> Well, when the preceding comment block contains five references to
> xemacs and the link for more information leads to www.xemacs.org,
> I don't think it's real helpful to add one sentence saying "oh
> by the way we're not actually following xemacs".
>
> I continue to think that we'd be better o
Tatsuo Ishii writes:
> Done along with comment that we follow emacs's implementation, not
> xemacs's.
Well, when the preceding comment block contains five references to
xemacs and the link for more information leads to www.xemacs.org,
I don't think it's real helpful to add one sentence saying "oh
Markus Wanner writes:
> I stumbled across an initialization of a BlockNumber with InvalidBuffer,
> which seems strange to me, as the values for "invalid" of the two types
> are different, see attached patch.
That's certainly bogus ...
> In case the 'stack' argument passed to that function is not
On Tue, Jul 10, 2012 at 09:10:25PM +0200, Magnus Hagander wrote:
> On Tue, Jul 10, 2012 at 7:27 PM, Bruce Momjian wrote:
> > On Tue, Jul 10, 2012 at 07:06:39PM +0200, Magnus Hagander wrote:
> >> >> >> rsync where and how? What are you actually trying to suggest people
> >> >> >> do?
> >> >> >
> >>
Tatsuo Ishii writes:
>> So far as I can see, the only LCPRVn marker code that is actually in
>> use right now is 0x9d --- there are no instances of 9a, 9b, or 9c
>> that I can find.
>>
>> I also read in the xemacs internals doc, at
>> http://www.xemacs.org/Documentati
On Tue, Jul 10, 2012 at 2:42 PM, Dimitri Fontaine
wrote:>
> What you explain you want reads to me "Async replication + Archiving".
Notable caveat: one can't very easily measure or bound the amount of
transaction loss in any graceful way as-is. We only have "unlimited
lag" and "2-safe or bust".
Alvaro Herrera writes:
> FWIW about the new include: I feel a strong dislike about the forward
> declaration you suggest. Defining Relation in elog.h seems completely
> out of place.
Agreed. Maybe a reasonable solution is to allow some ereport helper
functions (or, really, wrappers for the hel
On Tue, Jul 10, 2012 at 1:38 PM, Heikki Linnakangas <
heikki.linnakan...@enterprisedb.com> wrote:
> I think the ultimate question is, which ones of these should we include in
> core? We cannot drop the existing range_ops opclass, if only because that
> would break pg_upgrade. However, range_ops2 s
On 07/10/2012 12:50 AM, Magnus Hagander wrote:
> On Tue, Jul 10, 2012 at 3:23 AM, Joe Conway wrote:
>> I verified this behavior on head as well as 9.1 (didn't bother looking
>> any further back). Looks like the reason is that CreateFunction()
>> correctly checks lanpltrusted, whereas pg_language_a
On Sun, May 20, 2012 at 2:41 PM, Joel Jacobson wrote:
> Hi,
>
> I just read a very interesting post about "schema version management".
>
> Quote: "You could set it up so that every developer gets their own
> test database, sets up the schema there, takes a dump, and checks that
> in. There are goi
On Tue, Jul 10, 2012 at 11:39 PM, Peter Eisentraut wrote:
> On sön, 2012-07-08 at 18:52 -0400, Tom Lane wrote:
>> Peter Eisentraut writes:
>> > On lör, 2012-07-07 at 17:18 -0400, Tom Lane wrote:
>> >> Sure. You need not look further than "/" to find an operator name
>> that
>> >> absolutely *wil
On 07/10/2012 05:39 PM, Peter Eisentraut wrote:
On sön, 2012-07-08 at 18:52 -0400, Tom Lane wrote:
Peter Eisentraut writes:
On lör, 2012-07-07 at 17:18 -0400, Tom Lane wrote:
Sure. You need not look further than "/" to find an operator name
that
absolutely *will* cause trouble if it's dum
Shaun Thomas writes:
> When you re-connect a secondary device, it catches up as fast as possible by
> replaying waiting transactions, and then re-attaching to the cluster. Until
> it's fully caught-up, it doesn't exist. DRBD acknowledges the secondary is
> there and attempting to catch up, but doe
On sön, 2012-07-08 at 18:52 -0400, Tom Lane wrote:
> Peter Eisentraut writes:
> > On lör, 2012-07-07 at 17:18 -0400, Tom Lane wrote:
> >> Sure. You need not look further than "/" to find an operator name
> that
> >> absolutely *will* cause trouble if it's dumped into a filename
> >> literally.
>
On 10 July 2012 21:26, Alvaro Herrera wrote:
>> I haven't defined Relation in elog.h; I have pre-declared it there.
