From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Zeugswetter Andreas SB <ZeugswetterA(at)wien(dot)spardat(dot)at> |
Cc: | Vadim Mikheev <vmikheev(at)sectorbase(dot)com>, pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: AW: AW: Re: Backup and Recovery |
Date: | 2001-07-06 14:20:48 |
Message-ID: | 9920.994429248@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Zeugswetter Andreas SB <ZeugswetterA(at)wien(dot)spardat(dot)at> writes:
>> Wouldn't it be the same as the case where we *do* have UNDO? How is a
>> removed tuple different from a tuple that was never there?
> HiHi, the problem is a subtile one. What if a previously aborted txn
> produced a btree page split, that would otherwise not have happened ?
Good point. We'd have to recognize btree splits (and possibly some
other operations) as things that must be done anyway, even if their
originating transaction is aborted.
There already is a mechanism for doing that: xlog entries can be written
without any transaction identifier (see XLOG_NO_TRAN). Seems to me that
btree split XLOG records should be getting written that way now --- Vadim,
don't you agree?
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Trewern, Ben | 2001-07-06 14:25:49 | Vacuum and Transactions |
Previous Message | Matthew Hagerty | 2001-07-06 14:18:50 | Re: Proper use of select() parameter nfds? |