On 29 August 2014 01:04, Thomas Munro mu...@ip9.org wrote:
On 28 August 2014 23:45, Tom Lane t...@sss.pgh.pa.us wrote:
I don't claim to be an expert on this stuff, but I had the idea that
multithreaded environments were supposed to track signal state per-thread
not just per-process, precisely
On 08/29/2014 05:28 AM, Tom Lane wrote:
k...@rice.edu k...@rice.edu writes:
On Thu, Aug 28, 2014 at 03:33:56PM -0400, Bruce Momjian wrote:
So the standard requires storing of original timezone in the data type?
I was not aware of that.
I do not have a copy of the SQL 92 spec, but several
On Fri, 2014-08-22 at 12:34 -0400, Robert Haas wrote:
Still doesn't look good here. On the same PPC64 machine I've been
using for testing:
I have a new approach to the patch which is to call a callback at each
block allocation and child contexts inherit the callback from their
parents. The
On 08/28/2014 07:28 PM, Alexey Klyukin wrote:
On Mon, Aug 25, 2014 at 12:02 PM, Heikki Linnakangas
hlinnakan...@vmware.com wrote:
On 08/24/2014 03:11 PM, Alexey Klyukin wrote:
On Wed, Aug 20, 2014 at 11:53 AM, Heikki Linnakangas
hlinnakan...@vmware.com wrote:
The patch doesn't seem to
On 08/28/2014 02:46 PM, Fujii Masao wrote:
On Tue, Aug 26, 2014 at 4:55 AM, Heikki Linnakangas
hlinnakan...@vmware.com wrote:
On 08/25/2014 10:48 PM, Heikki Linnakangas wrote:
Actually, perhaps it would be better to just copy-paste PSQLexec, and
modify the copy to suite \watch's needs.
On 08/28/2014 02:14 PM, Craig Ringer wrote:
Hi all
Per the protocol todo:
https://wiki.postgresql.org/wiki/Todo#Wire_Protocol_Changes
do you think it's reasonable to allow for delayed sending of big varlena
values and arrays in the protocol?
Yes.
- Heikki
--
Sent via pgsql-hackers
On 08/27/2014 08:05 PM, Fabien COELHO wrote:
[...]
Yeah, something like that. I don't think it would be necessary to set
statement_timeout, you can inject that in your script or postgresql.conf if
you want. I don't think aborting a transaction that's already started is
necessary either. You
(2014/08/25 15:48), Etsuro Fujita wrote:
(2014/08/15 6:18), Rukh Meski wrote:
Based on the feedback on my previous patch, I've separated only the
LIMIT part into its own feature. This version plays nicely with
inheritance. The intended use is splitting up big UPDATEs and DELETEs
into batches
Hello Heikki,
This now begs the question:
In --rate mode, shouldn't the reported transaction latency also be calculated
from the *scheduled* start time, not the time the transaction actually
started? Otherwise we're using two different definitions of latency, one
for the purpose of the
On 8/29/14 12:20 PM, Etsuro Fujita wrote:
The patch places limit-counting inside ModifyTable, and works well for
inheritance trees, but I'm not sure that that is the right way to go. I
think that this feature should be implemented in the way that we can
naturally extend it to the ORDER-BY-LIMIT
On 8/29/14 1:53 PM, I wrote:
This is just my personal opinion, but what I think should happen is:
1) We put the LIMIT inside ModifyTable like this patch does. This
doesn't prevent us from doing ORDER BY in the future, but helps numerous
people who today have to
Oops, looks like I didn't
On 08/26/2014 05:13 AM, Tatsuo Ishii wrote:
While looking into backend/storage/freespace/freespace.c, I noticed
that struct FSMAddress is passed to functions by value, rather than
reference. I thought our code practice is defining pointer to a struct
data and using the pointer for parameter
Greetings,
Because of the memory bug in xmlCopyNode, this is a new patch with