>> Maybe that isn't to your taste, but there is surely something to be
>> said for adding exactly one line of code in preference to adding an
>> entire new header file, and having a
Excerpts from Peter Geoghegan's message of mar jul 10 15:54:59 -0400 2012:
> On 10 July 2012 20:28, Alvaro Herrera wrote:
> > FWIW about the new include: I feel a strong dislike about the forward
> > declaration you suggest. Defining Relation in elog.h seems completely
> > out of place. The o
Excerpts from Kyotaro HORIGUCHI's message of mar jul 03 04:59:38 -0400 2012:
> Hello, Here is regression test runs on pg's also built with
> cygwin-gcc and VC++.
>
> The patches attached following,
>
> - plperl_sql_ascii-4.patch : fix for pl/perl utf8 vs sql_ascii
> - plperl_sql_ascii_re
On 10 July 2012 20:28, Alvaro Herrera wrote:
> I think we should just define constants for the set of fields the patch
> currently uses. When and if we later add new fields to other callsites,
> we can define more constants.
Fair enough. Let's do that.
> FWIW about the new include: I feel a st
Excerpts from Peter Geoghegan's message of mar jul 10 14:56:40 -0400 2012:
> On 7 July 2012 13:57, Pavel Stehule wrote:
> >> +#define PG_DIAG_TRIGGER_SCHEMA 'h'
> >>
> >> Not all appear to have a way of setting the value within the ereport
> >> interface. For example, there is nothing l
On Tue, Jul 10, 2012 at 7:27 PM, Bruce Momjian wrote:
> On Tue, Jul 10, 2012 at 07:06:39PM +0200, Magnus Hagander wrote:
>> >> >> rsync where and how? What are you actually trying to suggest people
>> >> >> do?
>> >> >
>> >> > Updated docs attached.
>> >>
>> >> I suggest just removing the rsync pa
On 7 July 2012 13:57, Pavel Stehule wrote:
>> In my revision, I've just added a pre-declaration and removed the
>> dedicated header, which didn't make too much sense to me:
>>
>> + /* Pre-declare Relation, in order to avoid a build dependency on rel.h. */
>> + typedef struct RelationData *Relation
Am 10.07.2012 18:44, schrieb Tom Lane:
David Fetter writes:
On Tue, Jul 10, 2012 at 12:00:06PM -0400, Tom Lane wrote:
ISTM there was some discussion awhile back about user-definable
typing shortcuts in psql.
In some sense, we already have them:
Good point:
regression=# show search_path ;
On Mon, Jul 09, 2012 at 04:02:13PM -0400, Tom Lane wrote:
> ... Could you crank up the log verbosity so we can get
> file and line number, at least?
Here is what the increased verbosity reveals in aggregate. This
is about an 18-hour span, covering 12.5M transactions, on
version 8.3.18:
(13 tim
On Tue, Jul 10, 2012 at 07:06:39PM +0200, Magnus Hagander wrote:
> >> >> rsync where and how? What are you actually trying to suggest people
> >> >> do?
> >> >
> >> > Updated docs attached.
> >>
> >> I suggest just removing the rsync part completely. You're basically
> >> saying "you ca nset up a n
On Tue, Jul 10, 2012 at 6:59 PM, Bruce Momjian wrote:
> On Tue, Jul 10, 2012 at 06:29:24PM +0200, Magnus Hagander wrote:
>> > Testing maybe? I feel we have just avoided saying what you can and
>> > can't do with the standbys and pg_upgrade, so I think we have to state
>> > something. If we just
On Tue, Jul 10, 2012 at 3:23 AM, Fujii Masao wrote:
> Hi,
>
> I found several problems in pg_receivexlog, e.g., memory leaks,
> file-descripter leaks, ..etc. The attached patch fixes these problems.
>
> ISTM there are still some other problems in pg_receivexlog, so I'll
> read it deeply later.
Wh
On 7/10/12 5:22 AM, Dimitri Fontaine wrote:
> Jeff Janes writes:
>> I think we want this. There is some discussion about how much overlap
>> it has with pgfincore, but I don't think there is an active proposal
>> to put that into contrib, so don't see that as blocking this.
>
> It is my understa
On Tue, Jul 10, 2012 at 06:29:24PM +0200, Magnus Hagander wrote:
> > Testing maybe? I feel we have just avoided saying what you can and
> > can't do with the standbys and pg_upgrade, so I think we have to state
> > something. If we just want to say "recreate", let's say that.
