R
Rob Fitzpatrick
Guest
> But recids of those locks are NOT dbkey of index blocks (they are not devided by 32 or 64). That's right, they are RM blocks. My point was that the recid is valid for the table that has that index; there is a correlation between the two values. They aren't just random garbage values. The confusion is about index numbers appearing in the _lock-table column. What are these lock table entries? Are they valid records? Are they artifacts of dirty reads?
Continue reading...
Continue reading...