Run into something like this before. Once the /opt directory went nuts, TAC called it a database error had to rebuild/restore from backup. Check during the day to see if the used space rises dramatically.
Another time the VM volume was full. VM assigns the disk space but does not actually fence it off. The volume was close to full and when Prime went to write more data V center threw an error.