Hi,
> OK, Denis, can you run the regression tests with your patch and see what
> is going on?
>
> > Bruce Momjian writes:
> > > Applied. Thanks. I know it is a pain to generate a new patch against
> > > the release.
> >
> > Regression tests opr_sanity and sanity_check are now failing.
This was
Bruce Momjian <[EMAIL PROTECTED]> writes:
>> Next question: why is RelationInitLockInfo using
>> RelationGetPhysicalRelationName to get the input data for
>> IsSharedSystemRelationName --- shouldn't that be a test on logical
>> relation name? Or maybe the entire premise of
>> IsSharedSystemRelati
> Patrick Welche <[EMAIL PROTECTED]> writes:
> > [ core dump due to ]
> > #0 IsSystemRelationName (relname=0x0) at catalog.c:176
> > #1 0x807ed9a in IsSharedSystemRelationName (relname=0x0) at catalog.c:197
> > #2 0x80e9272 in RelationInitLockInfo (relation=0x82af018) at lmgr.c:119
> > #3 0x81
Patrick Welche <[EMAIL PROTECTED]> writes:
> [ core dump due to ]
> #0 IsSystemRelationName (relname=0x0) at catalog.c:176
> #1 0x807ed9a in IsSharedSystemRelationName (relname=0x0) at catalog.c:197
> #2 0x80e9272 in RelationInitLockInfo (relation=0x82af018) at lmgr.c:119
> #3 0x81202ef in for
Philip Warner <[EMAIL PROTECTED]> writes:
> ... I would prefer the
> 'CREATE DATABASE' code to set a value in a table somewhere to the last OID
> used when the DB was created. This would deal with the 'extended template1'
> scenario (which, incidentally, I am already a victim of). Could a new
> at
At 21:40 21/10/00 -0400, Tom Lane wrote:
>We could implement
>this by initializing the OID counter at 16384 as now, and then rewriting
>it to 32768 at the end of initdb.
...
>BTW, this raises a point I'd never thought hard about: if the dbadmin
>adds some site-local objects to template1, then make
Peter Eisentraut <[EMAIL PROTECTED]> writes:
> Philip Warner writes:
>> pg_dump uses the OID of template1 as the last builtin OID, but this now
>> seems broken in CVS (it returns 1). Should this work? If not, what is the
>> recommended way to find the last built-in OID?
> If you define the last b
Could we set a starting OID for user stuff like (0x400) or some
other round number?
LER
* Philip Warner <[EMAIL PROTECTED]> [001021 19:52]:
> At 16:00 21/10/00 +0200, Peter Eisentraut wrote:
> >
> >Perhaps you could move the CREATE TRIGGER pg_sync_pg_pwd (or something
> >else) to the very e
Hannu Krosing <[EMAIL PROTECTED]> writes:
> Kevin O'Gorman wrote:
>> I've been unable to follow the directions
>> in the Programmer's Guide
>> for getting to the anonymous CVS server.
> I ran into the same problem a while ago.
> Both username and password in docs are WRONG
IIRC, the username/pas
Tatsuo Ishii <[EMAIL PROTECTED]> writes:
> In 7.0.2
>select to_char(sum(n),'999') from t1;
> causes backend dump a core if n is a float/numeric ...data type AND if
> sum(n) returns NULL. This seems due to a bad null pointer handling for
> aruguments of pass-by-reference data types. I think
At 16:00 21/10/00 +0200, Peter Eisentraut wrote:
>
>Perhaps you could move the CREATE TRIGGER pg_sync_pg_pwd (or something
>else) to the very end to get a more predictable starting point.
>
Unfortunately, this is the sort of thing that caused the current problem
(ie. assuming that a certain item
"Kevin O'Gorman" <[EMAIL PROTECTED]> writes:
> so I tried going back to '7 Oct 2000 10:00:00 PST' and it's better,
> but regression tests fail on the rule system. It makes the server
> die. Since rules are what I want, this won't do.
Details? AFAIK, the system was operational on 7-Oct; I did n
OK, Denis, can you run the regression tests with your patch and see what
is going on?
> Bruce Momjian writes:
>
> > Applied. Thanks. I know it is a pain to generate a new patch against
> > the release.
>
> Regression tests opr_sanity and sanity_check are now failing.
>
> --
> Peter Eisentra
Bruce Momjian writes:
> Applied. Thanks. I know it is a pain to generate a new patch against
> the release.
Regression tests opr_sanity and sanity_check are now failing.
