On 10/5/2004 10:14 AM, Hannu Krosing wrote:

> On T, 2004-10-05 at 16:28, Jan Wieck wrote:
>> On 10/5/2004 4:45 AM, hannu at skype.net wrote:
>> >> On 10/4/2004 8:30 PM, Brad Hilton wrote:
>> >>
>> >>> Hello,
>> >>>
>> >>> I've spent the last few days learning slony and setting it up on some
>> >>> test databases with good results.  However, when I set up replication on
>> >>> my live database, the slon processes startup and appear to be preparing
>> >>> to copy things over when I get:
>> >>>
>> >>> "The postmaster has commanded this server process to roll back the
>> >>> current transaction and exit, because another server process exited
>> >>> abnormally and possibly corrupted shared memory."
>> > 
>> > Similar things have happened to me when someone changed schema (added
>> > columns) to master table directly, bypassing slonik.
>> 
>> Not entirely sure how this could cause a crash, since slony doesn't 
>> cache those things that do change, but it is for sure possible. We'd 
>> need a stack trace or a test case to answer that question definitively.
> 
> It started happening 3-4 times a day on a quite heavily used database
> (10+ queries sec, 27/4) , so it's nothing simple and straightforward.
> 
> Crashes went away when I dropped slony node.

This does worry me a little. Slony is still a relatively young project, 
so can people please try to provide stack traces if they suspect slony 
triggers to crash backends?


Jan

> 
> ------------
> Hannu
> 


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