I ran into something tonight that seems relevant here, or certainly related:
I recently updated my app(s) libpq version from 9.1 to 9.3 and immediately I starting seeing:
row number 0 is out of range 0..-1
spewed to stdout.
I traced it down to this code:
if (PQresultStatus(result) == PGRES_TUPLES_OK){
if (!PQgetisnull(result, 0, 0)){
......
}
}
if the result contains 0 tuples, the message is spit to stdout. This is new behavior, I went back and rebuilt using
libversions (according to PQlibVersion) 90102, 90203, and 90300. Neither version 90102 & 90203 spewed to stdout.
I don't know if this is causing the OP's crash, but it seems to me that there is something going on in libpq that needs
tobe looked at. If nothing else than to get rid of the spewing to stdout.
alan
On Oct 10, 2013, at 9:47 AM, Bruce Momjian <bruce@momjian.us> wrote:
> On Sat, Sep 14, 2013 at 09:40:01PM -0400, Robert Nix wrote:
>> Running a pg_upgrade task is causing Segmentation fault:
>>
>> command: "/usr/lib/postgresql/9.3/bin/pg_dump" --host "/var/lib/postgresql"
>> --port 50432 --username "postgres" --schema-only --quote-all-identifiers
>> --binary-upgrade --format=custom --file="pg_upgrade_dump_6064585.custom" "u"
>>>> "pg_upgrade_dump_6064585.log" 2>&1
>> pg_dump: row number 0 is out of range 0..-1
>> Segmentation fault (core dumped)
>
> This error is coming from libpq, specifically
> check_tuple_field_number(), which is called by PQgetvalue(),
> PQgetlength(), and PQgetisnull(). As I have never seen this error
> before, there might be something wrong with the system catalogs on that
> cluster. It would be good to run pg_dump --binary-upgrade on that
> database and try to identify the object being dumped that is causing the
> crash.
>
> --
> Bruce Momjian <bruce@momjian.us> http://momjian.us
> EnterpriseDB http://enterprisedb.com
>
> + Everyone has their own god. +
>
>
> --
> Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-general