--
Peter Eisentraut [EMAIL PROTECTED] http://yi.org/peter-e/
Applied
> First a core dump which can be relieved by:
>
> Index: catalog.c
> ===
> RCS file: /home/projects/pgsql/cvsroot/pgsql/src/backend/catalog/catalog.c,v
> retrieving revision 1.34
> diff -c -u -r1.34 catalog.c
> --- catalog.c
* Peter Eisentraut <[EMAIL PROTECTED]> [001021 13:25]:
> Larry Rosenman writes:
>
> > I was just thinking of checking uname -v and if it is 7.1.0 or 7.1.1
> > set a define that pq_comm.c sees and includes the fix. There isn't a
> > good #define yet.. :-(
>
> We could use the result of
>
> che
Larry Rosenman writes:
> I was just thinking of checking uname -v and if it is 7.1.0 or 7.1.1
> set a define that pq_comm.c sees and includes the fix. There isn't a
> good #define yet.. :-(
We could use the result of
checking host system type... i586-sco-sysv5uw7.1.1
But which one is the goo
* Peter Eisentraut <[EMAIL PROTECTED]> [001021 11:54]:
> Larry Rosenman writes:
>
> > We currently have a patch in the doc/FAQ_SCO file for the "accept
> > doesn't send AF_UNIX to the caller" problem on SCO UnixWare 7.1.[01].
> > Is there any problem with configure finding out we are on one of t
* Larry Rosenman <[EMAIL PROTECTED]> [001021 12:04]:
> * Peter Eisentraut <[EMAIL PROTECTED]> [001021 11:45]:
> > > 2) I forced CC=cc and CXX=CC in src/templates/unixware and removed the
> > > -K i486 option, and we still pass GCC options. This is *NOT* good...
> >
> > Don't do that then. Setti
Larry Rosenman writes:
> OK, then please allow --with-CC and --with-CXX. I tried it, and it
> didn't honor the --with-CXX option. Please also document same in
> ./configure --help.
The C compiler is chosen with the CC environment variable. That is
documented.
The C++ compiler can be chosen
First a core dump which can be relieved by:
Index: catalog.c
===
RCS file: /home/projects/pgsql/cvsroot/pgsql/src/backend/catalog/catalog.c,v
retrieving revision 1.34
diff -c -u -r1.34 catalog.c
--- catalog.c 2000/10/16 14:52:02
* Peter Eisentraut <[EMAIL PROTECTED]> [001021 12:02]:
> I wrote:
>
> > Larry Rosenman writes:
> >
> > > 1) when I included -with-openssl, it didn't add /usr/local/ssl/lib to
> > > the -L options, so couldn't find libssl.a
> >
> > Confirmed. It's being put into a different variable. I'll see
DOne.
> * Peter Eisentraut <[EMAIL PROTECTED]> [001021 11:45]:
> > Larry Rosenman writes:
> >
> > > 1) when I included -with-openssl, it didn't add /usr/local/ssl/lib to
> > > the -L options, so couldn't find libssl.a
> >
> > Confirmed. It's being put into a different variable. I'll see about
* Peter Eisentraut <[EMAIL PROTECTED]> [001021 11:45]:
> Larry Rosenman writes:
>
> > 1) when I included -with-openssl, it didn't add /usr/local/ssl/lib to
> > the -L options, so couldn't find libssl.a
>
> Confirmed. It's being put into a different variable. I'll see about
> fixing this.
Cool.
I wrote:
> Larry Rosenman writes:
>
> > 1) when I included -with-openssl, it didn't add /usr/local/ssl/lib to
> > the -L options, so couldn't find libssl.a
>
> Confirmed. It's being put into a different variable. I'll see about
> fixing this.
No, this is fine. It's in the LIBS variable, whe
Larry Rosenman writes:
> We currently have a patch in the doc/FAQ_SCO file for the "accept
> doesn't send AF_UNIX to the caller" problem on SCO UnixWare 7.1.[01].
> Is there any problem with configure finding out we are on one of those
> releases (uname -v), and setting a UW= variable so the use
Larry Rosenman writes:
> 1) when I included -with-openssl, it didn't add /usr/local/ssl/lib to
> the -L options, so couldn't find libssl.a
Confirmed. It's being put into a different variable. I'll see about
fixing this.
> 2) I forced CC=cc and CXX=CC in src/templates/unixware and removed the
What do other DBs do with their output variables if there is an embedded SQL
query resulting in a NULL return value? What I mean is:
exec sql select text into :txt:ind from ...
If text is NULL, ind will be set, but does txt change?
I was just told Informix blanks txt.
Michael
--
Michael Meske
That may be premature, it still doesn't build, but it's probably the
right answer when we fix the rest of it.
LER
* Bruce Momjian <[EMAIL PROTECTED]> [001021 10:42]:
> Applied. Thanks.
>
>
> > Per my ongoing discussion with PeterE, here is the patch I applied to
> > src/template/unixware:
>
BTW, if someone wants shell/telnet/ssh access to this box, please let
me know. I'm more than willing.
It's a P-III 500 w/128MB ram and 18GB disk, so I can support stuff.
Larry
--
Larry Rosenman http://www.lerctr.org/~ler
Phone: +1 972-414-9812 (voice) Internet: [EMAIL P
Applied. Thanks.
> Per my ongoing discussion with PeterE, here is the patch I applied to
> src/template/unixware:
>
> cvs diff: Diffing .