>
> Well, the botto
Shaun,
> Too many mental gymnastics. I get that async is "faster" than sync, but
> the inconsistent transactional state makes it *look* slower. If a
> customer makes an order, but just happens to check that order state on
> the secondary before it can catch up, that's a net loss. Like I said,
> th
On Tue, Jul 10, 2012 at 6:27 AM, Magnus Hagander wrote:
> On Mon, Jul 9, 2012 at 8:23 PM, Fujii Masao wrote:
>> Hi,
>>
>> I found several problems in pg_receivexlog, e.g., memory leaks,
>> file-descripter leaks, ..etc. The attached patch fixes these problems.
>
> While I don't doubt that what you
David Fetter writes:
> On Tue, Jul 10, 2012 at 12:00:06PM -0400, Tom Lane wrote:
>> ISTM there was some discussion awhile back about user-definable
>> typing shortcuts in psql.
> In some sense, we already have them:
Good point:
regression=# show search_path ;
search_path
On Tue, Jul 10, 2012 at 06:26:22PM +0200, Colin 't Hart wrote:
> On 10 July 2012 18:24, David Fetter wrote:
>
> > On Tue, Jul 10, 2012 at 12:00:06PM -0400, Tom Lane wrote:
> > > Josh Kupershmidt writes:
> > > > On Tue, Jul 10, 2012 at 2:09 AM, Colin 't Hart
> > wrote:
> > > >> Attached please f
On Tue, Jul 10, 2012 at 6:26 PM, Bruce Momjian wrote:
> On Tue, Jul 10, 2012 at 06:21:35PM +0200, Magnus Hagander wrote:
>> On Tue, Jul 10, 2012 at 6:17 PM, Bruce Momjian wrote:
>> > On Tue, Jul 10, 2012 at 12:04:50PM -0400, Tom Lane wrote:
>> >> Bruce Momjian writes:
>> >> > +While a Log-Sh
On Tue, Jul 10, 2012 at 06:21:35PM +0200, Magnus Hagander wrote:
> On Tue, Jul 10, 2012 at 6:17 PM, Bruce Momjian wrote:
> > On Tue, Jul 10, 2012 at 12:04:50PM -0400, Tom Lane wrote:
> >> Bruce Momjian writes:
> >> > +While a Log-Shipping Standby Server ()
> >> > can
> >> > +be upgraded,
On Tue, Jul 10, 2012 at 12:00:06PM -0400, Tom Lane wrote:
> Josh Kupershmidt writes:
> > On Tue, Jul 10, 2012 at 2:09 AM, Colin 't Hart wrote:
> >> Attached please find a trivial patch for psql which adds a \n
> >> meta command as a shortcut for typing set search_path =.
>
> > I think the use-ca
On Tue, Jul 10, 2012 at 6:17 PM, Bruce Momjian wrote:
> On Tue, Jul 10, 2012 at 12:04:50PM -0400, Tom Lane wrote:
>> Bruce Momjian writes:
>> > +While a Log-Shipping Standby Server ()
>> > can
>> > +be upgraded, the server must be in changed to a primary server to
>> > allow
>> > +w
On Tue, Jul 10, 2012 at 12:04:50PM -0400, Tom Lane wrote:
> Bruce Momjian writes:
> > +While a Log-Shipping Standby Server () can
> > +be upgraded, the server must be in changed to a primary server to allow
> > +writes, and after the upgrade it cannot be reused as a standby server.
> >
Bruce Momjian writes:
> +While a Log-Shipping Standby Server () can
> +be upgraded, the server must be in changed to a primary server to allow
> +writes, and after the upgrade it cannot be reused as a standby server.
> +(Running rsync after the upgrade allows reuse.)
"in changed"?
Alvaro Herrera writes:
> Excerpts from Tom Lane's message of mar jul 10 10:44:03 -0400 2012:
>> What we really want is REINDEX CONCURRENTLY.
> http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.47.9961 ?
Hm ... that paper looks like something we might want to incorporate into
btree's VACUUM
Josh Kupershmidt writes:
> On Tue, Jul 10, 2012 at 2:09 AM, Colin 't Hart wrote:
>> Attached please find a trivial patch for psql which adds a \n meta command
>> as a shortcut for typing set search_path =.
> I think the use-case is a bit narrow: saving a few characters typing
> on a command not
I occasionally get questions about how to run pg_upgrade on log-shipping
standby servers. The attached documentation patch outlines how to do
it.
I don't think we can assume that because pg_upgrade was run on the
master and standby that they are binary identical, can we? Technically
the user fil
Alvaro Herrera writes:
> Excerpts from Tom Lane's message of mar jul 10 10:56:50 -0400 2012:
>> What's to solve? Presumably the WITH function name would take
>> precedence over anything in the catalogs, the same as WITH query names
>> take precedence over actual tables.
