Hi,
The immediate shutdown (pg_ctl -m i stop) might not be able to
kill the startup process during archive recovery. It's because
the startup process calls system() which ignores SIGQUIT for
executing the restore_command. So, only the startup process
might survive the immediate shutdown and
Hi,
On Fri, Nov 28, 2008 at 6:56 PM, Fujii Masao [EMAIL PROTECTED] wrote:
Hi,
The immediate shutdown (pg_ctl -m i stop) might not be able to
kill the startup process during archive recovery. It's because
the startup process calls system() which ignores SIGQUIT for
executing the
Following test end up with the server crash into 8.4 cvs Head.
uname -a
Linux localhost.localdomain 2.6.18-53.el5 #1 SMP Wed Oct 10 16:34:19 EDT
2007 x86_64 x86_64 x86_64 GNU/Linux
Testcase with ISN contrib module:
=
CREATE OR REPLACE function isbn_issn_proc() returns
Stefan Kaltenbrunner wrote:
no it is not the -HEAD situation is significantly worse because what we
are actually shipping as the manpages there are the ones from 7.4 ...
So how do we get Marc to fix this?
--
Alvaro Herrerahttp://www.CommandPrompt.com/
The
Gianni Ciolli wrote:
On Tue, Nov 11, 2008 at 02:59:01PM +0100, Magnus Hagander wrote:
Gianni Ciolli wrote:
WARNING: detected write past chunk end in Postmaster 0x9b13650
Yes, that's a stupid bug:
(...)
Attached is an updated version of the patch that fixes this.
Hi Magnus,
Simon Riggs wrote:
On Tue, 2008-11-25 at 12:21 +0200, Peter Eisentraut wrote:
Peter Eisentraut wrote:
Here is an implementation of distinct types,
I'm withdrawing this patch from the current commit fest for further
work. For the record, I have attached the current patch, including the
On Thu, 2008-11-27 at 17:14 +, Simon Riggs wrote:
On Wed, 2008-11-26 at 18:02 +0530, Pavan Deolasee wrote:
I think whats happening is that
ResolveRecoveryConflictWithVirtualXIDs() is failing to abort
the open transaction
Btw, ISTM that SIGINT works
On Tue, Sep 9, 2008 at 3:04 AM, MUHAMMAD ASIF [EMAIL PROTECTED] wrote:
I am currently working in EnterpriseDB (www.enterprisedb.com). During
the integration of pldebugger ( http://pgfoundry.org/projects/edb-debugger )
with postgres on windows I faced a problem that plugins are not being
I had a deeper look at this now. The patch looks clean and applies without
any problems, regression tests passes. However, ATRewriteTables() has a
problem when adding columns with domains and constraints. Consider this
small test case:
CREATE TABLE bar (id INTEGER);
CREATE OR REPLACE VIEW
Peter Eisentraut wrote:
I wrote:
Some more information on this:
https://www.switch.ch/pki/meetings/2007-01/namebased_ssl_virtualhosts.pdf
slide 5 lists the matching rules for email, HTTP, and LDAP over TLS,
respectively, which are not all the same. Also note that these methods
have rules
On Fri, 2008-11-28 at 19:53 +0900, Fujii Masao wrote:
Hi,
On Fri, Nov 28, 2008 at 6:56 PM, Fujii Masao [EMAIL PROTECTED] wrote:
Hi,
The immediate shutdown (pg_ctl -m i stop) might not be able to
kill the startup process during archive recovery. It's because
the startup process calls
On Nov 28, 2008, at 12:46 PM, Peter Eisentraut wrote:
I understand, but the work required to make it work properly is too
much under the commit fest spirit right now. In particular, I'm
thinking we should try to devise a clever way to make the CREATE
ORDERING facility that SQL has for
David E. Wheeler wrote:
Speaking of other sorts of derived types: might they include something
just like enums, but sorting on the string values defined for the enum
rather than on the order in which the values were defined in the enum?
I'd use something like that all the time…
order
Fujii Masao [EMAIL PROTECTED] writes:
You export replication_timeout as a PGC_USERSET variable, but it is
dangerous. It allows non-superusers to kill servers easily by setting it
too low value. Walsender dies with FATAL on timeout.
Unlike other background processes, FATAL by walsender doesn't
On Nov 28, 2008, at 5:09 PM, Andrew Dunstan wrote:
Speaking of other sorts of derived types: might they include
something just like enums, but sorting on the string values defined
for the enum rather than on the order in which the values were
defined in the enum? I'd use something like
Simon Riggs [EMAIL PROTECTED] writes:
After some thought, the way I would handle this is by sending a slightly
different kind of signal.
We can send a shared invalidation message which means end the
transaction, whether or not you are currently running a statement.
No, a thousand times no.
Peter Eisentraut [EMAIL PROTECTED] writes:
Simon Riggs wrote:
Shame, this was sorely needed.
