Tom Lane writes:> Lee Kindness <lkindness@csl.co.uk> writes:> > I'd have to disagree with regards to the memory leaks
notbeing worth> > a mention - any such leak can cause problems when the PostgreSQL> > installation is either
unattended,long-living andor has very high> > connection levels.> I don't see how. We are talking about two strings,
nomore, no less,> that live for exactly the duration of the postmaster run. Explain to me> how any of your above
conditionswill affect this code in the slightest?
For the lock file leak yeah - it's just 2 strings once in a blue
moon. However in the more general case of freeaddrinfo and the eniron
wizardry (which is also leaking, and is done every new instance,
yeah?) it does build up.
Anyway, it's no reason to be lazy/messy, is it?
L.