"Oliver Elphick" <olly@lfix.co.uk> writes:
> Can anyone suggest what may be happening here?
> The original bug report was that initdb stuck with postgres consuming
> most of the CPU cycles.
> Now initdb has succeeded but `\d prova' in a test database consistently
> stalls, again with postgres taking most of the CPU.
No idea, but if you could attach to the looping backend with gdb and
provide a stack backtrace, it might help.
regards, tom lane