On Friday 06 January 2006 13:00, Jan Wieck wrote:
> On 1/5/2006 10:36 AM, Christopher Browne wrote:
> > Matt Michalowski wrote:
> >>Christopher Browne wrote:
> >>>Jan Wieck wrote:
> >>>>On 1/4/2006 2:41 PM, David Parker wrote:
> >>>>>Any ETA on the official 1.1.5 release?
> >>>>
> >>>>I have been looking into fixing the rebuild_listen issues, but all I
> >>>> can come up with is a complete rewrite from scratch, and that is a bit
> >>>> too much of a change for a STABLE bugfix release. So I am inclined to
> >>>> not fix it in 1.1.x and go with what we have.
> >>>
> >>>If that be the case, then we should see if there is any further cleaning
> >>>up of docs or anything of the sort to be done, and proceed towards the
> >>>release.
> >>
> >>I'm playing with 1.1.5RC2 now, and Log Shipping mode seems broken (more
> >>so than it was in 1.1.2)
> >>
> >>In the subscriber logs, you end up with queries like:
> >>select "_cluster".finishTableAfterCopy(1); analyze "public"."table";
> >>
> >>Normally, finishTableAfterCopy doesn't exist in a log shipped node. To
> >>work around it, I dumped my slony schema from my subscriber node into my
> >>log shipped node, and manually updated tab_reloid in sl_table.
> >>
> >>Is log shipping deemed too experimental anyway to worry about for
> >> release?
> >
> > Jan added finishTableAfterCopy() here...
> > http://gborg.postgresql.org/pipermail/slony1-commit/2005-November/000813.
> >html
> >
> > That obviously neglected to add the function to the log shipping
> > functionality.  I'm rectifying that right now, adding it to slon_dump.sh.
>
> So it got broken in the 1.1.5 development cycle and not noticed until we
> actually have RC2? That's uncool, I don't know but for some reason I
> feel like nobody is actually doing any testing at all until after a new
> release is available.

Notes as to testing logshipping as part of the standard test plan have been 
added to the wiki...


>
>
> Jan

-- 
Darcy Buskermolen
Wavefire Technologies Corp.

http://www.wavefire.com
ph: 250.717.0200
fx: 250.763.1759