[postgis-users] no non-null/empty features, unable to compute statistics when running ANALYSE

[Paul Ramsey]

Well, first thing just bump up to your relevant latest patch release (2.3.7)
and see if that makes things better.
It might.
Second of all, confirm that the error message is in fact wrong: do those
tables causing the message in fact have some non-null/non-empty features?
Third of all, attach gdb to the process and run analyze on those suspect
tables, and watch what it does. If you don't want to do #3, dump one of the
tables and attach it to a ticket. Before doing so, though, load it into a
new database and see if it *still* complains during analyze.
P

On Tue, Jul 31, 2018 at 4:43 PM, Nathan Woodrow <woodrow.nathan at gmail.com>
wrote:

> Because I was a bit quick on the send this is the version details:
>
> PostgreSQL 9.6.6 on x86_64-pc-linux-gnu, compiled by gcc (GCC) 4.9.3,
> 64-bit POSTGIS="2.3.4 r16009" GEOS="3.6.2-CAPI-1.10.2 4d2925d6" PROJ="Rel.
> 4.9.3, 15 August 2016" GDAL="GDAL 2.1.3, released 2017/20/01"
> LIBXML="2.9.3" LIBJSON="0.12" (core procs from "2.3.2 r15302" need upgrade)
> RASTER (raster procs from "2.3.2 r15302" need upgrade)
>
> - Nathan
>
> On Wed, Aug 1, 2018 at 9:39 AM Nathan Woodrow <woodrow.nathan at gmail.com>
> wrote:
>
>> Hey,
>>
>> When I run  ANALYSE on some of our tables we seem to be getting the
>> following:
>>
>> [2018-08-01 09:20:31] [00000] no non-null/empty features, unable to compute statistics
>> [2018-08-01 09:20:31] [00000] no non-null/empty features, unable to compute statistics
>>
>>
>> I found an older ticket that was meant to be fixed related to 3D data and
>> large values on the z and m, but from what I can see we don't have any of
>> that in our database.  Dumping the WKT doesn't return any POINTZ or
>> anything like that.
>>
>> Any hints on how to debug this one.  I'm going to check other tables we
>> have but so far 2 are showing it happen and it seems to make the stats not
>> update correctly and given the table is going to be quite large I would
>> like to make sure they stay updated if possible.
>>
>> Regards,
>> Nathan
>>
>>
> _______________________________________________
> postgis-users mailing list
> postgis-users at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/postgis-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/postgis-users/attachments/20180801/1158aac1/attachment.html>