> Hm, would the newly def
On Tue, Jul 10, 2012 at 6:28 AM, Shaun Thomas wrote:
> On 07/10/2012 01:11 AM, Daniel Farina wrote:
>
>> So if I get this straight, what you are saying is "be asynchronous
>> replication unless someone is around, in which case be synchronous"
>> is the mode you want.
>
>
> Er, no. I think I see wh
Excerpts from Tom Lane's message of mar jul 10 10:44:03 -0400 2012:
> All of these things seem like ugly, hard-to-use kluges anyway (the
> make-sure-the-indexes-match business is just as much of a PITA for the
> DBA as it is for the system). What we really want is REINDEX
> CONCURRENTLY.
http:/
On 07/10/2012 09:40 AM, Heikki Linnakangas wrote:
You are mistaken. It only guarantees that it's been sync'd to disk in
the standby, but if there are open snapshots or the system is simply
busy, it might takes minutes or more until the effects of that
transaction become visible.
Well, crap. It
On Tue, Jul 10, 2012 at 3:44 PM, Tom Lane wrote:
>> The problem you describe is one of constraints and dependencies and
>> not one of indexes. It seems what you really want is a way to alter
>> foreign key dependencies to depend on a new index. Either an explicit
>> command that lets you set the n
Excerpts from Tom Lane's message of mar jul 10 10:56:50 -0400 2012:
> Pavel Stehule writes:
> > 2012/7/10 Dimitri Fontaine :
> >> I'm not sure I can understand the difference between that and the use
> >> case for which you want to implement DO blocks with parameters.
>
> > this is similar to te
On Tue, Jul 10, 2012 at 2:09 AM, Colin 't Hart wrote:
> Attached please find a trivial patch for psql which adds a \n meta command
> as a shortcut for typing set search_path =.
I think the use-case is a bit narrow: saving a few characters typing
on a command not everyone uses very often (I don't)
Pavel Stehule writes:
> 2012/7/10 Dimitri Fontaine :
>> I'm not sure I can understand the difference between that and the use
>> case for which you want to implement DO blocks with parameters.
> this is similar to temporary functions - you need some temporary name
> - it is insert to pg_proc, and
2012/7/10 Dimitri Fontaine :
> Pavel Stehule writes:
WITH FUNCTION foo(param list) returns rettype language foo AS (
definition here
)
;
>>>
>>> I like this idea. This gets rid of both the "how to pass parameters"
>>> and the "how to return results" issues that exist
Pavel Stehule writes:
>>> WITH FUNCTION foo(param list) returns rettype language foo AS (
>>> definition here
>>> )
>>> ;
>>
>> I like this idea. This gets rid of both the "how to pass parameters"
>> and the "how to return results" issues that exist with DO, as well as
>> assorted imple
Greg Stark writes:
> On Sat, Jul 7, 2012 at 4:53 AM, Gurjeet Singh wrote:
>> All we need to do is allow swapping of pg_class.relfilenode of two indexes.
> Fwiw I don't like swapping relfilenodes on indexes the user created.
> REINDEX currently does this but it's a bit of a hack and only works
>
On 10.07.2012 17:31, Shaun Thomas wrote:
On 07/09/2012 05:15 PM, Josh Berkus wrote:
So I'm unclear on why sync rep would be faster than async rep given
that they use exactly the same mechanism. Explain?
Too many mental gymnastics. I get that async is "faster" than sync, but
the inconsistent tr
2012/7/10 Tom Lane :
> Dimitri Fontaine writes:
>> We could imagine several syntax to show up the idea, common keywords
>> here include LAMBDA, FLET or LABELS, but I think that expanding WITH
>> would be preferable for us.
>
>> WITH FUNCTION foo(param list) returns rettype language foo AS (
>>
On 07/09/2012 05:15 PM, Josh Berkus wrote:
"Total-consistency" replication is what I think you want, that is, to
guarantee that at any given time a read query on the master will return
the same results as a read query on the standby. Heck, *most* people
would like to have that. You would also
Dimitri Fontaine writes:
> We could imagine several syntax to show up the idea, common keywords
> here include LAMBDA, FLET or LABELS, but I think that expanding WITH
> would be preferable for us.
> WITH FUNCTION foo(param list) returns rettype language foo AS (
> definition here
> )
>
On Sat, Jul 7, 2012 at 4:53 AM, Gurjeet Singh wrote:
> All we need to do is allow swapping of pg_class.relfilenode of two indexes.
