I tagged a "REL_1_2_0_PRE1" yesterday, preparatory to doing some
testing on AIX 5.3...

Subsequently, I have made a couple of minor patches to test scripts in
order to get them to work properly.

I was able to successfully run through the entire set of tests with
AIX 5.3 + PostgreSQL 8.1.3 + Slony-I 1.2 PRE1:

  test1  - which generally runs a bunch of data thru replication
  testddl - which performs a fairly complex DDL script
  testseqnames - which generates sequences with funky names
  testutf8 - which replicates some UTF-8 data
  testschemanames - which generates some schemas with wild names
  testtabnames - generating tables with wild names

The tests of names were particularly important in that Jan applied a
couple of patches this week that substantially revised the quoting
logic.

After the recent SQL injection problem, he re-examined the quoting
functions, and wasn't entirely happy with them...  It's worth
observing that Slony-I probably didn't any issues surrounding that
security bug, as there are a *very* few places where Slony-I quotes
things, and it does so quite carefully.

It seems to me we're at the point where it makes sense to solicit test
results on as many platforms as possible in order to have as large a
set of supported platforms as possible; I'll see about templating
something under separate cover...

I have tagged REL_1_2_0_PRE2; that should be a reasonable thing to
check out and try out...
-- 
(reverse (concatenate 'string "gro.mca" "@" "enworbbc"))
http://cbbrowne.com/info/spiritual.html
Smith's Test for Artificial Life:
When animal-rights activists and right-to-life protesters are marching
outside your laboratory, then you know you've definitely made progress
in your artificial life research.  -- Donald A. Smith