Slony1 PostgreSQL Replication - 1.2 Release schedule

Author

1.2 Release schedule

Dave Page

2005-09-25, 8:24 pm

Hi guys,

I was wondering if there was a plan for version 1.2, or is it just a
case of hacking until the new feature list looks long enough? :-)

The reason I ask is because it would be good (well, for some of us at
least!) to get an official release out supporting Windows. I've already
been getting people asking for a build to run, and of course have turned
them down in case they end up with any incompatibilities in the snapshot
I might have given them and the eventual release version.

Regards, Dave.

Christopher Browne

2005-09-25, 8:24 pm

"Dave Page" <dpage- 6fRByWcp2Fwg5PB3cbIC
Q7VCufUGDwFn@public.gmane.org> writes:
> I was wondering if there was a plan for version 1.2, or is it just a
> case of hacking until the new feature list looks long enough? :-)
>
> The reason I ask is because it would be good (well, for some of us
> at least!) to get an official release out supporting Windows. I've
> already been getting people asking for a build to run, and of course
> have turned them down in case they end up with any incompatibilities
> in the snapshot I might have given them and the eventual release
> version.

I'd like to see Darcy's "new test bed" in, as well as something of a
set of tests based on that.

I'd also like to see his "fast COPY" in place; that would be a very
worthwhile performance improvement.

I'd somewhat like to get my "rotating between sl_log_1 and sl_log_2"
patch in; that has been proving troublesome on the SPI function side
of things. (I'm suspicious that I'm getting some sort of "off by 1"
pointer problem there.)

There are a couple of strategies I have been looking at to cut down on
pg_listener "dead tuples" which are probably worth trying further.

There is a patch based on my work to improve the "big sl_log_1 query"
which is pretty near readiness, assuming Hannu's revisions are OK.

If we could get about 4 of those, along with the changes already in to
address Windows support, "holding nodes behind", and such, that's
pretty worthy of a 1.2 release. The "big work" is, frankly, to get
the "new test bed" in and to construct tests for various known
troublesome conditions so that this test bed essentially replaces
what's in the "ducttape" directory.
--
output = reverse("ofni.sailifa.ac" "@" "enworbbc")
<http://dev6.int.libertyrms.com/>
Christopher Browne
(416) 673-4124 (land)

Dave Page

2005-09-26, 3:24 am

=20

> -----Original Message-----
> From: Christopher Browne [mailto:cbbrowne-swQf4SbcV9C7WVzo/KQ3Mw@public.gmane.org]=20
> Sent: 25 September 2005 21:41
> To: Dave Page
> Cc: Slony Mailing List
> Subject: Re: [Slony1-general] 1.2 Release schedule
>=20
> "Dave Page" <dpage- 6fRByWcp2Fwg5PB3cbIC
Q7VCufUGDwFn@public.gmane.org> writes:
>=20
> I'd like to see Darcy's "new test bed" in, as well as something of a
> set of tests based on that.

I was expecting to have seen that already - unless he's had a major
rethink, it seemed pretty much complete (well, the framework at least)
when I was helping him out with it a while back.

> I'd also like to see his "fast COPY" in place; that would be a very
> worthwhile performance improvement.

OK.

> I'd somewhat like to get my "rotating between sl_log_1 and sl_log_2"
> patch in; that has been proving troublesome on the SPI function side
> of things. (I'm suspicious that I'm getting some sort of "off by 1"
> pointer problem there.)

OK.

> There are a couple of strategies I have been looking at to cut down on
> pg_listener "dead tuples" which are probably worth trying further.

OK.

> There is a patch based on my work to improve the "big sl_log_1 query"
> which is pretty near readiness, assuming Hannu's revisions are OK.

OK.
=20
> If we could get about 4 of those, along with the changes already in to
> address Windows support, "holding nodes behind", and such, that's
> pretty worthy of a 1.2 release. The "big work" is, frankly, to get
> the "new test bed" in and to construct tests for various known
> troublesome conditions so that this test bed essentially replaces
> what's in the "ducttape" directory.

OK, thanks. It's sounds like it's going to be later rather than sooner
at the moment - which is a relief in some ways if I'm honest!!.

Regards, Dave.

Sponsored Links