Re: [HACKERS] Postgres mailing lists

Chris Bitmead <chris(dot)bitmead(at)bigfoot(dot)com> writes:
> Just wondering, when I do a COPY in the latest CVS into a file, it makes
> the file both where I ask it to AND in
> /home/postgres/data/base/<filename>. I've tried removing both files and
> doing it again, and it still happens. The one in /home/postgres looks
> like some kind of binary file.

Can't reproduce it here. Anyone else seeing this?

If you do something like
COPY int4_tbl TO 'int4out';
you will get the file dumped into the database directory, because that
is the current working directory of the backend; you need to give a
full path in this form of the COPY command to get the file put someplace
more useful. But I don't see how COPY could produce two output files
when it's only doing one fopen()...

It occurs to me that it's a very dangerous thing for server-side COPY
to default to storing into the DB directory; pick the name of an
existing table and boom, you just overwrote your table. Maybe the
server-side COPY command ought to insist on being given an absolute
path?

regards, tom lane

Is it really necessary that the mailing lists block "non-member
submissions"?. I have several email addresses and this is rather
inconvenient.

--
Chris Bitmead
mailto:chris(at)tech(dot)com(dot)au
http://www.techphoto.org - Photography News, Stuff that Matters

At 02:16 PM 6/16/99 +1000, Chris Bitmead wrote:
>
>Is it really necessary that the mailing lists block "non-member
>submissions"?. I have several email addresses and this is rather
>inconvenient.

In my experiences with other mailing lists - yes. If the
mailing list address shows up anywhere accessible by
web spiders, it will be found, and the list will be hit
by spam.

So I guess the question is really "which is worse, getting
hit by occassional spam or having to post from a subscribed
e-mail address?"

I have two e-mail addresses, one which I use for personal
e-mail and lists, so the answer's easy for me. Keep it
closed to subscribers.

If no one anywhere has the e-mail address of the list on
a spider-accesible web page, then opening it *might*
work unless...

Anyone ever posts the list address to Usenet.

- Don Baccus, Portland OR <dhogaza(at)pacifier(dot)com>
Nature photos, on-line guides, and other goodies at
http://donb.photo.net

On Wed, 16 Jun 1999, Chris Bitmead wrote:

>
> Is it really necessary that the mailing lists block "non-member
> submissions"?. I have several email addresses and this is rather
> inconvenient.

There is a pgsql-loopback address you can subscribe to, so that you don't
have to be a member of every list, but, yes, it is required...

Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy(at)hub(dot)org secondary: scrappy(at){freebsd|postgresql}.org

Well I reckon there should be a pseudo-subscription mechanism, where you
can register yourself as a contributor without actually getting a copy
of the email. I could subscribe under my several email addresses just so
that I can mail from all accounts but I don't really want to receive X
copies of everything. Is there a way to subscribe without getting
anything sent? A kind of suspended account I guess you would call it.

The Hermit Hacker wrote:
>
> On Wed, 16 Jun 1999, Chris Bitmead wrote:
>
> >
> > Is it really necessary that the mailing lists block "non-member
> > submissions"?. I have several email addresses and this is rather
> > inconvenient.
>
> There is a pgsql-loopback address you can subscribe to, so that you don't
> have to be a member of every list, but, yes, it is required...
>
> Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
> Systems Administrator @ hub.org
> primary: scrappy(at)hub(dot)org secondary: scrappy(at){freebsd|postgresql}.org

--
Chris Bitmead
mailto:chris(at)tech(dot)com(dot)au
http://www.techphoto.org - Photography News, Stuff that Matters

> At 02:16 PM 6/16/99 +1000, Chris Bitmead wrote:
> >
> >Is it really necessary that the mailing lists block "non-member
> >submissions"?. I have several email addresses and this is rather
> >inconvenient.
>
> In my experiences with other mailing lists - yes. If the
> mailing list address shows up anywhere accessible by
> web spiders, it will be found, and the list will be hit
> by spam.
>
> So I guess the question is really "which is worse, getting
> hit by occassional spam or having to post from a subscribed
> e-mail address?"
>
> I have two e-mail addresses, one which I use for personal
> e-mail and lists, so the answer's easy for me. Keep it
> closed to subscribers.
>
> If no one anywhere has the e-mail address of the list on
> a spider-accesible web page, then opening it *might*
> work unless...
>
> Anyone ever posts the list address to Usenet.