different method. In this patch, instead of using xmlCopyNode to bring the
namespace back, we added the required namespaces to the node before
dumping the node to string, and cleaning it up afterwards (because the same
Hi all,
Using a plugin producing binary output, I came across this error:
=# select data from pg_logical_slot_peek_changes('foo', NULL, NULL);
ERROR: 0A000: output plugin cannot produce binary output
LOCATION: pg_logical_slot_get_changes_guts, logicalfuncs.c:404
Shouldn't the error message be
Mark Kirkwood wrote:
On 29/08/14 08:56, Alvaro Herrera wrote:
Robert Haas wrote:
I agree that you might not like that. But you might not like having
the table vacuumed slower than the configured rate, either. My
impression is that the time between vacuums isn't really all that
negotiable
On 2014-08-29 22:42:46 +0900, Michael Paquier wrote:
Hi all,
Using a plugin producing binary output, I came across this error:
=# select data from pg_logical_slot_peek_changes('foo', NULL, NULL);
ERROR: 0A000: output plugin cannot produce binary output
LOCATION:
On Fri, Aug 29, 2014 at 10:48 PM, Andres Freund and...@2ndquadrant.com
wrote:
On 2014-08-29 22:42:46 +0900, Michael Paquier wrote:
Hi all,
Using a plugin producing binary output, I came across this error:
=# select data from pg_logical_slot_peek_changes('foo', NULL, NULL);
ERROR:
Jeff Davis pg...@j-davis.com writes:
I have a new approach to the patch which is to call a callback at each
block allocation and child contexts inherit the callback from their
parents. The callback could be defined to simply dereference and
increment its argument, which would mean block
On Thu, Aug 21, 2014 at 11:29:31PM -0400, Noah Misch wrote:
On Thu, Aug 21, 2014 at 09:18:22AM -0400, Andrew Dunstan wrote:
What's happening about this? Buildfarm animal jacana is consistently
red because of this.
If nobody plans to do the aforementioned analysis in the next 4-7 days, I
On 2014-08-29 23:07:44 +0900, Michael Paquier wrote:
On Fri, Aug 29, 2014 at 10:48 PM, Andres Freund and...@2ndquadrant.com
wrote:
On 2014-08-29 22:42:46 +0900, Michael Paquier wrote:
Hi all,
Using a plugin producing binary output, I came across this error:
=# select data from
On Fri, Aug 29, 2014 at 11:15 PM, Andres Freund and...@2ndquadrant.com
wrote:
No, a textual output plugin is *NOT* allowed to produce binary
output. That'd violate e.g. pg_logical_slot_peek_changes's return type
because it's only declared to return text.
A textual output plugin can call
Marko Tiikkaja ma...@joh.to writes:
The LIMIT part *has* to happen after the rows have been locked or it
will work very surprisingly under concurrency (sort of like how FOR
SHARE / FOR UPDATE worked before 9.0).
Good point.
So either it has to be inside
ModifyTable or the ModifyTable has
Heikki Linnakangas hlinnakan...@vmware.com writes:
On 08/26/2014 05:13 AM, Tatsuo Ishii wrote:
While looking into backend/storage/freespace/freespace.c, I noticed
that struct FSMAddress is passed to functions by value, rather than
reference.
There isn't really any strict coding rule on that.
On 2014-08-29 23:31:49 +0900, Michael Paquier wrote:
On Fri, Aug 29, 2014 at 11:15 PM, Andres Freund and...@2ndquadrant.com
wrote:
No, a textual output plugin is *NOT* allowed to produce binary
output. That'd violate e.g. pg_logical_slot_peek_changes's return type
because it's only
On 08/29/2014 10:15 AM, Noah Misch wrote:
On Thu, Aug 21, 2014 at 11:29:31PM -0400, Noah Misch wrote:
On Thu, Aug 21, 2014 at 09:18:22AM -0400, Andrew Dunstan wrote:
What's happening about this? Buildfarm animal jacana is consistently
red because of this.
