Jan Wieck wrote:
> On 11/8/2006 2:17 PM, Christopher Browne wrote:
>> Jan Wieck wrote:
>>> On 10/31/2006 6:03 PM, Christopher Browne wrote:
>>>> In the 1.1 branch, a number of minor bugs have been dealt with; in
>>>> both branches, the following more significant changes have been made:
>>>
>>> These release candidates have been out for over a week now and there
>>> has been literally zero feedback. Is ANYONE at all testing these at
>>> all?
>> There has been a bunch of feedback on the altperl tools, patches for RPM
>> spec files, fixes to version numbers.  Someone pointed out that one of
>> the scripts in HEAD hadn't gotten into 1_2_STABLE; I fixed that.
>>
>> Admittedly, no comments on the MERGE SET change.
>
> Maybe I wasn't too clear. I didn't mean the tools and gadgets around
> Slony, I meant how the replication engine itself works, what people
> found when testing their failover an other maintenance procedures. How
> are we supposed to tell folks that upgrading from 1.1 to 1.2 is a good
> thing for production systems, if nobody has done that on their test
> systems?
Actually, consider that part verified.

pgdba at ydb2.int.libertyrms.com:/tmp/slony-regress.hvvhcgylh $ cat
slonik.script
CLUSTER NAME = slony_regress1;
NODE 1 ADMIN CONNINFO = 'dbname=slonyregress1 host=localhost user=pgdba
port=9472';
NODE 2 ADMIN CONNINFO = 'dbname=slonyregress2 host=localhost user=pgdba
port=9472';
update functions (id=1);
update functions (id=2);

I just updated a 1.1.6 system to 1.2.1, as shown above, and all is
working fine.

It's a test instance, as opposed to anything in production, but there
you go...