Bruce Momjian <[EMAIL PROTECTED]> writes:
> Have you considered having the background writer check the pages it is
> about to write to see if they can be added to the FSM, thereby reducing
> the need for vacuum? Seems we would need to add a statistics parameter
> so pg_autovacuum would know how
> Verbus Counts wrote:
>
>> {/opt/postgres/postgresql-7.4RC1}$ ll core
>> -rw--- 1 postgres postgres726684 Nov 6 10:55 core
>
>
> Can you show us the stack trace ?
>
> Regards
> Gaetano Mendola
Here is:
(gdb) backtrace
#0 0xc0198e40 in free+0x130 () from /usr/lib/libc.2
#1 0xc00cc
Christopher Kings-Lynne wrote:
> Hey - now that we have a branch, is Bruce going to start committed the
> pgpatches2 list?
Yes, once my email backlog is cleared --- probably early next week.
--
Bruce Momjian| http://candle.pha.pa.us
[EMAIL PROTECTED] |
Tom Lane wrote:
> Bruce Momjian <[EMAIL PROTECTED]> writes:
> > Have you considered having the background writer check the pages it is
> > about to write to see if they can be added to the FSM, thereby reducing
> > the need for vacuum?
>
> The 7.4 rewrite of FSM depends on the assumption that all
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Have you considered having the background writer check the pages it is
> about to write to see if they can be added to the FSM, thereby reducing
> the need for vacuum?
The 7.4 rewrite of FSM depends on the assumption that all the free space
in a given re
Hey - now that we have a branch, is Bruce going to start committed the
pgpatches2 list?
Chris
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTE
Jan Wieck wrote:
> If the system is write-bound, the checkpointer will find that many dirty
> blocks that he has no time to nap and will burst them out as fast as
> possible anyway. Well, at least that's the theory.
>
> PostgreSQL with the non-overwriting storage concept can never have
> hot-wr
Jan Wieck wrote:
> It also contains the starting work of the discussed background buffer
> writer. Thus far, the BufferSync() done at a checkpoint only writes out
> all dirty blocks in their LRU order and over a configurable time
> (lazy_checkpoint_time in seconds). But that means at least, whil
On Thursday 06 November 2003 17:18, Peter Eisentraut wrote:
> Robert Treat writes:
> >
> > we don't need links, we need patches
> >
>
> Let me ask you the questions that people always ask of us:
>
> How does one get involved?
post proposals to pgsql-www and start coding
> Where is the code?
http
Just wondering how often the stats collector resets it self. Is this a
parameter i can change?
At my knowledge each time that you do an analyze on
your db your statistics are changed ( are not incremental
I mean), anyway you can set to reset statistics at the
start of postgres.
I think you're mix
--On Thursday, November 06, 2003 20:19:05 -0500 Tom Lane
<[EMAIL PROTECTED]> wrote:
Larry Rosenman <[EMAIL PROTECTED]> writes:
A "make check" fails at createdb with errors in the postmaster logfile:
LOG: setsockopt(TCP_NODELAY) failed: Protocol not available
[snip]
Better ask them what their p
At 11:38 AM 7/11/2003, Tom Lane wrote:
Your argument that we should add the table name to the view does have
some merit though.
Sounds good to me. It would need to be added to each view that has
constraint_name, then we should be able to cross the info schema views and
get meaningful data.
-
Larry Rosenman <[EMAIL PROTECTED]> writes:
> A "make check" fails at createdb with errors in the postmaster logfile:
> LOG: setsockopt(TCP_NODELAY) failed: Protocol not available
This is coming from
inton;
#ifdefTCP_NODELAY
on = 1;
if (setsockopt(port->soc
Stephan Szabo <[EMAIL PROTECTED]> writes:
> To -hackers: Is it still safe to send small documentation patches for 7.4
> at this point?
Of course. Docs patches are fair game up till release (although I think
Peter wants us to minimize edits to the reference pages, because
regenerating the man page
I asked my friends at SCO (who are productizing PG) to test
7.4RC1 on OpenServer 5.0.7, and received this back:
A "make check" fails at createdb with errors in the postmaster logfile:
LOG: setsockopt(TCP_NODELAY) failed: Protocol not available
Plus he needed to add:
if test "$GCC" != yes ; then
Josh Berkus <[EMAIL PROTECTED]> writes:
> Peter,
>> Let me ask you the questions that people always ask of us:
> This isn't helping. What Robert was pointing out is that we don't
> currently have enough people writing HTML and PHP to finish improving
> the site anytime soon.