If nobody plans to do the
Tom Lane t...@sss.pgh.pa.us wrote:
Kevin Grittner kgri...@ymail.com writes:
But the standard doesn't say anything about storing a time zone
*name* or *abbreviation* -- it requires that it be stored as UTC
with the *offset* (in hours and minutes). That makes it pretty
close to what we have
On Fri, Aug 29, 2014 at 4:03 PM, Kevin Grittner kgri...@ymail.com wrote:
It was actually rather disappointing to hear that we had a
conforming implementation and changed away from it circa the 7.2
release; and even more disturbing to hear that decision is still
being defended on the grounds
On Fri, Aug 29, 2014 at 11:39 PM, Andres Freund and...@2ndquadrant.com
wrote:
Yes. The output plugin declares whether it requires the *output method*
to support binary data. pg_logical_slot_peek_changes *can not* support
binary data because it outputs data as
text.
Kevin Grittner kgri...@ymail.com writes:
It was actually rather disappointing to hear that we had a
conforming implementation and changed away from it circa the 7.2
release;
That is not the case. The existing implementation is work that Tom
Lockhart did around 6.3 or so. It was called
On Fri, Aug 29, 2014 at 11:12 AM, Greg Stark [via PostgreSQL]
ml-node+s1045698n5816903...@n5.nabble.com wrote:
On Fri, Aug 29, 2014 at 4:03 PM, Kevin Grittner [hidden email]
http://user/SendEmail.jtp?type=nodenode=5816903i=0 wrote:
It was actually rather disappointing to hear that we had a
On Fri, Aug 29, 2014 at 4:19 PM, David G Johnston
david.g.johns...@gmail.com wrote:
And forcing people to change their data types to migrate to PostgreSQL is
undesirable IF our type is usefully equivalent to others in the majority of
situations - though I don't know if that is actually the
Prompted by a comment in the UPDATE/LIMIT thread, I saw Marko Tiikkaja
reference Tom's post
http://www.postgresql.org/message-id/1598.1399826...@sss.pgh.pa.us
which mentions the possibility of a different partitioning
implementation than what we have so far. As it turns out, I've been
thinking
Alvaro Herrera alvhe...@2ndquadrant.com writes:
[ partition sketch ]
In this design, partitions are first-class objects, not normal tables in
inheritance hierarchies. There are no pg_inherits entries involved at all.
Hm, actually I'd say they are *not* first class objects; the problem with
On Fri, Aug 29, 2014 at 4:56 PM, Alvaro Herrera
alvhe...@2ndquadrant.com wrote:
For scan plans, we need to prepare Append lists which are used to scan
for tuples in a partitioned relation. We can setup fake constraint
expressions based on the partitioning expressions, which let the planner
2014-08-29 18:35 GMT+02:00 Tom Lane t...@sss.pgh.pa.us:
Alvaro Herrera alvhe...@2ndquadrant.com writes:
[ partition sketch ]
In this design, partitions are first-class objects, not normal tables in
inheritance hierarchies. There are no pg_inherits entries involved at
all.
Hm, actually
Greg Stark st...@mit.edu writes:
And I think there are still unanswered questions about indexes.
One other interesting thought that occurs to me: are we going to support
UPDATEs that cause a row to belong to a different partition? If so, how
are we going to handle the update chain links?
Tom Lane wrote:
Greg Stark st...@mit.edu writes:
And I think there are still unanswered questions about indexes.
One other interesting thought that occurs to me: are we going to support
UPDATEs that cause a row to belong to a different partition? If so, how
are we going to handle the
Alvaro Herrera alvhe...@2ndquadrant.com writes:
Tom Lane wrote:
One other interesting thought that occurs to me: are we going to support
UPDATEs that cause a row to belong to a different partition? If so, how
are we going to handle the update chain links?
Bah, I didn't mention it? My
On 2014-08-29 13:15:16 -0400, Tom Lane wrote:
Alvaro Herrera alvhe...@2ndquadrant.com writes:
Tom Lane wrote:
One other interesting thought that occurs to me: are we going to support
UPDATEs that cause a row to belong to a different partition? If so, how
are we going to handle the update
Hi,
We have pg_get_serial_sequence() mapping (relation, colum) to the
sequence. What I'm missing right now is the inverse. I.e. given a
sequence tell me the owner.
describe.c has a query for that, and it's not too hard to write, but it
still seems 'unfriendly' not to provide it.
