On 10/6/2004 1:42 PM, Rod Taylor wrote:

> On Wed, 2004-10-06 at 13:30, Jan Wieck wrote:
>> On 10/6/2004 1:01 PM, Brad Hilton wrote:
>> > Jan Wieck wrote:
>> >>> However, this is a serious bug since user defined tyes without any 
>> >>> operators are of course legal in Postgres and have to be supported by 
>> >>> Slony. I am working on a fix for it.
>> >> 
>> >> This is fixed in REL_1_0_STABLE and HEAD, so it will be included in the 
>> >> upcoming version 1.0.3.
>> > 
>> > Thank you - it does seem to have fixed the crashes! :)
>> > 
>> > I'm now seeing a new issue when the initial copy starts up.  Slon logs 
>> > this on my slave database:
>> 
>> Oh crap ... well, so much for using truncate instead of delete at all. I 
>> will back out that attempt :-/
> 
> Out of curiosity, how do you deal with a non cascading delete, like
> RESTRICT?

By doing some ugly catalog modifications before screwing with the data. 
In short all the triggers and rules of a replicated table are hidden 
while the table is subscribed (on the slave only).

> 
> Part #2, if you deal with the above, perhaps a cascading truncate would
> solve the issue?

There is a "trucate table foo CASCADE" ?


Jan

-- 
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me.                                  #
#================================================== JanWieck at Yahoo.com #