I think there is a way to add all your e-mail addresses to the list
without getting mail to each address. Not sure how, though.

--
Bruce Momjian | http://www.op.net/~candle
maillist(at)candle(dot)pha(dot)pa(dot)us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026

> Well I reckon there should be a pseudo-subscription mechanism, where you
> can register yourself as a contributor without actually getting a copy
> of the email. I could subscribe under my several email addresses just so
> that I can mail from all accounts but I don't really want to receive X
> copies of everything. Is there a way to subscribe without getting
> anything sent? A kind of suspended account I guess you would call it.

That's what loopback does.

>
> The Hermit Hacker wrote:
> >
> > On Wed, 16 Jun 1999, Chris Bitmead wrote:
> >
> > >
> > > Is it really necessary that the mailing lists block "non-member
> > > submissions"?. I have several email addresses and this is rather
> > > inconvenient.
> >
> > There is a pgsql-loopback address you can subscribe to, so that you don't
> > have to be a member of every list, but, yes, it is required...
> >
> > Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
> > Systems Administrator @ hub.org
> > primary: scrappy(at)hub(dot)org secondary: scrappy(at){freebsd|postgresql}.org
>
> --
> Chris Bitmead
> mailto:chris(at)tech(dot)com(dot)au
> http://www.techphoto.org - Photography News, Stuff that Matters
>
>

--
Bruce Momjian | http://www.op.net/~candle
maillist(at)candle(dot)pha(dot)pa(dot)us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026

On Wed, 16 Jun 1999, Bruce Momjian wrote:

> > At 02:16 PM 6/16/99 +1000, Chris Bitmead wrote:
> > >
> > >Is it really necessary that the mailing lists block "non-member
> > >submissions"?. I have several email addresses and this is rather
> > >inconvenient.
> >
> > In my experiences with other mailing lists - yes. If the
> > mailing list address shows up anywhere accessible by
> > web spiders, it will be found, and the list will be hit
> > by spam.
> >
> > So I guess the question is really "which is worse, getting
> > hit by occassional spam or having to post from a subscribed
> > e-mail address?"
> >
> > I have two e-mail addresses, one which I use for personal
> > e-mail and lists, so the answer's easy for me. Keep it
> > closed to subscribers.
> >
> > If no one anywhere has the e-mail address of the list on
> > a spider-accesible web page, then opening it *might*
> > work unless...
> >
> > Anyone ever posts the list address to Usenet.
>
> I think there is a way to add all your e-mail addresses to the list
> without getting mail to each address. Not sure how, though.

Subscribe all addresses to loopback. It doesn't send mail out, it's
just a place for majordomo to look for ok addresses.

Vince.
--
==========================================================================
Vince Vielhaber -- KA8CSH email: vev(at)michvhf(dot)com flame-mail: /dev/null
# include <std/disclaimers.h> TEAM-OS2
Online Campground Directory http://www.camping-usa.com
Online Giftshop Superstore http://www.cloudninegifts.com
==========================================================================

Thus spake Chris Bitmead
> Is it really necessary that the mailing lists block "non-member
> submissions"?. I have several email addresses and this is rather
> inconvenient.

Spam is inconvenient. The term for mailing lists that don't do this
sort of blocking is "spam amplifier."

--
D'Arcy J.M. Cain <darcy(at){druid|vex}.net> | Democracy is three wolves
http://www.druid.net/darcy/ | and a sheep voting on
+1 416 424 2871 (DoD#0082) (eNTP) | what's for dinner.

