I apologize if this post is inappropriate.
Doing some development work, me and my co-worker discussed some
optimizations strategies. One of the ideas that came up was materialized
views. Trading disk space to summarize queries, and paying for a trigger

I have loaded the patch queue with all patches that were in my main
mailbox:
    http://momjian.postgresql.org/cgi-bin/pgpatches
I still have to go through the saved patches in:

I have table with column of type "inet". Then I have my own function
resolveip(text). It works okay when I store IP addres into column typed
as text or varchar. But with type of inet I run into this problem:
select host(acc_ip), resolveip(host(acc_ip)) from access_test limit 15

Late in the 7.4 release we've added a patch that changed -fpic to -fPIC
for Linux Sparc.  We wanted to investigate that issue further later on.
I've tried building PostgreSQL with -fpic on Sparc and saw no problems.
So I suggest that we change back to -fpic until we get detailed ...

I upgraded to 7.4 a few days ago and am getting these
errors when importing dumps: "permission denied to set
session authority" and "permission denied for schema
public". Not all of the dump is imported successfully.

I haven't found anything in terms of startup- and check-scripts for
pg_autovacuum yet; usually I like to have some sort of mechanism to
check if some daemon is running and restart it if it isn't.
Of course this sort of script shouldn't be too much of a bother for more

Here is what I get:
peter ~$ pg-install/bin/initdb pg-install/var/data
...
creating directory pg-install/var/data ... initdb: failed

bytea-test.cxx -- contains detailed description of the bug and test
plan and code.
varlena.c.diff -- patch for PostgreSQL 7.5devel
src/backend/utils/adt/varlena.c#byteaout

In file postgresql-7.4/src/interfaces/ecpg/ecpglib/misc.c, this version
of FreeBSD has no definition of LONG_LONG_MIN, although it does
support an int64_t integral type.  Compilation was successful after
inserting

I have a table which contains three
columns of domain X.
If I drop the domain X with cascade,
the table remains with no columns.

Is there a function or other way to get the user's ip address the moment
  an insert is performed?
Supposed that many people with the same "pg_username" are conected to
the database so no username tracking is usefull.

Hypothetical situation: a table containing, say, 10 billion rows is backed up
and then restored with pg_restore. Would this lead to the transaction id
wraparound issue since 10B rows are imported in one "batch"?
Thanks

Based on discussions on -hackers, and baring any objections betwen now and
then, I'm going to go through all files in CVS and change:
    $Id$ -> $PostgreSQL$
I will do this the evening of Friday, November 29th ...

Have the fonts gone wild in the new 7.4 documentation?
In the tables describing the functions and
operators and in the table of contents,
there are at least three different fonts

Bruce is going to do whatever magic he needs to on Tuesday, after while
I'll bundle it up for testing (at which point in time Peter should point
out any oops I happen to have made, no after, eh? *evil grin*) ... then
we'll do an announce of its availability on Wednesday ...