On Wed, Feb 6, 2019 at 4:22 PM Thomas Munro
<thomas.munro@enterprisedb.com> wrote:
> On Wed, Feb 6, 2019 at 1:10 PM Justin Pryzby <pryzby@telsasoft.com> wrote:
> > This is a contrived query which I made up to try to exercise/stress bitmap
> > scans based on Thomas's working hypothesis for this error/bug. This seems to
> > be easier to hit than the other error ("could not attach to segment") - a loop
> > around this query has run into "free pages" several times today.
>
> Thanks. I'll go and try to repro this with queries that look like that.
No luck so far. My colleague Robert pointed out that the
fpm->contiguous_pages_dirty mechanism (that lazily maintains
fpm->contiguous_pages) is suspicious here, but we haven't yet found a
theory to explain how fpm->contiguous_pages could have a value that is
too large. Clearly such a bug could result in a segment that claims
too high a number, and that'd result in this error.
--
Thomas Munro
http://www.enterprisedb.com