I understand, but the work required to make it work properly is too much
under the commit fest spirit right now.
Personally I was wondering exactly why it's sorely needed. There has
been not one
Alvaro Herrera [EMAIL PROTECTED] writes:
So how do we get Marc to fix this?
Actually I think the ball is in Peter's court now. Marc updated the
snapshot build script, but all the attempts are dying now because the
new docbook2man part of the Makefiles doesn't work on svr1.
On Fri, 2008-11-28 at 11:14 -0500, Tom Lane wrote:
Simon Riggs [EMAIL PROTECTED] writes:
After some thought, the way I would handle this is by sending a slightly
different kind of signal.
We can send a shared invalidation message which means end the
transaction, whether or not you are
On Fri, 2008-11-28 at 11:20 -0500, Tom Lane wrote:
Peter Eisentraut [EMAIL PROTECTED] writes:
Simon Riggs wrote:
Shame, this was sorely needed.
I understand, but the work required to make it work properly is too much
under the commit fest spirit right now.
Personally I was
Simon Riggs [EMAIL PROTECTED] writes:
On Fri, 2008-11-28 at 11:14 -0500, Tom Lane wrote:
The sinval queue is an *utterly* inappropriate
mechanism for such a thing.
To be honest, it did seem quite a neat solution. Any particular
direction of thought you'd like me to pursue instead?
I hadn't
Simon Riggs [EMAIL PROTECTED] writes:
On Fri, 2008-11-28 at 11:20 -0500, Tom Lane wrote:
If it's going to take a significant amount of work then I think someone
ought to provide an actual justification why it's worth the work.
Few thoughts:
* Domains don't work very well in conjunction
Magnus Hagander [EMAIL PROTECTED] writes:
See attached patch which takes out the parts of fnmatch that we're not
interested in, and puts it directly in fe-secure.c. Obviously, if we go
down that way, we can remove fnmatch.c from port again :-)
Thoughts?
Generally +1, but a couple of
Rushabh Lathia [EMAIL PROTECTED] writes:
Following test end up with the server crash into 8.4 cvs Head.
Hmm, this'd have been noticed sooner if contrib/isn had even
minimal regression tests :-( Anyone feel like writing some?
regards, tom lane
--
Sent via pgsql-hackers
On Fri, 2008-11-28 at 11:44 -0500, Tom Lane wrote:
Simon Riggs [EMAIL PROTECTED] writes:
On Fri, 2008-11-28 at 11:14 -0500, Tom Lane wrote:
The sinval queue is an *utterly* inappropriate
mechanism for such a thing.
To be honest, it did seem quite a neat solution. Any particular
The problem reported by Rushabh Lathia boils down to the fact that
contrib/isn intends to define a datatype that is represented just like
int8, but it supposes that int8 must be pass by reference. There is
not anything wrong with the C code; the problem is the CREATE TYPE
command in isn.sql. To
Simon Riggs [EMAIL PROTECTED] writes:
On Fri, 2008-11-28 at 11:44 -0500, Tom Lane wrote:
I hadn't been following the discussion closely enough to know what the
problem is.
When we replay an AccessExclusiveLock on the standby we need to kick off
any current lock holders, after a configurable
On Fri, 2008-11-28 at 12:45 -0500, Tom Lane wrote:
Simon Riggs [EMAIL PROTECTED] writes:
On Fri, 2008-11-28 at 11:44 -0500, Tom Lane wrote:
I hadn't been following the discussion closely enough to know what the
problem is.
When we replay an AccessExclusiveLock on the standby we need
KaiGai Kohei [EMAIL PROTECTED] writes:
In my understanding, the matter comes from the mixture of two kind of
tuples. The one has object identifier, and the other don't have.
It seems to me the current implementation assumes fetched tuples have
proper rowtype which matches to the current table
On Fri, Nov 28, 2008 at 6:45 PM, Tom Lane [EMAIL PROTECTED] wrote:
Hm. People have complained of that fact from time to time in normal
usage as well. Should we simply change SIGINT handling to allow it to
cancel an idle transaction?
If this means that we would be able to able to easily
I'm trying to actually run the experiments Tom suggested running EXPLAIN on
the DBT3 DSS queries for various default_stats_target sizes. It's a bit of a
headache because DBT3 seems to be unmaintained these days so lots of things
don't really work.
I ran into an error I've never seen before:
Tom Lane wrote:
Magnus Hagander [EMAIL PROTECTED] writes:
See attached patch which takes out the parts of fnmatch that we're not
interested in, and puts it directly in fe-secure.c. Obviously, if we go
down that way, we can remove fnmatch.c from port again :-)
Thoughts?
Generally +1, but
Guillaume Smet wrote:
On Fri, Nov 28, 2008 at 6:45 PM, Tom Lane [EMAIL PROTECTED] wrote:
Hm. People have complained of that fact from time to time in normal
usage as well. Should we simply change SIGINT handling to allow it to
cancel an idle transaction?
