On Jan 25, 2008, at 7:27 AM, Decibel! wrote:
On Wed, Jan 23, 2008 at 05:50:02PM -0500, Tom Lane wrote:
Simon Riggs <[EMAIL PROTECTED]> writes:
From looking at how Oracle does them, autonomous transactions are
completely independent of the transaction that originates them
-- they
take a new
Caleb Welton <[EMAIL PROTECTED]> writes:
> Is there any reason that int2_sum, int4_sum, and int8_sum are not marked as
> being strict?
They wouldn't work otherwise, because the transition datatypes aren't
the same as the inputs.
regards, tom lane
-
Is there any reason that int2_sum, int4_sum, and int8_sum are not marked as
being strict? All the other transition functions for sum, and every other
built in aggregation function is marked as strict, as demonstrated with:
select x.proname, t.proname, t.proisstrict
from ((pg_aggregate a left join
yet another inverse function I wrote before, though it applies for only 1D
array.
typedef struct _enuminfo{
ArrayType *data;
char*ptr;
int16 typlen;
booltypbyval;
chartypalign;
} EnumInfo;
Datum array_enum
Hi Magnus,
Would you take a look at the patch I just committed in fe-connect.c?
I found out today that PQreset() wasn't working on a GSSAPI connection,
because closePGconn hadn't been patched to clear out the GSSAPI state
(resulting in "duplicate GSS authentication request" failure). I think
I
Jeff Davis wrote:
On Sun, 2008-01-27 at 22:11 -0800, Neil Conway wrote:
p. 564 discusses the required behavior. The result of array_agg() is an
array with one element per input value, sorted according to the optional
ORDER BY clause. NULL input values are included in the array, and the
result fo
Alvaro Herrera wrote:
> Alvaro Herrera wrote:
>
> > objdump -W $object_file | \
> > awk '/DW_TAG_/ { grab=0 } /DW_TAG_typedef/ { grab=1 } /DW_AT_name/ { if
> > (grab) { print $0 } }' | \
> > sed -e 's/^.*: \([^ ]*\)/\1/' | \
> > sort | \
> > uniq
>
> I oversimplified the awk line, causing some
Jeff Davis wrote:
> On Mon, 2008-01-28 at 23:13 +, Heikki Linnakangas wrote:
>
>> "clusteredness" didn't get screwed up by a table that looks like this:
>> "5 6 7 8 9 1 2 3 4"
>>
> ...test table with a similar
> distribution to your example, and it shows a correlation of about -0.5,
>
Decibel! wrote:
> On Fri, Jan 25, 2008 at 11:40:19AM +, Simon Riggs wrote:
> > (for 8.4 ...)
> > I'd like to introduce triggers that fire when we issue a truncate:
>
> Rather than focusing exclusively on TRUNCATE, how about "triggers" that
> fire whenever any kind of DDL operation is performed
Tom Lane wrote:
> Decibel! <[EMAIL PROTECTED]> writes:
> > Would it be worthwhile to allow for logging when a lock gets upgraded?
> > That would make it easier to protect against deadlocks...
>
> There is some debug code for that in the backend, but my experience
> is that it's too noisy to have o
On Mon, 2008-01-28 at 23:13 +, Heikki Linnakangas wrote:
> Tables that are seq scanned are typically very small, like a summary
> table with just a few rows, or huge tables in a data warehousing
> system. Between the extremes, I don't think the threshold actually has
> a very big impact.
And
On Mon, 2008-01-28 at 23:13 +, Heikki Linnakangas wrote:
> It's a good point that we don't want pg_dump to screw up the cluster
> order, but that's the only use case I've seen this far for disabling
> sync scans. Even that wouldn't matter much if our estimate for
> "clusteredness" didn't get
On Dec 6, 2007 6:28 PM, Decibel! <[EMAIL PROTECTED]> wrote:
> FWIW, I've never seen anything but a performance increase or no change
> when going from 10 to 100. In most cases there's a noticeable
> improvement since it's common to have over 100k rows in a table, and
> there's just no way to captur
Simon Riggs wrote:
On Mon, 2008-01-28 at 16:21 -0500, Tom Lane wrote:
Simon Riggs <[EMAIL PROTECTED]> writes:
Rather than having a boolean GUC, we should have a number and make the
parameter "synchronised_scan_threshold".
This would open up a can of worms I'd prefer not to touch, having to do
On Sun, 2008-01-27 at 13:37 -0500, Tom Lane wrote:
> Also, does anyone object to making pg_dump just disable it
> unconditionally? Greg's original gripe only mentioned the case of
> clustered tables, but it'd be kind of a pain to make pg_dump turn it
> on and off again for different tables. And I
On Mon, 2008-01-28 at 16:21 -0500, Tom Lane wrote:
> Simon Riggs <[EMAIL PROTECTED]> writes:
> > Rather than having a boolean GUC, we should have a number and make the
> > parameter "synchronised_scan_threshold".
