Re: [Patch] Check file type before calling AllocateFile() for filesout of pg data directory to avoid potential issues (e.g. hang). - Mailing list pgsql-hackers
From
Paul Guo
Subject
Re: [Patch] Check file type before calling AllocateFile() for filesout of pg data directory to avoid potential issues (e.g. hang).
On Wed, Apr 24, 2019 at 10:36 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
Alvaro Herrera <alvherre@2ndquadrant.com> writes: > On 2019-Apr-24, Paul Guo wrote: >> On Wed, Apr 24, 2019 at 12:49 PM Andres Freund <andres@anarazel.de> wrote: >>> This seems like a bad idea to me. IMO we want to support using a pipe >>> etc here. If the admin creates a fifo like this without attaching a >>> program it seems like it's their fault.
>> Oh, I never know this application scenario before. So yes, for this, we >> need to keep the current code logic in copy code.
> But the pgstat.c patch seems reasonable to me.
Nah, I don't buy that one either. Nobody has any business creating any non-Postgres files in the stats directory ... and if somebody does want to stick a FIFO in there, perhaps for debug purposes, why should we stop them?
For the pgstat case, the files for AllocateFile() are actually temp files which
are soon renamed to other file names. Users might not want to set them as fifo files.
For developers 'tail -f' might be sufficient for debugging purpose.
I'm not sure if those hardcoded temp filenames (not just those in pgstat) are used across postgres reboot.
If no, we should instead call glibc function to create unique temp files and also remove those hardcode temp
filename variables, else we also might want them to be regular files.
The case with COPY is a bit different, since there it's reasonable to be worried about collisions with other users' files --- but I agree with Andres that this change would eliminate too many valid use-cases.
