Peter Eisentraut <peter_e@gmx.net> writes:
> A while ago, I blogged about the following problem:
> (http://petereisentraut.blogspot.com/2011/07/undefined-symbol.html)
While not wishing to deny that this can be a problem, I think you're
overstating this aspect:
> Now if this had been, say, plpython, which is also developed closely
> together with the backend, but is probably shipped in a separate binary
> package and has extra dependencies, so it might reasonably not be
> upgraded at the same time, there would be additional problems. We
> should figure out a way to advise packagers about putting in tight
> enough version dependencies when this happens.
This is not possible at least in the Red Hat world, because all the
subpackages have exact-version-and-release dependencies tying them
together. That's distro policy not just my whim, and I'd expect other
server-grade distros to have similar policies.
You're right though that doing a "yum update" underneath a running
server could cause transient failures until the server was restarted
with the new postgres executable.
regards, tom lane