Re: in place alter messing things up

On Nov 9, 12:45 pm, "Floyd Wellershaus" <fl...@fwellers.com> wrote: > Our developers are getting ready for a big upgrade tomorrow. Their script modifies tables, which is causing in place alters. No biggy right ? > But, then they are running an internal edit application which edits rows through a cursor. They are having issues where certain rows are trying to be updated twice, as if the cursor got messed up. > > The workaround for us on this, is to update all the rows in the table so that there aren't any rows using the other version of the table. > > Is this a bug ? I haven't heard or seen this behavior before. > Is there another simpler workaround I should be looking at ? ( update stats on the table doesn't fix it ). I thought of making a checkpoint , but their application has no authority to run an onmode -c > > Any thoughts or advice ? > > Thanks, > floyd I have run into problems with in place alters, but only on upgrades. We have a nibble script that gets rid of ipa rows slowly so as not to interfere with our users.