Does anybody
Andres Freund and...@2ndquadrant.com writes:
On 2014-08-29 13:15:16 -0400, Tom Lane wrote:
Hm. I certainly agree that it's a case that could be disallowed for a
first cut, but it'd be nice to have some clue about how we might allow it
eventually.
Not pretty, but we could set t_ctid to some
On Wed, Aug 27, 2014 at 6:51 PM, Kouhei Kaigai kai...@ak.jp.nec.com wrote:
I'd like to follow this direction, and start stripping the DDL support.
...please make it so.
The attached patch eliminates DDL support.
Instead of the new CREATE CUSTOM PLAN PROVIDER statement,
it adds an internal
Tom Lane wrote:
Alvaro Herrera alvhe...@2ndquadrant.com writes:
Tom Lane wrote:
One other interesting thought that occurs to me: are we going to support
UPDATEs that cause a row to belong to a different partition? If so, how
are we going to handle the update chain links?
Bah, I
On 2014-08-29 13:29:19 -0400, Tom Lane wrote:
Andres Freund and...@2ndquadrant.com writes:
On 2014-08-29 13:15:16 -0400, Tom Lane wrote:
Hm. I certainly agree that it's a case that could be disallowed for a
first cut, but it'd be nice to have some clue about how we might allow it
Hello Heikki,
[...] I would be fine with both.
After giving it some thought, ISTM better to choose consistency over
intuition, and have latency under throttling always defined wrt the
scheduled start time and not the actual start time, even if having a
latency of 1 ms for an OLTP load
Hi
I see a lot of warnings
[pavel@localhost postgresql]$ make all | grep warning
exec.c:280:2: warning: implicitly declaring library function 'strlcpy' with
type 'unsigned long (char *, const char *, unsigned long)'
strlcpy(link_buf, fname, sizeof(link_buf));
^
exec.c:280:2:
Andres Freund and...@2ndquadrant.com writes:
On 2014-08-29 13:29:19 -0400, Tom Lane wrote:
An actual fix would presumably involve adding a partition number to the
ctid chain field in tuples in partitioned tables. The reason I bring it
up now is that we'd have to commit to doing that (or at
Pavel Stehule pavel.steh...@gmail.com writes:
I see a lot of warnings
[pavel@localhost postgresql]$ make all | grep warning
exec.c:280:2: warning: implicitly declaring library function 'strlcpy' with
type 'unsigned long (char *, const char *, unsigned long)'
strlcpy(link_buf, fname,
On 08/29/2014 07:15 PM, Tom Lane wrote:
Alvaro Herrera alvhe...@2ndquadrant.com writes:
Tom Lane wrote:
One other interesting thought that occurs to me: are we going to support
UPDATEs that cause a row to belong to a different partition? If so, how
are we going to handle the update chain
Hi
I am looking to your last patch and I have a few questions, notes
1. I am thinking so reduction to only numeric types is not necessary -
although we can live without it - but there are lot of non numeric
categories: chars, date, ...
2. Still I strongly afraid about used searching method -
On 29.8.2014 16:12, Tom Lane wrote:
Jeff Davis pg...@j-davis.com writes:
I have a new approach to the patch which is to call a callback at each
block allocation and child contexts inherit the callback from their
parents. The callback could be defined to simply dereference and
increment its
On 08/29/2014 05:56 PM, Alvaro Herrera wrote:
Prompted by a comment in the UPDATE/LIMIT thread, I saw Marko Tiikkaja
reference Tom's post
http://www.postgresql.org/message-id/1598.1399826...@sss.pgh.pa.us
which mentions the possibility of a different partitioning
implementation than what we
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Le 28/08/2014 05:58, Michael Paquier a écrit :
On Thu, Aug 28, 2014 at 7:37 AM, Julien Rouhaud
julien.rouh...@dalibo.com wrote:
Attached v2 patch implements this approach. All the work is still
done in pg_stat_get_archiver, as I don't think
Hannu Krosing wrote:
Cross-partition indexes are especially needed if we want to allow putting
UNIQUE constraints on non-partition-key columns.
I'm not going to implement cross-partition indexes in the first patch.
