc/services and sqlhosts BTW : This was our method to test HDR on exactly the = same machine. Yves -----Original Message----- From: nobody@ace.iiug.org [<A = HREF=3D"mailto:nobody@ace.iiug.org">mailto:nobody@ace.iiug.org</A>]</FON= T> Sent: 10 December 2003 16:15 To: ids@iiug.org Subject: Re: Build duplicate databases [2334] = ----- Original Message ----- From: Carol Dennis  = <c.dennis@nib.com.au> At: 12/ 9 23:52 > Is it possible to build 2 databases exactly the = same on the one server but give > them different names and restore them with = ONBAR Yes to building identical databases with different = names on one server.  No to restoring with onbar.  IDS archive/restore = utilities do not operate at the database level and even if they did would not = restore one database to a different name.  Use dbexport/dbimport, the = HPLoader (onpload), or my dbexport/dbimport replacement package myexport = (available from the IIUG Software  Repository) which can be run using Jonathan = Leffler's sqlcmd package for the unload/reload or using onpload.  Myexport = includes a feature to remap the dbspaces into which tables/fragments/indexes are = loaded if they are not all loaded into the database's root. Art S. Kagel </BODY> </HTML> ------_=_NextPart_001_01C3C2D8.CD27FBE0--

↪ replying to Support Informix

bjleb@harland.net — Mon, 15 Dec 2003 15:10:12 — source: IIUG Forums & Mailing Lists

We've been using relative paths for years. It can be done, but you obviously have to first rebuild your source instance to use relative path. You can then use onbar or ontape to restore an "entire instance" on the same server. However, keep in mind what Art said. This is at the instance level, not the database. If you have multiple databases in the instance, you can't pick and choose. The other option is informix 9.4. You can restore instances with the file rename option on the same server. This is the equivalent of a redirected restore in DB2. Barry Leb Senior DBA - Informix/DB2 John H. Harland bjleb@harland.net 770.593.5050 x6137 "Support Inf...." To: ids@iiug.org <supinformix@dof cc: i.fgov.be> Subject: RE: Build duplicate databases [2353] Sent by: forum.subscriber @iiug.org 12/15/2003 03:07 AM Yes, you can do it with onbar. Someone from IBM told me the trick for Unix. Use only relative pathnames for the chunks. Even for the ROOTDBS in onconfig. From the current directory you can use files (awfull) as chunk or symbolic links to disk partitions. When you start the engine with oninit, informix looks for the primary chunk in current directory. The same way it does with onbar backup and onbar restore. so for the restore you have to be in another directory, with the same symbolic link names pointing to other partitions. You have to change the servername, the servernum and the services in /etc/services and sqlhosts BTW : This was our method to test HDR on exactly the same machine. Yves -----Original Message----- From: nobody@ace.iiug.org [mailto:nobody@ace.iiug.org] Sent: 10 December 2003 16:15 To: ids@iiug.org Subject: Re: Build duplicate databases [2334] ----- Original Message ----- From: Carol Dennis <c.dennis@nib.com.au> At: 12/ 9 23:52 > Is it possible to build 2 databases exactly the same on the one server but give > them different names and restore them with ONBAR Yes to building identical databases with different names on one server. No to restoring with onbar. IDS archive/restore utilities do not operate at the database level and even if they did would not restore one database to a different name. Use dbexport/dbimport, the HPLoader (onpload), or my dbexport/dbimport replacement package myexport (available from the IIUG Software Repository) which can be run using Jonathan Leffler's sqlcmd package for the unload/reload or using onpload. Myexport includes a feature to remap the dbspaces into which tables/fragments/indexes are loaded if they are not all loaded into the database's root. Art S. Kagel ------_=_NextPart_001_01C3C2D8.CD27FBE0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> <HTML> <HEAD> <META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; = charset=3Diso-8859-1"> <META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version = 5.5.2653.12"> <TITLE>RE: Build duplicate databases [2334] </TITLE> </HEAD> <BODY> Yes, you can do it with onbar. Someone from IBM told me the trick for Unix. Use only relative pathnames for the chunks. Even for the ROOTDBS in onconfig. From the current directory you can use files = (awfull) as chunk or symbolic links to disk partitions. When you start the engine with oninit, informix = looks for the primary chunk in current directory. The same way it does with  onbar backup and = onbar restore. so for the restore you have to be in another = directory, with the same symbolic link names pointing to other = partitions. You have to change the servername, the servernum and = the services in /etc/services and sqlhosts BTW : This was our method to test HDR on exactly the = same machine. Yves -----Original Message----- From: nobody@ace.iiug.org [<A = HREF=3D"mailto:nobody@ace.iiug.org">mailto:nobody@ace.iiug.org</A>]</FON= T> Sent: 10 December 2003 16:15 To: ids@iiug.org Subject: Re: Build duplicate databases [2334] = ----- Original Message ----- From: Carol Dennis  = <c.dennis@nib.com.au> At: 12/ 9 23:52 > Is it possible to build 2 databases exactly the = same on the one server but give > them different names and restore them with = ONBAR Yes to building identical databases with different = names on one server.  No to restoring with onbar.  IDS archive/restore = utilities do not operate at the database level and even if they did would not = restore one database to a different name.  Use dbexport/dbimport, the = HPLoader (onpload), or my dbexport/dbimport replacement package myexport = (available from the IIUG Software  Repository) which can be run using Jonathan = Leffler's sqlcmd package for the unload/reload or using onpload.  Myexport = includes a feature to remap the dbspaces into which tables/fragments/indexes are = loaded if they are not all loaded into the database's root. Art S. Kagel </BODY> </HTML> ------_=_NextPart_001_01C3C2D8.CD27FBE0-- This message and any attachments have been scanned by Symantec Anti-Virus prior to leaving the originator's network.

Re: Build duplicate databases

Related threads

Glossary

Your privacy choices