Bruce Momjian <bruce@momjian.us> writes:
> Tom Lane wrote:
>> elog(PANIC) => abort() => core dump. This is completely expected.
> Well, then I should have said there is no reason you should ever get a
> panic.
The reason it's a PANIC is that the whole of StartupXLOG() runs as a
critical section. Given the recent changes to make elog(FATAL) do
exit(1) instead of exit(0), I think it might be possible to dispense
with that, but it's rather a ticklish thing to be changing the day
before we wrap 8.2. How excited are you about this?
regards, tom lane