> Index: unixware
> ===
> RCS file: /cvsroot/pgsql-snap/src/template/unixware,v
> retrieving
We currently have a patch in the doc/FAQ_SCO file for the "accept
doesn't send AF_UNIX to the caller" problem on SCO UnixWare 7.1.[01].
Is there any problem with configure finding out we are on one of those
releases (uname -v), and setting a UW= variable so the user doesn't
have to remember to pa
Per my ongoing discussion with PeterE, here is the patch I applied to
src/template/unixware:
cvs diff: Diffing .
Index: unixware
===
RCS file: /cvsroot/pgsql-snap/src/template/unixware,v
retrieving revision 1.1.1.1
diff -c -r1.1.1.1
* Larry Rosenman <[EMAIL PROTECTED]> [001021 07:34]:
> I will, in the next couple of days...
Well, I pulled todays (2000/10/21) snapshot.
I didn't get very far...
1) when I included -with-openssl, it didn't add /usr/local/ssl/lib to
the -L options, so couldn't find libssl.a
2) I forced CC=cc
Philip Warner writes:
> pg_dump uses the OID of template1 as the last builtin OID, but this now
> seems broken in CVS (it returns 1). Should this work? If not, what is the
> recommended way to find the last built-in OID?
If you define the last builtin oid as the highest oid in existence after
in
On Sat, Oct 21, 2000 at 11:27:54PM +1000, Horst Herb wrote:
> > > > http://www.l-t.ee/marko/pgsql/pgcrypto-0.1.tar.gz(11k)
>
> First of all, thankd for tis contribution. I had impemented a
> similar thing for my own purposes. A problem I still have is using
> the digest for "checksumming" r
* Peter Eisentraut <[EMAIL PROTECTED]> [001021 03:42]:
> Larry Rosenman writes:
>
> > Question: Why do we (for UnixWare) force i486 optimization?
>
> No particularly good reason, I suppose. We could remove it and leave it
> up to the installer to choose the optimization level.
>
> > CXXFLAGS=
pg_dump uses the OID of template1 as the last builtin OID, but this now
seems broken in CVS (it returns 1). Should this work? If not, what is the
recommended way to find the last built-in OID?
Philip Warner|
> > > http://www.l-t.ee/marko/pgsql/pgcrypto-0.1.tar.gz(11k)
First of all, thankd for tis contribution. I had impemented a similar thing for my own
purposes. A problem I still have is using the digest for "checksumming" rows in my
tables - which is a 'MUST' for medico-legal reasons in my
At 22:33 21/10/00 +1000, Philip Warner wrote:
>
>OK. I just built with absolute latest CVS again, and pg_dump has started
>dumping the entire schema, including system tables, functions etc. Looks
>like the way it decides something is user-defined is now broken.
>
>Same is true for non-empty databa
BTW, also, if it is possible it is a good idea to remove the following
warning if there are no BLOBs in the archive: Archiver: WARNING - skipping
BLOB restoration
--
Sincerely Yours,
Denis Perchine
--
E-Mail: [EMAIL PROTECTED]
HomePage: http://www.perchine.com/
At 18:33 21/10/00 +0700, Denis Perchine wrote:
>
>;
>; Archive created at Sat Oct 21 18:05:33 2000
>; dbname: test3
>; TOC Entries: 2899
>; Compression: -1
>; Dump Version: 1.4-17
>; Format: CUSTOM
>;
>;
>; Selected TOC Entries:
>;
>2338; 15 OPERATOR = postgres
>
OK. I just bu
> >I tried to do backup/restore of empty database.
> >And get the following error:
> >
> >Archiver(db): Could not execute query. Code = 7. Explanation from backend:
> >'ERROR: OperatorDef: operator "=" already defined
> >'.
> >
> >Another funny thing is that dump of empty database occupies 657614
At 18:13 21/10/00 +0700, Denis Perchine wrote:
>
>I tried to do backup/restore of empty database.
>And get the following error:
>
>Archiver(db): Could not execute query. Code = 7. Explanation from backend:
>'ERROR: OperatorDef: operator "=" already defined
>'.
>
>Another funny thing is that dump
Hello,
I tried to do backup/restore of empty database.
And get the following error:
Connecting to database for restore
Connecting to test4 as postgres
Creating OPERATOR =
Archiver(db): Could not execute query. Code = 7. Explanation from backend:
'ERROR: OperatorDef: operator "=" already define
Bruce Momjian <[EMAIL PROTECTED]> writes:
> I just counted lines, not line content. Not sure which is more
> meaningful. Our comments are as important as the code, sometimes,
> though they do not add functionality to the application. I am not
> inclined to inflate numbers, but I am not sure th
Larry Rosenman writes:
> Question: Why do we (for UnixWare) force i486 optimization?
No particularly good reason, I suppose. We could remove it and leave it
up to the installer to choose the optimization level.
> CXXFLAGS=-O ./configure --with-perl --with-CC=cc --with-CXX=CC
>--with-includes
47 matches
Mail list logo