On Wed, 16 Jun 1999, Chris Bitmead wrote:

> Well I reckon there should be a pseudo-subscription mechanism, where you
> can register yourself as a contributor without actually getting a copy
> of the email. I could subscribe under my several email addresses just so
> that I can mail from all accounts but I don't really want to receive X
> copies of everything. Is there a way to subscribe without getting
> anything sent? A kind of suspended account I guess you would call it.
>

> > There is a pgsql-loopback address you can subscribe to, so that you
> > don't have to be a member of every list, but, yes, it is required...

> The Hermit Hacker wrote:
> >
> > On Wed, 16 Jun 1999, Chris Bitmead wrote:
> >
> > >
> > > Is it really necessary that the mailing lists block "non-member
> > > submissions"?. I have several email addresses and this is rather
> > > inconvenient.
> >
> > There is a pgsql-loopback address you can subscribe to, so that you don't
> > have to be a member of every list, but, yes, it is required...
> >
> > Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
> > Systems Administrator @ hub.org
> > primary: scrappy(at)hub(dot)org secondary: scrappy(at){freebsd|postgresql}.org
>
> --
> Chris Bitmead
> mailto:chris(at)tech(dot)com(dot)au
> http://www.techphoto.org - Photography News, Stuff that Matters
>

Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy(at)hub(dot)org secondary: scrappy(at){freebsd|postgresql}.org

At 9:16 PM -0700 6/15/99, Chris Bitmead wrote:
>Is it really necessary that the mailing lists block "non-member
>submissions"?. I have several email addresses and this is rather
>inconvenient.

I don't know how postgres does it exactly, but NetBSD has open lists and
some aggressive kind of spam filtering based on known spammers. I think
it's been very nice to have discussions, for example, which span both the
port-mac68k(at)netbsd(dot)org list and e.g. a MkLinux list. The spam filtering
seems to be very effective since I only see a piece of spam on the netbsd
lists once every 3 months or so. I would suggest you consider cooperating
with them to use whatever mechanism they use.

As another specific example there was a person on port-mac68k who was
getting postgres up on a Mac 68k system and having some problems. I tried
to cross-post my responses to the postgres-ports lists, but they never
appeared. I was doing it from this very email address on this very
machine. Aside from the 'now' - 'current' problem I reported earlier, his
were mostly related to interactions among the NetBSD installation, the
NetBSD package system and our install instructions rather than to any
fundamental deficiencies in postgres. However I think he had valid
problems which I think the Postgres documenters could address, and which
would improve our product. I think it is unfortunate that I was not able
to involve one of the postgres lists in the discussion.

Signature failed Preliminary Design Review.
Feasibility of a new signature is currently being evaluated.
h(dot)b(dot)hotz(at)jpl(dot)nasa(dot)gov, or hbhotz(at)oxy(dot)edu

On Wed, 16 Jun 1999, Henry B. Hotz wrote:

> As another specific example there was a person on port-mac68k who was
> getting postgres up on a Mac 68k system and having some problems. I tried
> to cross-post my responses to the postgres-ports lists, but they never
> appeared.

Curious about this, since if its rejected as "non-member submission", it
will get sent back to you...

Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy
Systems Administrator @ hub.org
primary: scrappy(at)hub(dot)org secondary: scrappy(at){freebsd|postgresql}.org

This is a test to make sure I really can post to the ports list.

At 5:23 AM -0700 6/17/99, The Hermit Hacker wrote:
>On Wed, 16 Jun 1999, Henry B. Hotz wrote:
>
>> As another specific example there was a person on port-mac68k who was
>> getting postgres up on a Mac 68k system and having some problems. I tried
>> to cross-post my responses to the postgres-ports lists, but they never
>> appeared.
>
>Curious about this, since if its rejected as "non-member submission", it
>will get sent back to you...
>

I didn't get a bounce message.

The ports list should allow bug reports from anybody. Also the rest of
what I said still applies: I think it is nice to allow cross-posted
discussions among normally unrelated lists (the person with the problem
wouldn't have been a known subscriber and someone with a solution might be
a subscriber only of the non-postgres lists) and NetBSD has (or had a year
ago) a filtering mechanism that allows it while still effectively blocking
spam.

