Hi all,
I have problem: i need mass delete from table. Next operation must be reorg (can be online only) and statistics capture using runstats. Problem is, that online reorg works asynchronous. How I can ensure, that runstat will run on reorganized table ?
thanks

Assuming that you are on z/OS, you can collect statistics during reorg itself. Or do you use DB2 on a different platform? Which version of DB2?

stolze: i have db2 udb LUW on linux, 8.2.3

In DB2 LUW, the DB2 CLP will be blocked until the REORG is finished. REORG is not asynchronous in this case. The same applies to the invocation of the reorg API. So that wouldn't be a problem at all.

The question is now how you start REORG in your environment and why you think it is asynchrounous.

stolze: thanks for help
when i run reorg offline, CLP is blocked, as you wroted. but when I use keyword "inplace" then
is reorg executed asynchronous.
Next cmd I like runstats. How to ensure that table in this time will be reorganized ?


db2 => connect to mydb

Database Connection Information

Database server = DB2/LINUX 8.2.3
SQL authorization ID = DB2INST1
Local database alias = MYDB

db2 => reorg table log.task
DB20000I The REORG command completed successfully

db2 => reorg table log.task inplace allow write access start
DB20000I The REORG command completed successfully.
DB21024I This command is asynchronous and may not be effective immediately.

db2 =>

first command is done in few second, second command immediate.
some ideas?
thanks

Now we are finally getting to the point where we know what you are really doing...

In your case, you will have to poll the system regularly to check if the REORG is completed. After that, you con start the RUNSTATS. That's your only option - unless you can resort to a REORG that doesn't use the INPLACE option - do you really need that?

system is 7x24... what i can use instead INPLACE option?

INPLACE means that the reorg cannot use a shadow copy to do the reorg on. I guess you want to use the option ALLOW READ/WRITE ACCESS instead.