Restore Individual Database

We have a customer with a number (>100) of databases within the same IDS V10 instance who run a different end of day process at a different time on each database and in some cases want to be able to restore the database to its state prior to the end of day run. Can anybody think of a way to do this ? The problem is that however you restore an individual dbspace (assuming that we put each database in its own dbspace) you must always restore the logs, which would re-do the end of day process.

dlinthwaite@lintel.co.uk said: > We have a customer with a number (>100) of databases within the same > IDS V10 instance who run a different end of day process at a different > time on each database and in some cases want to be able to restore the > database to its state prior to the end of day run. Can anybody think > of a way to do this ? The problem is that however you restore an > individual dbspace (assuming that we put each database in its own > dbspace) you must always restore the logs, which would re-do the end of > day process. dbexport? And you don't *have* to restore the logs, although I accept no responsibility for the consequences thereof. :o) -- Bye now, Obnoxio "... no bill is required as no value was provided." -- Christine Normile

\

Found the link to the documentation. See below. > From the keyboard of dlinthwaite@lintel.co.uk: > >> We have a customer with a number (>100) of databases within the >> same IDS V10 instance who run a different end of day process at >> a different time on each database and in some cases want to be >> able to restore the database to its state prior to the end of >> day run. Can anybody think of a way to do this ? The problem >> is that however you restore an individual dbspace (assuming that >> we put each database in its own dbspace) you must always restore >> the logs, which would re-do the end of day process. > > With IDS V10 and not too many tables in an individual database, > TLPITR (Table Level Point In Time Restore) might be an option. http://publib.boulder.ibm.com/infocenter/idshelp/v10/topic/com.ibm.b ar.doc/barmst335.htm or http://tinyurl.com/z5yxy Regards Christian -- #include <std_disclaimer.h> /* The opinions stated above are my own and not necessarily those of my employer. */

dlinthwaite@lintel.co.uk wrote: > We have a customer with a number (>100) of databases within the same > IDS V10 instance who run a different end of day process at a different > time on each database and in some cases want to be able to restore the > database to its state prior to the end of day run. Can anybody think > of a way to do this ? The problem is that however you restore an > individual dbspace (assuming that we put each database in its own > dbspace) you must always restore the logs, which would re-do the end of > day process. > Several options: 1- With onbar you can do a point-in-time restore rolling forward the logical logs only to the point you want. 2- With ontape, if you take a full level 0 archive between each data run and the databases are restricted to independent sets of dbspaces then you can do dbspace level restores. You do not have to restore the logical logs if you are comfortable that there were no transactions following the archive affecting those dbspaces that you want to recover. 3- Alternatively, if you save logical logs to disk using the ALARMPROGRAM there will normally be only one logical log per file, so you can control which logical log files you restore. Then you can take ONE level 0 archive before ALL processing and just change logical logs before starting to process each database. Then to recover one you do dbspace level restore rolling forward only the logical log files up until the one saved prior to that database's run. 4- Take archives between database runs and use archecker to do table level restores. Art S. Kagel