| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> |
| Cc: | Hiroshi Inoue <Inoue(at)tpf(dot)co(dot)jp>, Manfred Koizar <mkoi-pg(at)aon(dot)at>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: Nested transactions: low level stuff |
| Date: | 2003-03-20 03:54:14 |
| Message-ID: | 19899.1048132454@sss.pgh.pa.us |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
> Tom Lane wrote:
>> There are other problems besides WAL bloat, too.
> I assumed the UNDO would have had to be in a separate place, or allow
> compression of the WAL file to keep needed UNDO stuff but get rid of
> unneeded stuff --- it was all quite complicated.
There is a mechanism in the XLOG code to distinguish "in transaction"
from "out of transaction" WAL entries. The thing I had not realized
before working on the btree code is that that mechanism is inadequate
for UNDO.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Bruce Momjian | 2003-03-20 03:54:28 | Re: Faster NUMERIC implementation |
| Previous Message | Hiroshi Inoue | 2003-03-20 03:52:14 | Re: Nested transactions: low level stuff |