> This will let the dependency entries stand as they are and allow us to drop
> the bloated primary key index structure without having to rebuild the
> foreign key con
On Sat, Jul 7, 2012 at 5:43 PM, Noah Misch wrote:
> I like the patch, as far as it goes. It's the natural addition to the
> completions we already offer; compare the simplistic completion after WHERE.
> Like Pavel and Robert, I think a delightful implementation of tab completion
> for SELECT stat
Hi,
Attached please find a trivial patch for psql which adds a \n meta command
as a shortcut for typing set search_path =.
This allows you to use psql as follows:
\dn
\n my_schema
\d
\d my_table
etc.
Not yet done: updating documentation (psql internal help, psql man page,
main documentatio
Thom Brown writes:
> I also attach various typo/grammar fixes.
In fact Robert's cleanup of the docs make that patch of yours not apply
anymore, and I think a part of it is maybe already fixed. Do you have
time to look at this with the new v1.8 patch that you will receive in a
minute, or with the
On Tue, Jul 10, 2012 at 9:28 AM, Shaun Thomas wrote:
> Async is simply too slow for our OLTP system except for the disaster
> recovery node, which isn't expected to carry on within seconds of the
> primary's failure. I briefly considered sync mode when it appeared as a
> feature, but I see it's s
On 07/10/2012 01:11 AM, Daniel Farina wrote:
So if I get this straight, what you are saying is "be asynchronous
replication unless someone is around, in which case be synchronous"
is the mode you want.
Er, no. I think I see where you might have gotten that, but no.
This is a pretty tricky de
On Mon, Jul 9, 2012 at 08:00:48PM -0700, David Kerr wrote:
>
> On Jul 9, 2012, at 7:48 PM, Bruce Momjian wrote:
>
> > Rsync is popular with Postgres users, but I don't understand how they
> > are using the default check mode (file size, modification time) to
> > synchronize shut-down data direct
Hackers,
I stumbled across an initialization of a BlockNumber with InvalidBuffer,
which seems strange to me, as the values for "invalid" of the two types
are different, see attached patch.
In case the 'stack' argument passed to that function is not NULL, the
variable in question gets overridden i
2012/7/10 Dimitri Fontaine :
> Pavel Stehule writes:
>> When I worked on parametrised DO statement, I had to solve following issue:
>
> DO currently is a utility command, not a query. Do you mean to change
> that?
>
> Also, did you think about a lambda construct, which is basically
> allowing func
Jeff Janes writes:
> I think we want this. There is some discussion about how much overlap
> it has with pgfincore, but I don't think there is an active proposal
> to put that into contrib, so don't see that as blocking this.
It is my understanding that Cédric wants to propose a patch for
pgfinc
Pavel Stehule writes:
> When I worked on parametrised DO statement, I had to solve following issue:
DO currently is a utility command, not a query. Do you mean to change
that?
Also, did you think about a lambda construct, which is basically
allowing functions to be defined inline in a query?
We
Hi,
Attached please find a trivial patch for psql which adds a \n meta command
as a shortcut for typing set search_path =.
This allows you to navigate a database very quickly in psql as follows:
\dn
\n my_schema
\d
\d my_table
etc.
Not yet done: updating documentation (psql internal help,
On 10.07.2012 02:33, Alexander Korotkov wrote:
Hackers,
I've tested various opclasses for ranges (including currently in-core one
and my patches). I've looked into scholar papers for which datasets they
are using for testing. The lists below show kinds of datasets used in
papers.
Great! That's
On Tue, Jul 10, 2012 at 3:23 AM, Joe Conway wrote:
> I noticed today that has_language_privilege() returns incorrect answer
> for non-superuser, e.g.:
>
> 8<---
> select has_language_privilege('nobody',
> 'plperlu',
>
On Tue, Jul 10, 2012 at 8:42 AM, Amit Kapila wrote:
>> From: pgsql-hackers-ow...@postgresql.org
> [mailto:pgsql-hackers-ow...@postgresql.org] On Behalf Of Daniel Farina
>> Sent: Tuesday, July 10, 2012 11:42 AM
>>>On Mon, Jul 9, 2012 at 1:30 PM, Shaun Thomas
> wrote:
>>>
>>> 1. Slave wants to be s
Hey all,
I've begun an implementation of the proposed HTTP API [1] (with some
changes) using node.js
The project lives at
https://github.com/bjornharrtell/postgresql-http-server and
basic functionality is in place.
Feedback appriciated!
[1] http://wiki.postgresql.org/wiki/HTTP_API
Regards
/Bj
72 matches
Mail list logo