On Tue, 2007-04-03 at 15:04 +0100, Heikki Linnakangas wrote:
> Simon Riggs wrote:
> >> Here's the next version (v10) of the patch, ready for review.
Thanks for reviewing.
> >> I've struggled with what to call all of the new concepts inherent in
> >> this patch, but I think I've got something now. COMMIT NOWAIT doesn't
> >> describe this feature, since there is no command of that name in the
> >> implementation that we've agreed. So what's it called?
>
> Hmm. Deferred commit? Asynchronous commit?
I'll change everything to whatever we agree. Other ideas?
> >> wal_writer_delay = 0 (default) ms enables this feature at server start.
> >> Once enabled, individual sessions or transactions may request
> >> transaction_guarantee = off, or it may be set for the whole server.
>
> This error message talks about commit_fsync_delay:
> ERROR: cannot set transaction guarantee when server commit_fsync_delay = 0
>
> Should be wal_writer_delay, I suppose.
Thanks, will change.
> The HeapTupleSetVisibilityInfo function with it's action-argument looks
> funny to me.
Lets see what others think as well.
> You don't need to check if a transaction has been flushed when setting
> the HEAP_XMAX/XMIN_INVALID flags for aborted transactions.
I was after as little change as possible in the code, which is why I
left that as it was. But I think you're right and that's a one line
change to avoid that is possible. Will do.
--
Simon Riggs
EnterpriseDB http://www.enterprisedb.com