Signature failed Preliminary Design Review.
Feasibility of a new signature is currently being evaluated.
h(dot)b(dot)hotz(at)jpl(dot)nasa(dot)gov, or hbhotz(at)oxy(dot)edu

What is the status of nested structures?
Like

CREATE TABLE foo (x int4, y text);
CREATE TABLE bar (z foo, y foo);

Postgres doesn't complain but then again it doesn't seem to do anything
good with it. I can't remember seeing it mentioned in the docs anywhere
but I had the idea from somewhere that the postgres of old supported it.

Using the very latest CVS I'm encountering a bug in SUBSELECTs.
This query returns 22 rows...

SELECT id,title FROM question;

This query returns 15 rows...

SELECT id,title FROM question WHERE question.id IN (SELECT webobject
FROM comment);

Therefore I would expect this query to return 7 rows. Instead it returns
none....

SELECT id,title FROM question WHERE question.id NOT IN (SELECT webobject
FROM comment);

I've tried it with a newly dumped and created database and it still
happens. However I can't get it to happen on a new and empty database
with simple data. In other words I can't come up with a simple test
case. All I can think to do therefore is put my data in a file for
download. It is 100k.

ftp://ftp.tech.com.au/pub/datadump.gz

As an aside, while creating this dump I tried to load it into another
database and drop a few tables to make it smaller. I then tried to dump
it again and got the following error...

dumpRules(): SELECT failed for table productv. Explanation from
backend: 'ERROR: cache lookup of attribute 1 in relation 1864370 failed

--
Chris Bitmead
mailto:chris(at)tech(dot)com(dot)au

Ok, I've come up with a test case now. It's got to do with nulls so I
don't know if someone's going to come back and say that this is the way
it's meant to work. It sure doesn't seem intuitive but perhaps someone
can tell me if it's correct behaviour. I feel sure it can't be because
it means one spurious record in the database could destroy lots of
previously working queries. In other words you could have a whole lot of
queries that work. Then if some joker puts a record in the database with
a null, all the other records will no longer be returned. Anyway, here
is the simple test case...

httpd=> create table a (i int, aa text);
CREATE
httpd=> create table b (i int, bb text);
CREATE
httpd=> insert into a values(1, 'foo');
INSERT 1878534 1
httpd=> insert into b values(null, 'bar');
INSERT 1878535 1
httpd=> select * from a where i not in (select i from b);
i|aa
-+--
(0 rows)

I would expect the single record in a to be returned here. Imagine I
have thousands of records in the database that this query returns. Then
someone adds a record to b with a null. Now all those previous return
values will no longer be returned. Seems really dangerous but maybe
that is how nulls work???

Chris Bitmead wrote:
>
> Using the very latest CVS I'm encountering a bug in SUBSELECTs.
> This query returns 22 rows...
>
> SELECT id,title FROM question;
>
> This query returns 15 rows...
>
> SELECT id,title FROM question WHERE question.id IN (SELECT webobject
> FROM comment);
>
> Therefore I would expect this query to return 7 rows. Instead it returns
> none....
>
> SELECT id,title FROM question WHERE question.id NOT IN (SELECT webobject
> FROM comment);
>
> I've tried it with a newly dumped and created database and it still
> happens. However I can't get it to happen on a new and empty database
> with simple data. In other words I can't come up with a simple test
> case. All I can think to do therefore is put my data in a file for
> download. It is 100k.
>
> ftp://ftp.tech.com.au/pub/datadump.gz
>
> As an aside, while creating this dump I tried to load it into another
> database and drop a few tables to make it smaller. I then tried to dump
> it again and got the following error...
>
> dumpRules(): SELECT failed for table productv. Explanation from
> backend: 'ERROR: cache lookup of attribute 1 in relation 1864370 failed
>
> --
> Chris Bitmead
> mailto:chris(at)tech(dot)com(dot)au

--
Chris Bitmead
mailto:chris(at)tech(dot)com(dot)au