On Mon, 6 Aug 2007, Jeff Frost wrote:

>> [root at evim ~]# rpm -q --queryformat="%{ARCH}" postgresql-server
>> x86_64
>> 
>> [root at evim ~]# file `rpm -ql postgresql-server|grep /usr/bin`
>> /usr/bin/initdb:         ELF 64-bit LSB executable, x86-64, version 1 
>> (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.9, 
>> stripped
>> /usr/bin/ipcclean:       Bourne shell script text executable
>> /usr/bin/pg_controldata: ELF 64-bit LSB executable, x86-64, version 1 
>> (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.9, 
>> stripped
>> /usr/bin/pg_ctl:         ELF 64-bit LSB executable, x86-64, version 1 
>> (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.9, 
>> stripped
>> /usr/bin/pg_resetxlog:   ELF 64-bit LSB executable, x86-64, version 1 
>> (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.9, 
>> stripped
>> /usr/bin/postgres:       ELF 64-bit LSB executable, x86-64, version 1 
>> (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.9, 
>> stripped
>> /usr/bin/postmaster:     symbolic link to `postgres'
>
> Unfortunately, I don't have those 32-bit 8.2.4 RPMs installed anywhere at the 
> moment, but will circle back on this as soon as a 32-bit machine is available 
> again.

Finally got to circle back around on this.  It seems a different 32-bit system 
acted the same.  However, doing the maximum uptime procedure of:

* drop node
* upgrade postgresql
* subscribe node

works fine.  I'm not sure why I can't subscribe a dumped version of the 
master's database with the slony schema removed on a new 8.2.4 master, but 
this is a reasonable way to go.

Thanks to everyone for the help and suggestions!


-- 
Jeff Frost, Owner 	<jeff at frostconsultingllc.com>
Frost Consulting, LLC 	http://www.frostconsultingllc.com/
Phone: 650-780-7908	FAX: 650-649-1954