[postgis-users] Loading a PostGIS dump

[Lee Hachadoorian]

Paul,

Thank you. postgres *is* still consuming CPU. In the meantime,
terminal has finally changed, outputting several more setval selects.
But it seriously had shown know change for 3-4 hours when I sent this
message.

Glad I didn't kill it and thanks for the info.

Best,
--Lee

On Wed, Aug 26, 2015 at 3:18 PM, Paul Ramsey <pramsey at cleverelephant.ca> wrote:
> The COPY command used to load tables is basically 100% silent, so it
> just might have been working away. A good thing to check is 'top' to
> see if the postgres process is busy using CPU. If so, it's hard at
> work on your process. If not, then perhaps something is awry.
>
> P
>
> On Wed, Aug 26, 2015 at 12:01 PM, Lee Hachadoorian
> <Lee.Hachadoorian+L at gmail.com> wrote:
>> New Linux machine, new PostGIS install (postgresql-9.3-postgis-2.1). I
>> created DB and added extensions. Now trying to load database using
>> postgis_restore.pl. This is a large database (63GB dump, created from
>> DB that was probably 600GB or more).
>>
>> The script ran for a while. Then it stopped at
>>
>>  setval
>> --------
>>      53
>> (1 row)
>>
>> It's been hanging out there for hours now, so I assume something is
>> hopelessly wrong.
>>
>> The loader created foo.dump.lst at 10:50am and I piped errors to a
>> text file. Errors file shows that I did not create a user who owns one
>> of the schemas. `ERROR:  role "geo_user" does not exist` appears 260
>> times.
>>
>> Can I just kill the loader script with Ctrl+C? Is database
>> recoverable? Should I just drop database and try again from the
>> beginning? Is failure related to missing role, or if not how do I
>> figure out what the error is so that I can get this to run to
>> completion?
>>
>> Best,
>> --Lee
>>
>> --
>> Lee Hachadoorian
>> Asst Professor of Geography, Temple University
>> http://geospatial.commons.gc.cuny.edu/