Re: why aren't ORA-1555 errors MORE frequent?

From: Richard Kuhler <noone_at_nowhere.com>
Date: Tue, 19 Nov 2002 18:55:22 GMT
Message-ID: <uOvC9.14875$2z1.6153644@twister.socal.rr.com>

Richard Foote wrote:

<snip>

> ... Oracle takes note of the latest SCN that it
> could reconstruct in the header (lets say 1100). Therefore, we know *for
> sure* as the query SCN snapshot is 1234, all changes made to the block
> *must* have been committed prior to 1100. We don't know when the change was
> committed, but it must have been committed prior to 1100.

Thanks, that's the piece of information I was missing.

<snip>

> However, if the query started again at 1234 but there are heaps and heaps of
> cleanouts occurring on this table during the running of the query resulting
> in heaps and heaps of changes to the RBS then it could be these *current*
> cleanouts that are causing the problems.

Interesting. You're saying that the cleanouts themselves have to be written to a RBS? I can't envision any reason for this. Why is this done? This also seems to imply that a pure SELECT could run out of rollback, right?

Thanks,
Richard Kuhler Received on Tue Nov 19 2002 - 12:55:22 CST