If this means that we
Magnus Hagander [EMAIL PROTECTED] writes:
Tom Lane wrote:
* This seems to be still mostly NetBSD code, so I think you need to do
more than just credit them in an aside. Should we repeat the full
NetBSD copyright notice for this one function?
Do you mean the
* Copyright (c) 1989, 1993,
Greg Stark [EMAIL PROTECTED] writes:
I ran into an error I've never seen before:
ERROR: XX000: could not devise a query plan for the given query
That's cause I just recently broke it :-(
Fixed.
regards, tom lane
--
Sent via pgsql-hackers mailing list
I wrote:
The problem reported by Rushabh Lathia boils down to the fact that
contrib/isn intends to define a datatype that is represented just like
int8, but it supposes that int8 must be pass by reference. There is
not anything wrong with the C code; the problem is the CREATE TYPE
command in
-BEGIN PGP SIGNED MESSAGE-
Hash: RIPEMD160
NotDashEscaped: You need GnuPG to verify this message
Looks like inheritance causes a statement-level trigger to fire on
the last evaluated table in the inheritance chain. Is this the
desired behavior? If so, is there any way to predict or
Greg Sabino Mullane [EMAIL PROTECTED] writes:
Looks like inheritance causes a statement-level trigger to fire on
the last evaluated table in the inheritance chain. Is this the
desired behavior?
Hm, I think whoever wrote the statement-level trigger code completely
forgot to consider the
I was going to say something like that but couldn't come up with a
nice syntax. The syntax you gave seems obvious in retrospect.
I wonder if this should be tied in some way with creating binary-
compatible casts or anything. Probably not since the data type can
just make them itself.
greg
Greg Stark [EMAIL PROTECTED] writes:
I was going to say something like that but couldn't come up with a
nice syntax. The syntax you gave seems obvious in retrospect.
Here's a draft patch for this. It actually works out pretty well --
the bulk of the diff is just to allow processing the
On Friday 28 November 2008 17:13:54 Magnus Hagander wrote:
Matching *only* as the first character will make it impossible to make
certificates for www*.domain.com, which is AFAIK fairly popular - and
one of the examples you'll find on CA sites. But it would be fairly easy
to add this
On Friday 28 November 2008 18:25:23 Tom Lane wrote:
Alvaro Herrera [EMAIL PROTECTED] writes:
So how do we get Marc to fix this?
Actually I think the ball is in Peter's court now. Marc updated the
snapshot build script, but all the attempts are dying now because the
new docbook2man part of
On Friday 28 November 2008 22:02:52 Tom Lane wrote:
I had a better but more invasive idea: invent a new attribute for
CREATE TYPE
LIKE = typename
which means copy any unspecified representational details from the
given pre-existing type. This would allow the .sql file to know
On Friday 28 November 2008 18:49:17 Tom Lane wrote:
* Strong typing is preferable in complex applications to avoid errors
like sum(ordinal_column). Most developers use this all the time in their
3GL code but cannot use it in SQL.
The problem I see with distinct types is that the typing is
Peter Eisentraut [EMAIL PROTECTED] writes:
On Friday 28 November 2008 18:49:17 Tom Lane wrote:
The problem I see with distinct types is that the typing is *too*
strong --- the datatype has in fact got no usable operations whatever.
You are supposed to define your own. It's a new type after
Peter Eisentraut [EMAIL PROTECTED] writes:
On Friday 28 November 2008 18:25:23 Tom Lane wrote:
Actually I think the ball is in Peter's court now. Marc updated the
snapshot build script, but all the attempts are dying now because the
new docbook2man part of the Makefiles doesn't work on svr1.
Tom Lane [EMAIL PROTECTED] wrote:
If it's going to take a significant amount of work then I think
someone
ought to provide an actual justification why it's worth the work.
This by itself doesn't justify the effort, but I've worked with
databases which would refuse to allow comparison
I have a feeling what it would actually take to make this useful might
be to turn every type into a kind of polymorphic type like our anyelem
and anyarray. So substring(mystring) would work and return a mystring
but mystring=string would fail.
--
greg
--
Sent via pgsql-hackers mailing list
On Saturday 29 November 2008 00:52:03 Tom Lane wrote:
AFAICT, we are building both man.tar.gz and postgres.tar.gz on-the-fly
in current releases. At least, both of them have internal timestamps
supporting that theory in the official 8.3.5 and 8.2.11 tarballs, and
the contained files reflect
Hello,
On Sat, Nov 29, 2008 at 12:40 AM, Simon Riggs [EMAIL PROTECTED] wrote:
On Fri, 2008-11-28 at 19:53 +0900, Fujii Masao wrote:
Hi,
On Fri, Nov 28, 2008 at 6:56 PM, Fujii Masao [EMAIL PROTECTED] wrote:
Hi,
The immediate shutdown (pg_ctl -m i stop) might not be able to
kill the
50 matches
Mail list logo