>
> This would open up a can of worms I'd prefer not to touch, having to do
> with wh
On Sun, 2008-01-27 at 15:02 +, Gregory Stark wrote:
> It occurred to me the other day that synchronized scans could play havoc with
> clustered tables. When you dump and reload a table even if it was recently
> clustered if any other sequential scans are happening in the system at the
> time yo
On Sun, 2008-01-27 at 12:45 -0500, Tom Lane wrote:
> Maybe a GUC variable to enable/disable syncscan?
The first iterations of the patch included a GUC.
I don't have any objection to re-introducing a GUC to enable/disable it.
However, I would suggest that it defaults to "on", because:
1. There a
On Sun, 2008-01-27 at 22:11 -0800, Neil Conway wrote:
> p. 564 discusses the required behavior. The result of array_agg() is an
> array with one element per input value, sorted according to the optional
> ORDER BY clause. NULL input values are included in the array, and the
> result for an empty gr
> On Wed, Jan 23, 2008 at 05:50:02PM -0500, Tom Lane wrote:
> > Simon Riggs <[EMAIL PROTECTED]> writes:
> > >> From looking at how Oracle does them, autonomous transactions are
> > >> completely independent of the transaction that originates them --
> they
> > >> take a new database snapshot. This
Simon Riggs <[EMAIL PROTECTED]> writes:
> Rather than having a boolean GUC, we should have a number and make the
> parameter "synchronised_scan_threshold".
This would open up a can of worms I'd prefer not to touch, having to do
with whether the buffer-access-strategy behavior should track that or
Decibel! <[EMAIL PROTECTED]> writes:
> Would it be worthwhile to allow for logging when a lock gets upgraded?
> That would make it easier to protect against deadlocks...
There is some debug code for that in the backend, but my experience
is that it's too noisy to have on by default.
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Alvaro Herrera wrote:
>> To avoid the usage of unadorned "help" (which I don't think is going to
>> ever cause conflicts with a SQL command but perhaps it's better to be
>> prepared), one idea would be to respond with "please execute \help
>> instead", an
-BEGIN PGP SIGNED MESSAGE-
Hash: RIPEMD160
> I am not thrilled about moving _some_ of pgcrypto into the backend ---
> pgcrypto right now seems well designed and if we pull part of it out it
> seems it will be less clear than what we have now. Perhaps we just need
> to document that md5(
I had a mail issue on my end which resulted in a number of outbound
emails getting stuck in a queue. They all just went out; sorry for the
flood.
--
Decibel!, aka Jim C. Nasby, Database Architect [EMAIL PROTECTED]
Give your computer some brain candy! www.distributed.net Team #1828
pgp2tr8ezOZO
On Wed, Jan 23, 2008 at 05:50:02PM -0500, Tom Lane wrote:
> Simon Riggs <[EMAIL PROTECTED]> writes:
> >> From looking at how Oracle does them, autonomous transactions are
> >> completely independent of the transaction that originates them -- they
> >> take a new database snapshot. This means that u
-BEGIN PGP SIGNED MESSAGE-
Hash: RIPEMD160
> Uh, imagine:
>
> test=> SELECT * from pg_class
> test-> help
>
> Technically 'help' is now an alias for 'pg_class'. Are you suggesting
> supporting 'help' in this usage? People were saying they forget
> semicolons, so this 'help'
On Wed, Dec 05, 2007 at 06:49:00PM +0100, Guillaume Smet wrote:
> On Dec 5, 2007 3:26 PM, Greg Sabino Mullane <[EMAIL PROTECTED]> wrote:
> > Agreed, this would be a nice 8.4 thing. But what about 8.3 and 8.2? Is
> > there a reason not to make this change? I know I've been lazy and not run
> > any a
On Mon, Jan 21, 2008 at 04:54:06PM -0500, Tom Lane wrote:
> It's probably not a good idea to have shdepReassignOwned() take only
> AccessShareLock on pg_shdepend. Even though the function itself
> merely reads the table, it is going to call functions that will take
> RowExclusiveLock, meaning that
On Sat, Dec 22, 2007 at 09:25:05AM -0500, Bruce Momjian wrote:
> So, what solutions exist? We could require the use of port numbers less
> than 1024 which typically require root and then become a non-root user,
> but that requires root to start the server. We could put the unix
I don't know abou
On Wed, Dec 05, 2007 at 01:49:20AM +, Gregory Stark wrote:
> Regardless of what mechanism is used and who is responsible for doing it
> someone is going to have to figure out which blocks are specifically
> interesting to prefetch. Bitmap index scans happen to be the easiest since
> we've alrea
On Fri, Jan 25, 2008 at 11:40:19AM +, Simon Riggs wrote:
> (for 8.4 ...)