Peter appeared to be
Was this problem fixed? Can I request the problem report @ FreeBSD to be
closed?
Thanks,
Palle
--On torsdag, juni 12, 2003 18.51.18 -0400 "Yves R. Crevecoeur"
<[EMAIL PROTECTED]> wrote:
Don't break BeOS support.
A new version of BeOS will be released very soon.
http://www.yellowtab.com
http://w
Philip Warner <[EMAIL PROTECTED]> writes:
> Peter may have been alluding to this, and I misunderstood, but one idea
> might be to present a mangled name in the information schema; since the
> spec expects them to be unique, perhaps the schema should present them as
> unique.
Doesn't seem like t
At 10:54 AM 7/11/2003, Philip Warner wrote:
add table OID (or something else) to the information schema
Peter may have been alluding to this, and I misunderstood, but one idea
might be to present a mangled name in the information schema; since the
spec expects them to be unique, perhaps the schem
Philip Warner wrote:
At 03:37 AM 7/11/2003, Peter Eisentraut wrote:
It would
already help a lot if we just added the table name, or something that
was
until a short time before the action believed to be the table name, or
even only the table OID, before (or after) the $1.
Can we allow/bypass
At 03:37 AM 7/11/2003, Peter Eisentraut wrote:
It would
already help a lot if we just added the table name, or something that was
until a short time before the action believed to be the table name, or
even only the table OID, before (or after) the $1.
Can we allow/bypass the pg_* restriction, and c
While investigating Scott Goodwin's recent report of trouble on Mac OS
X, I have realized that we have an unpleasant little misbehavior in our
last several releases. After a backend crash, the postmaster will
attempt to recycle (delete and recreate) the old shared memory segment.
However, if the s
On Thu, 7 Nov 2003, Grant McLean wrote:
> On Fri, 2003-11-07 at 11:31, Stephan Szabo wrote:
> >
> > On Thu, 7 Nov 2003, Grant McLean wrote:
> >
> > > So it would seem that if I include the clauses:
> > >
> > > on delete restrict on update restrict
> > >
> > > Then the 'deferrable' which fo
On Fri, 2003-11-07 at 11:31, Stephan Szabo wrote:
>
> On Thu, 7 Nov 2003, Grant McLean wrote:
>
> > So it would seem that if I include the clauses:
> >
> > on delete restrict on update restrict
> >
> > Then the 'deferrable' which follows is only applied to creates and
> > not to updates o
Nailah Ogeer wrote:
Just wondering how often the stats collector resets it self. Is this a
parameter i can change?
At my knowledge each time that you do an analyze on
your db your statistics are changed ( are not incremental
I mean), anyway you can set to reset statistics at the
start of postgres.
Just wondering how often the stats collector resets it self. Is this a
parameter i can change?
---(end of broadcast)---
TIP 2: you can get off all lists at once with the unregister command
(send "unregister YourEmailAddressHere" to [EMAIL PROTE
Verbus Counts wrote:
{/opt/postgres/postgresql-7.4RC1}$ ll core
-rw--- 1 postgres postgres726684 Nov 6 10:55 core
Can you show us the stack trace ?
Regards
Gaetano Mendola
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postma
Tom Lane wrote:
Kurt Roeckx <[EMAIL PROTECTED]> writes:
Should we just not give that error message, in case we already
binded to AF_INET6 ::?
Seems like a cure worse than the disease to me --- it could mask
real problems. I suppose we could think about dropping it from LOG
to DEBUG1 level
Peter,
> Let me ask you the questions that people always ask of us:
This isn't helping. What Robert was pointing out is that we don't currently
have enough people writing HTML and PHP to finish improving the site anytime
soon. Robert doesn't need "managerial direction."
Or did you have amb
On Thu, 7 Nov 2003, Grant McLean wrote:
> So it would seem that if I include the clauses:
>
> on delete restrict on update restrict
>
> Then the 'deferrable' which follows is only applied to creates and
> not to updates or deletes.
>
> Since 'restrict' is the default, the clauses aren't a
Kurt Roeckx <[EMAIL PROTECTED]> writes:
> Should we just not give that error message, in case we already
> binded to AF_INET6 ::?