They are a huge can of worms.
--
Álvaro Herrera
On Fri, Aug 29, 2014 at 11:56 AM, Alvaro Herrera
alvhe...@2ndquadrant.com wrote:
In this design, partitions are first-class objects, not normal tables in
inheritance hierarchies. There are no pg_inherits entries involved at all.
Whoa. I always assumed that table inheritance was a
On 8/28/14 9:01 AM, Kyotaro HORIGUCHI wrote:
I found this issue when trying per-pg_user (role) loading of
auto_analyze and some tweaking tool. It is not necessarily set by
the user by own, but the function to decide whether to load some
module by the session-user would be usable, at least, as
On 8/27/14 2:55 AM, Magnus Hagander wrote:
I think the easy way of doing that is to just create an xlog.tar file.
Since we already create base.tar and possibly n*tablespace.tar,
adding one more file shouldn't be a big problem, and would make such
an implementation much easier. Would be trivial
On Fri, Aug 29, 2014 at 10:34 PM, Peter Eisentraut pete...@gmx.net wrote:
On 8/27/14 2:55 AM, Magnus Hagander wrote:
I think the easy way of doing that is to just create an xlog.tar file.
Since we already create base.tar and possibly n*tablespace.tar,
adding one more file shouldn't be a big
On 8/29/14 11:27 AM, Greg Stark wrote:
You know... I wonder if we have enough leverage in the standards
committee these days that we could usefully push that direction
instead of being pushed around. The standard timestamp with time zone
is not very useful and I'm sure the standards committee
Heikki Linnakangas hlinnakan...@vmware.com wrote:
On 08/28/2014 12:03 AM, Kevin Grittner wrote:
Heikki Linnakangas hlinnakan...@vmware.com wrote:
I suggest adding a new hook to the ParseState struct, (p_rangevar_hook
?). The planner calls it whenever it sees a reference to a table, and
the
Hi Craig,
On Fri, Aug 29, 2014 at 10:17:17AM +0800, Craig Ringer wrote:
(...) It should also discuss the approach of storing a (instant
timestamptz, timezone text) or (instant timestampts, tzoffset
smallint) tuple for when unambiguous representation is required.
(I guess I just volunteered
On 2014-08-26 22:04:03 -0400, Robert Haas wrote:
That's all I see on a first-read through.
I think I fixed all of them. Thanks.
There might be other
issues, and I haven't checked through it in great detail for mundane
bugs, but generally, I favor pressing on relatively rapidly toward a
Andres Freund-3 wrote
Hi,
We have pg_get_serial_sequence() mapping (relation, colum) to the
sequence. What I'm missing right now is the inverse. I.e. given a
sequence tell me the owner.
describe.c has a query for that, and it's not too hard to write, but it
still seems 'unfriendly' not to
On 2014-08-29 17:55:38 -0700, David G Johnston wrote:
Andres Freund-3 wrote
Hi,
We have pg_get_serial_sequence() mapping (relation, colum) to the
sequence. What I'm missing right now is the inverse. I.e. given a
sequence tell me the owner.
describe.c has a query for that, and it's
Andres Freund-3 wrote
On 2014-08-29 17:55:38 -0700, David G Johnston wrote:
Andres Freund-3 wrote
pg_get_sequence_ownedby(...)
My problem is that that possibly be confused with the user owning the
sequence :/
Though as soon as that person reads the output their misunderstanding would
On Sat, Aug 30, 2014 at 12:56 AM, Alvaro Herrera
alvhe...@2ndquadrant.com wrote:
Prompted by a comment in the UPDATE/LIMIT thread, I saw Marko Tiikkaja
reference Tom's post
http://www.postgresql.org/message-id/1598.1399826...@sss.pgh.pa.us
which mentions the possibility of a different
On Wed, Aug 27, 2014 at 7:16 PM, Fujii Masao masao.fu...@gmail.com wrote:
The patch looks good to me. One minor comment is; probably you need to
update the tab-completion code.
Thanks for the review. I have updated the patch to support
tab-completion.
As this is a relatively minor change, I
67 matches
Mail list logo