> I'd like to introduce triggers that fire when we issue a truncate:
Rather than focusing exclusively on TRUNCATE, how about "triggers" that
fire whenever any kind of DDL operation is performed? (Ok, truncate is
more DML
Steve Atkins wrote:
On Jan 28, 2008, at 8:36 AM, Tom Lane wrote:
Andrew Dunstan <[EMAIL PROTECTED]> writes:
Kevin Grittner wrote:
It would seem reasonable to me for pg_dump to use ORDER BY to select
data from clustered tables.
What will be the performance hit from doing that?
That worrie
Alvaro Herrera wrote:
> To avoid the usage of unadorned "help" (which I don't think is going to
> ever cause conflicts with a SQL command but perhaps it's better to be
> prepared), one idea would be to respond with "please execute \help
> instead", and then \help would emit the verbose output. Per
On Jan 28, 2008, at 6:14 PM, Simon Riggs wrote:
On Sun, 2008-01-27 at 21:04 -0500, Tom Lane wrote:
[ redirecting thread to -hackers ]
Neil Conway <[EMAIL PROTECTED]> writes:
On Sun, 2008-01-27 at 21:54 +, Gregory Stark wrote:
I liked the "synchronized_sequential_scans" idea myself.
I
I am not thrilled about moving _some_ of pgcrypto into the backend ---
pgcrypto right now seems well designed and if we pull part of it out it
seems it will be less clear than what we have now. Perhaps we just need
to document that md5() isn't for general use and some function in
pgcrypto should
This has been saved for the 8.4 release:
http://momjian.postgresql.org/cgi-bin/pgpatches_hold
---
Andrew Gilligan wrote:
>
> On 20 Jan 2008, at 04:34, Tom Lane wrote:
> > Andrew Gilligan <[EMAIL PROTECTED]> writes:
On Sun, 2008-01-27 at 21:04 -0500, Tom Lane wrote:
> [ redirecting thread to -hackers ]
>
> Neil Conway <[EMAIL PROTECTED]> writes:
> > On Sun, 2008-01-27 at 21:54 +, Gregory Stark wrote:
> >> I liked the "synchronized_sequential_scans" idea myself.
>
> > I think that's a bit too long. How ab
>>> On Mon, Jan 28, 2008 at 10:36 AM, in message <[EMAIL PROTECTED]>,
Tom Lane <[EMAIL PROTECTED]> wrote:
> in general pg_dump's charter is to reproduce
> the state of the database as best it can, not to "improve" it.
Seems that I've often seen it recommended as a way to eliminate bloat.
It se
On Jan 28, 2008, at 8:36 AM, Tom Lane wrote:
Andrew Dunstan <[EMAIL PROTECTED]> writes:
Kevin Grittner wrote:
It would seem reasonable to me for pg_dump to use ORDER BY to select
data from clustered tables.
What will be the performance hit from doing that?
That worries me too. Also, in
Thank you :)
> -Original Message-
> From: Heikki Linnakangas [mailto:[EMAIL PROTECTED]
> Sent: Monday, January 28, 2008 5:35 PM
> To: Gevik Babakhani
> Cc: pgsql-hackers@postgresql.org
> Subject: Re: [HACKERS] system catalog constraints question
>
> Gevik Babakhani wrote:
> > Is there a
Magnus Hagander <[EMAIL PROTECTED]> writes:
>>> While I'm complaining: that's got to be one of the least useful error
>>> messages I've ever seen, and it's for a case that's surely going to be
>>> fairly common in practice.
> AFAIK, that one is for Kerberos only. For GSSAPI, we already use the
> g
Andrew Dunstan <[EMAIL PROTECTED]> writes:
> Kevin Grittner wrote:
>> It would seem reasonable to me for pg_dump to use ORDER BY to select
>> data from clustered tables.
> What will be the performance hit from doing that?
That worries me too. Also, in general pg_dump's charter is to reproduce
th
Gevik Babakhani wrote:
Is there a way to query the column constraints between the tables of system
catalog.
For example pg_attribute.atttypid>pg_type.oid. This is described in the
docs of course , but does the system use something like pg_constaint
or the system catalog constraints are enfo
"Guillaume Smet" <[EMAIL PROTECTED]> writes:
> It was my first idea but I didn't propose it as it's really a
> different thing IMHO. enable_* variables don't change the way
> PostgreSQL really does the job as synchronize_scans (or whatever the
> name will be) does.
