Re: slru.c race condition

jnasby@pervasive.com ("Jim C. Nasby") writes:
> AFAIK they're not using subtransactions at all, but I'll check.

Are they perchance using pl/PerlNG?

We discovered a problem with Slony-I's handling of subtransactions
which was exposed by pl/PerlNG, which evidently wraps its SPI calls
inside subtransactions.

For more details...
<http://gborg.postgresql.org/project/slony1/bugs/bugupdate.php?1293>

That is the only subtransaction issue I am aware of...
-- 
select 'cbbrowne' || '@' || 'acm.org';
http://www.ntlug.org/~cbbrowne/nonrdbms.html
:FATAL ERROR -- ERROR IN ERROR HANDLER