Seems like a cure worse than the disease to me --- it could mask
real problems. I suppose we could think about dropping it from LOG
to DEBUG1 level, so that it wouldn'
Hi Hackers
I have noticed an issue with deferrable triggers not always being
deferrable. Eg:
alter table subsession
add constraint fk_subsession_session foreign key (session_id)
references session (session_id)
deferrable;
alter table subsession2
add constraint fk
Robert Treat writes:
>
> we don't need links, we need patches
>
Let me ask you the questions that people always ask of us:
How does one get involved?
Where is the code?
What is the plan?
Where is the roadmap?
Where can issues be discussed?
Who is working on this?
How can we help?
--
Peter Ei
On Thu, Nov 06, 2003 at 03:42:39PM -0500, Tom Lane wrote:
> >> Is it possible that that kernel considers binding to an IPv6 port to
> >> conflict with binding to the "same" port number as an IPv4 port?
>
> Actually, I think that that may be expected behavior depending on the
> vintage of the kerne
Josh Berkus writes:
> But it does! You pointed it out yourself for the hackers & OSS tech
> people, they can just look at the descriptions of the major contributors and
> figure things out for themselves. They don't need a list with company logos
> & links.
Other people have pointed out
On Thu, 2003-11-06 at 09:46, Alvaro Herrera wrote:
> On Thu, Nov 06, 2003 at 09:25:38AM -0500, Robert Treat wrote:
>
> > The advocacy site does have different requirements than the main site,
> > namely its bi-lingualness and the different target audience, but perhaps
> > with adding bi-lingual ca
Greetings,
PostgreSQL built with:
export CC=cc
./configure --without-readline --without-zlib
The build and install went OK.
{/opt/postgres/postgresql-7.4RC1}$ initdb -D /usr/local/pgsql/data
The files belonging to this database system will be owned by user
"postgres". This user must also own
Peter,
> Hence my original point: the list of supporting companies
> does not primarily belong in the advocacy realm.
But it does! You pointed it out yourself for the hackers & OSS tech
people, they can just look at the descriptions of the major contributors and
figure things out for the
Josh Berkus writes:
> Yes? That's exactly the intention -- so that existing users and interested
> parties can see the companies that give major resources to the project.
Yes, but existing users and most interested parties don't fall into the
PHB category, nor do most PHB's fall into the existi
Andrew Sullivan writes:
> I'm not sure that's all it's for. Every time we talk about using
> Postgres, people want to know who else uses it.
True, but for that you're looking at the wrong list. This is the list of
contributors, not of users.
--
Peter Eisentraut [EMAIL PROTECTED]
-
>> Is it possible that that kernel considers binding to an IPv6 port to
>> conflict with binding to the "same" port number as an IPv4 port?
Actually, I think that that may be expected behavior depending on the
vintage of the kernel. Note the following comment in
StreamServerPort():
/*
Peter,
> Right now, the list is nothing more than a
> marketing tool for the listed companies for attracting existing users to
> them.
Yes? That's exactly the intention -- so that existing users and interested
parties can see the companies that give major resources to the project.
This has
Tom Lane wrote:
Peter Eisentraut <[EMAIL PROTECTED]> writes:
When I start up with -i, I get the following log:
LOG: could not bind IPv4 socket: Address already in use
There is no other postmaster running anywhere. I suspect that this has to
do with IPv6. This is a SuSE 8.something ma
Peter Eisentraut <[EMAIL PROTECTED]> writes:
> I don't understand this business, but if it helps, below is my ifconfig
> output.
Hmm, you have a bunch of addresses don't you? It looks like we should
have included more information in the report of bind failures, like
exactly which address failed.
Andrew Sullivan wrote:
On Thu, Nov 06, 2003 at 09:08:57PM +0100, Peter Eisentraut wrote:
I think that list is a pretty dumb idea in the first place. We have a
list of developers with company names next to them. Let readers make
their own recognition evaluation.
Your assuming that peopl
Tom Lane writes:
> Is it possible that that kernel considers binding to an IPv6 port to
> conflict with binding to the "same" port number as an IPv4 port?
I don't understand this business, but if it helps, below is my ifconfig
output.
eth0 Link encap:Ethernet HWaddr 00:40:F6:74:BE:71
Hello,
My feeling is that advocacy should be just that: Advocacy.
It doesn't matter who the intended audience is in reality. However,
it is also important to remember that technical experts typically
don't need to be sold on PostgreSQL.
