Alfredo writes in his essay on Jumbo Datasets: >DBSTORE does not store jumbo datasets. You have to use STORE with >Posix names (for instance, instead of saying ":store [log in to unmask]; >*dds" you would issue a command such as ":store /ACCOUNT/GROUP/DB@; >*dds" (note the upper-case for the account, group & db names; Posix >names are case sensitive). >Note for HP, Orbit & Unison/Tymlabs: In the Chronicle column, I >neglected to mention the approaches (regarding Jumbo datasets) of the >OnLine versions of your products. Could you please shed some light? >Thanks. To which ORBiT thanks Alfredo for the opening and responds: ORBiT has always supported the DBSTORE functionality as an integral part of its MPE/V and MPE/iX backup products, including the setting of the dirty bit and date/timestamp in the root file. This functionality is being enhanced in the MPE/iX product to include Jumbo Datasets, as was done for TPI files, and will be in the next release of BACKUP+/iX. At that point, users will be able to type: :store DB.GROUP.ACCOUNT;dds;dbstore and the entire database including the Jumbo Datasets will be backed up. This will further preserve the logging and recovery timestamps in the root file that DBRECOV depend upon. For the time being, users need to specify the database for store using the same syntax as for MPE :STORE, which will cause the chunk files in the HFS namespace to be stored as well. Additionally, users can specify SELECT TYPE=IMAGE to select all IMAGE files on the system, and this will pick up all chunk control files and chunk files, as it today qualifies all root files, datasets, TPI files, etc. As far as our online backup technology, there should be no special impact as far as we can tell.