Re: PQexec() hangs on OOM

On Thu, Jul 9, 2015 at 6:34 PM, Michael Paquier <michael.paquier@gmail.com>
wrote:
>
> On Wed, Jul 8, 2015 at 1:01 AM, Heikki Linnakangas <hlinnaka@iki.fi>
wrote:
> > On 07/07/2015 09:32 AM, Michael Paquier wrote:
> >
> >
> > Ok, committed and backpatched the latest patch I posted. Yeah, we'll
need
> > additional patches for the refactoring and the remaining issues. I'm not
> > sure if it's worth trying to backpatch them; let's see how big the
patch is.
>
> So, here are two patches aimed at fixing the hangling issues with
> getStartCopy and getParamDescriptions. After trying a couple of
> approaches, I found out that the most elegant way to prevent the
> infinite loop in parseInput is to introduce a new PGASYNC flag to
> control the error and let the caller know what happened.

One thing that looks slightly non-elegant about this approach is that
new async status (PGASYNC_FATAL) is used to deal with errors only
in few paths in function pqParseInput3() and not-other paths which should
be okay if there is no other better way.  I have not spent enough time on
it to suggest any better alternative, but would like to hear what other
approaches you have considered?

With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com