PHBs on the other hand probably do and thus much of our
Adv
On Thu, Nov 06, 2003 at 09:08:57PM +0100, Peter Eisentraut wrote:
>
> I think that list is a pretty dumb idea in the first place. We have a
> list of developers with company names next to them. Let readers make
> their own recognition evaluation.
I'm not sure that's all it's for. Every time we
Peter Eisentraut <[EMAIL PROTECTED]> writes:
> I think that list is a pretty dumb idea in the first place. We have a
> list of developers with company names next to them. Let readers make
> their own recognition evaluation.
That works if you think that the only form of corporate support is
spons
Josh Berkus writes:
> > > I was discussing specifically the "Recognized Corporate Contributors" which
> > > is, AFAIK, strictly a PHB thing, no?
> >
> > No.
>
> Please explain.
I don't see anything in this project that should be strictly a PHB thing,
the exception maybe being the weird whitepaper
Peter Eisentraut <[EMAIL PROTECTED]> writes:
> When I start up with -i, I get the following log:
> LOG: could not bind IPv4 socket: Address already in use
> There is no other postmaster running anywhere. I suspect that this has to
> do with IPv6. This is a SuSE 8.something machine that is relat
When I start up with -i, I get the following log:
LOG: could not bind IPv4 socket: Address already in use
HINT: Is another postmaster already running on port 5432? If not, wait a few seconds
and retry.
LOG: database system was shut down at 2003-11-06 20:47:54 CET
LOG: checkpoint record is at
> > Could we see the pg_stats rows for service.account_id and
> > account.account_id?
Sorry, ignore previous numbers. My prior tests were done in a
transaction (to roll back stats changes) and I forgot to re-analyze.
relname | attname | stanullfrac | stawidth | stadistinct | stakind1
| staki
Peter,
> > I was discussing specifically the "Recognized Corporate Contributors"
which
> > is, AFAIK, strictly a PHB thing, no?
>
> No.
Please explain.
--
-Josh Berkus
Aglio Database Solutions
San Francisco
---(end of broadcast)---
TIP 1: su
On Thu, 2003-11-06 at 10:35, Tom Lane wrote:
> Rod Taylor <[EMAIL PROTECTED]> writes:
> >> -> Hash Join (cost=3D1230.79..60581.82 rows=3D158 width=3D54)=
> > (actual time=3D1262.35..151200.29 rows=3D1121988 loops=3D1)
> >> Hash Cond: ("outer".account_id =3D "inner".account_id)
> >> -> Hash Join
Josh Berkus writes:
> I was discussing specifically the "Recognized Corporate Contributors" which
> is, AFAIK, strictly a PHB thing, no?
No.
--
Peter Eisentraut [EMAIL PROTECTED]
---(end of broadcast)---
TIP 4: Don't 'kill -9' the postmaster
Peter,
> I'm a bit lost here.
I was discussing specifically the "Recognized Corporate Contributors" which
is, AFAIK, strictly a PHB thing, no?
--
-Josh Berkus
Aglio Database Solutions
San Francisco
---(end of broadcast)---
TIP 5: Have you che
> -Original Message-
> From: Alvaro Herrera [mailto:[EMAIL PROTECTED]
> Sent: 06 November 2003 14:47
> To: Robert Treat
> Cc: Dave Page; Peter Eisentraut; Josh Berkus;
> [EMAIL PROTECTED]; [EMAIL PROTECTED]
> Subject: Re: [pgsql-www] [HACKERS] Changes to Contributor List
>
> On Thu, N
Josh Berkus writes:
> And if we're gonna continue this thread, we should move it to -Advocacy.
I'm a bit lost here.
It was recently said very clearly, "The target audience of the advocacy
site is PHB's, not technical people." And the content of the site
supports that in my mind. Yet, the advoc
Tom Lane wrote:
Peter Eisentraut <[EMAIL PROTECTED]> writes:
I don't think we really need a method to guarantee unique names. It would
already help a lot if we just added the table name, or something that was
until a short time before the action believed to be the table name, or
even only the
Alvaro Herrera <[EMAIL PROTECTED]> writes:
> On Thu, Nov 06, 2003 at 11:42:13AM -0500, Tom Lane wrote:
>> I don't have a problem with switching from "$1" to "tablename_$1", or
>> some such, for auto-generated constraint names. But if it's not
>> guaranteed unique, does it really satisfy Philip's c
Peter,
> Btw., what process is used to determine which organizations become a
> "recognised contributor"?
