Re: Truncation on restore - Mailing list pgsql-novice

From Andrew Kelly
Subject Re: Truncation on restore
Date
Msg-id 3FD5CBB7.716DDD6F@transparency.org
Whole thread Raw
In response to Truncation on restore  (Andrew Kelly <akelly@transparency.org>)
Responses Re: Truncation on restore
Re: Truncation on restore
List pgsql-novice

Tom Lane wrote:
>
> Andrew Kelly <akelly@transparency.org> writes:
> > It seems that many of the index names are being truncated
> > 31 characters and this cascades to other errors.
>
> 31 was the standard name length limit in PG 7.2.  If you see longer
> names in a dump from a 7.2 server, then that server is *not* standard
> but was built with a custom value of NAMEDATALEN.  You will need to
> build yours likewise.  I would recommend holding a gun to someone's
> head till they tell you what other nonstandard build options were
> used, too ;-)
>
>                         regards, tom lane


Oh Tom, if you only knew...

The difficulties I'm having getting information out of
these chuckle-heads is at a last stage prior to
open warfare. In fact, I'm flying to London on Monday
for a lousy 30 minute face-to-face.
Maybe seeing that some computer weenies like to train with
free weights while they mentally debug will put a little
"here you go, sir" on the table. I'll have a gun, though,
just in case...

Can I pass the value for NAMEDATALEN to the configure
sript (or use a spec file in an rpm build from src), or
do I have to make the change in postgres_ext.h?

And thanks for the help, I really appreciate it.

Andy

pgsql-novice by date:

Previous
From: Tom Lane
Date:
Subject: Re: LDFLAGS and libstdc++ on osx10.3.1
Next
From: Tom Lane
Date:
Subject: Re: Truncation on restore