rowid corruption with heavy deletes

We have experienced some strange behaviour on an HP 9000 G30 running OnLine 5.01.UD1. We have a large, wide (255K rows, 501 bytes per row) table that gets a lot of delete activity. Eventually, we get problems with accessing from the table by rowid. Explain shows it doing a sequential scan on a rowid select. The actual values of the rowids seem to be awfully high. Tbcheck shows no problems with the tables or indexes. All we can do is drop and recreate the table, then the problem goes away. I'm wondering if the heavy delete activity is causing us to overflow some limits on rowid size or something similar. Or maybe there's a problem with the index structure that tbcheck isn't seeing. Has anyone else seen this? Joe -- =========================================================================== jlumbley@netcom.com (Joe Lumbley) BancTec Service Corportation 214-450-9894 Dallas, Texas Watch for my _INFORMIX DBA SURVIVAL GUIDE_ in Fall '94 from Prentice Hall! ===========================================================================