> And it's not very consistent wi
Hi,
Is there a way to query the column constraints between the tables of system
catalog.
For example pg_attribute.atttypid>pg_type.oid. This is described in the
docs of course , but does the system use something like pg_constaint
or the system catalog constraints are enforced only in the cod
>>> On Mon, Jan 28, 2008 at 9:00 AM, in message <[EMAIL PROTECTED]>,
Andrew Dunstan <[EMAIL PROTECTED]> wrote:
> Kevin Grittner wrote:
> On Sun, Jan 27, 2008 at 9:02 AM, in message
>> <[EMAIL PROTECTED]>, Gregory Stark <[EMAIL PROTECTED]>
>> wrote:
>>
>>> Perhaps we should have some form
Kevin Grittner wrote:
On Sun, Jan 27, 2008 at 9:02 AM, in message
<[EMAIL PROTECTED]>, Gregory Stark <[EMAIL PROTECTED]>
wrote:
Perhaps we should have some form of escape hatch for pg_dump to request real
physical order when dumping clustered tables.
It would seem reaso
>>> On Sun, Jan 27, 2008 at 9:02 AM, in message
<[EMAIL PROTECTED]>, Gregory Stark <[EMAIL PROTECTED]>
wrote:
> Perhaps we should have some form of escape hatch for pg_dump to request real
> physical order when dumping clustered tables.
It would seem reasonable to me for pg_dump to use ORDER
Hi Florian,
Glad to see you back!
On Jan 28, 2008 3:25 PM, Florian G. Pflug <[EMAIL PROTECTED]> wrote:
> How about enable_syncscan, or enable_seqscan_sync? It's not strictly
> something the influences the planner, but maybe it's similar enough to
> justify a similar naming?
It was my first idea
Guillaume Smet wrote:
On Jan 27, 2008 9:07 PM, Markus Bertheau
<[EMAIL PROTECTED]> wrote:
2008/1/28, Tom Lane <[EMAIL PROTECTED]>:
Do we have nominations for a name? The first idea that comes to
mind is "synchronized_scanning" (defaulting to ON).
"synchronized_sequential_scans" is a bit long,
On Jan 28, 2008 8:21 AM, Gokulakannan Somasundaram <[EMAIL PROTECTED]> wrote:
> I am not seeing my mail getting listed in the archives. So i am just
> resending it, in case the above one has got missed.
It was sent. Archive processing is delayed.
--
Jonah H. Harris, Sr. Software Architect | pho
On Sun, Jan 27, 2008 at 09:32:54PM -0500, Stephen Frost wrote:
>
> > While I'm complaining: that's got to be one of the least useful error
> > messages I've ever seen, and it's for a case that's surely going to be
> > fairly common in practice. Can't we persuade GSSAPI to produce
> > something mo
Simon Riggs wrote:
On Fri, 2008-01-25 at 17:56 +0100, Zdenek Kotala wrote:
Regarding to Robert Mach's work during Google SOC on data integrity
check. I would like to improve storage module and implement some
Robert's code into the core.
I would like to make following modification:
1) Add Rea
> >> I liked the "synchronized_sequential_scans" idea myself.
>
> > I think that's a bit too long. How about "synchronized_scans", or
> > "synchronized_seqscans"?
>
> We have enable_seqscan already, so that last choice seems to fit in.
Yes looks good, how about synchronized_seqscan without plur
* Robert Treat:
> Note we've been using Theo's plperl bytea patch on one of our
> production servers for some time; if anyone wants access to that
> lmk.
I'm interested. Could you post a pointer to this code, please?
--
Florian Weimer<[EMAIL PROTECTED]>
BFK edv-consulting GmbH
"Neil Conway" <[EMAIL PROTECTED]> writes:
> AFAIK the conclusion reached by the previous thread was that to be type
> safe, you'd need one distinct pseudotype per aggregate function, along
> with some way to let the planner distinguish this class of pseudotypes
> from other types (in order to appl
On Sun, Jan 27, 2008 at 08:09:10PM -0500, Tom Lane wrote:
> If you do a manual "kill -9" (for testing purposes) on its connected
> server process, psql normally recovers nicely:
>
> regression=# select 1;
> ?column?
> --
> 1
> (1 row)
>
> -- issue kill here in another window
> r
On Sun, Jan 27, 2008 at 09:51:48PM -0500, Tom Lane wrote:
> Stephen Frost <[EMAIL PROTECTED]> writes:
> > * Tom Lane ([EMAIL PROTECTED]) wrote:
> >> Whilst trying to reproduce bug #3902 I noticed that the code doesn't
> >> work with an abbreviated host name:
>
> > Testing w/ 8.3RC2, everything see
59 matches
Mail list logo