Yeah, that's another "ToDo" item ... your company needs to go up there.
Criteria are major code contributions and/or sponsoring a full-time developer.
We've discussed it on -CORE some, but
On Thu, Nov 06, 2003 at 11:42:13AM -0500, Tom Lane wrote:
> Peter Eisentraut <[EMAIL PROTECTED]> writes:
> > I don't think we really need a method to guarantee unique names. It would
> > already help a lot if we just added the table name, or something that was
> > until a short time before the act
Peter Eisentraut <[EMAIL PROTECTED]> writes:
> I don't think we really need a method to guarantee unique names. It would
> already help a lot if we just added the table name, or something that was
> until a short time before the action believed to be the table name, or
> even only the table OID, b
Tom Lane writes:
> > Would a good halfway house be to ensure that generated names were unique
> > within a schema (e.g. instead of generating "$1" generate
> > "tablename$1")?
>
> No, because that buys into all of the serialization and deadlocking
> problems that doing it the spec's way entail
I
Andrew Dunstan <[EMAIL PROTECTED]> writes:
>> The reason the spec defines these views this way is that it expects
>> constraint names to be unique across a whole schema. We don't enforce
>> that, and I don't think we want to start doing so (that was already
>> proposed and shot down at least once)
Tom Lane wrote:
The reason the spec defines these views this way is that it expects
constraint names to be unique across a whole schema. We don't enforce
that, and I don't think we want to start doing so (that was already
proposed and shot down at least once). You are of course free to use
const
Rod Taylor <[EMAIL PROTECTED]> writes:
>> -> Hash Join (cost=3D1230.79..60581.82 rows=3D158 width=3D54)=
> (actual time=3D1262.35..151200.29 rows=3D1121988 loops=3D1)
>> Hash Cond: ("outer".account_id =3D "inner".account_id)
>> -> Hash Join (cost=3D1226.78..52863.43 rows=3D1542558 w=
> idth=3D
Philip Warner <[EMAIL PROTECTED]> writes:
> Notice that the two records are identical because the two constraint names
> are the same. ISTM that we should have a way of usefully examining specific
> constraints without having to name them. Can we add the constraint OID or
No. The schemas of th
On Thu, Nov 06, 2003 at 09:25:38AM -0500, Robert Treat wrote:
> The advocacy site does have different requirements than the main site,
> namely its bi-lingualness and the different target audience, but perhaps
> with adding bi-lingual capabilities to the main site these two sites
> could be brough
On Thu, 2003-11-06 at 03:21, Dave Page wrote:
>
>
> > -Original Message-
> > From: Peter Eisentraut [mailto:[EMAIL PROTECTED]
> > Sent: 05 November 2003 23:11
> > To: Josh Berkus
> > Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]
> > Subject: Re: [HACKERS] Changes to Contributor List
> >
> >
Just looking at the information schema in 7.4 and noticed something
odd/annoying/problematic:
create table pk(f1 int primary key);
create table fk1(f1 int references pk(f1));
create table fk2(f1 int references pk(f1));
select * from information_schema.referential_constraints;
-[ RECORD 1 ]--
Emeritus is verbatin from Latin and is really very spread into most Western
languages.
--
Paulo Scardine
>
> I think the "Emeritus" word might be too hard for non-native English
> speakers, and even for less educated English speakers.
---(end of broadcast)---
Josh Berkus writes:
> Yeah, you're right,it's confusing we should have two seperate pages, one
> for "What is the PGDG" possibly linking to developer., and one page for
> "Contact Us". As it is, they two are munged together.
Btw., what process is used to determine which organizations becom
> -Original Message-
> From: Peter Eisentraut [mailto:[EMAIL PROTECTED]
> Sent: 05 November 2003 23:11
> To: Josh Berkus
> Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]
> Subject: Re: [HACKERS] Changes to Contributor List
>
> Josh Berkus writes:
>
> > If possible, for the upcoming release
> -Original Message-
> From: Josh Berkus [mailto:[EMAIL PROTECTED]
> Sent: 05 November 2003 22:27
> To: Bruce Momjian; Tom Lane
> Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]
> Subject: Re: [pgsql-www] [HACKERS] Changes to Contributor List
>
> Guys,
>
> Oh, and how about we kill the Image
76 matches
Mail list logo
