repeated corruption in 7.11.UC2

We are having a problem with repeated corruption in our version 7.11.UC2 database. The odd thing is that the onchecks do not see the corruption but unloading and loading the table clears it up (without data loss). Simply rebuilding the indices worked once but has not in the other 4 instance of the corruption. It does not seem to be a memory corruption problem. Restarting online does not clear up the problem. I also have not been able to tie the corruption to any hardware or OS incidences. The online log gives the following messages: ========= 09:06:38 Who: Session(71, medmgr@batman, 4709, 8) Thread(94, sqlexec, 80225d7c, 1) 09:06:38 Results: Possible inconsistencies in 'workmgr_db:"medmgr".wm_elements' 09:06:38 Action: Run 'oncheck -cD 3145730' 09:06:38 See Also: /tmp/af.76814, shmem.76814.1 ========= The 'oncheck -cD 3145730' will run clean. Has anyone seen this before? This is running on HP-UX 9.07. Thanks, Scott Patterson Norand pattersons@norand.com