On Thu, May 24, 2012 at 1:26 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Peter Eisentraut <peter_e@gmx.net> writes:
>> On mån, 2012-05-21 at 13:14 -0400, Tom Lane wrote:
>>> ... wait, scratch that. AFAICS, that commit was totally useless,
>>> because BlockSig should always already contain SIGQUIT.
>
>> No, because PostgresMain() deletes it from BlockSig.
>
> Ah. So potentially we have an issue in all the background processes
> that have copied-and-pasted that sigdelset call, which seems to be most
> of them.
>
> I'm inclined to think this does need to be fixed. Even if it
> accidentally works now, it seems fragile as can be; and certainly
> it's underdocumented.
+1
And I found that PG_SETMASK(&UnBlockSig) in WalSenderMain() is useless because
it's always called in PostgresMain() before entering WalSenderMain().
Regards,
--
Fujii Masao