Interestingly: Another cluster configuration from an 8.1.4 -> 8.2.4 =

server seems to be working just fine. The main difference is that this =

second cluster configuration was created/installed from scratch.

My plan now is simply to drop the slon schema on all nodes and recreate =

the whole thing.

I'll let you know how that goes. If you have any other suggestions on =

how to get to the root cause of the problem described below I would be =

happy to try them out.

Regards
Jason

Andrew Sullivan wrote:
> On Sat, May 12, 2007 at 06:38:25AM -0700, Jason L. Buberel wrote:
>   =

>> May 12 00:42:53 srv3 slon[8698]: [54-1] 2007-05-12 00:42:53 CDT FATAL  =

>> syncThread: "select =

>> "_srv1_srv2_cluster".createEvent('_srv1_srv2_cluster', 'SYNC', NULL);" -
>>     =

>
> It looks like the createEvent() function is blowing up. . . =

>
>   =

>> May 12 00:42:53 srv3 slon[8698]: [54-2]  server closed the connection =

>> unexpectedly
>> May 12 00:42:53 srv3 slon[8698]: [54-3]         This probably means the =

>> server terminated abnormally
>> May 12 00:42:53 srv3 slon[8698]: [54-4]         before or while =

>> processing the request.
>> May 12 00:42:53 srv3 postgres[7320]: [82-1] LOG:  server process (PID =

>> 8714) was terminated by signal 11
>>     =

>
> with sig 11.  You could look at the corefile that oughta be dumped. =

> But my guess is that the slony functions haven't been compiled
> correctly here.
>
> A
>
>   =

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.slony.info/pipermail/slony1-general/attachments/20070512/=
ec4555b6/attachment.htm