On 12/31/2013 09:55 AM, Tom Lane wrote:
Josh Berkus j...@agliodbs.com writes:
Tom,
There's an abbreviated version of this argument in the comments in
my proposed patch at
http://www.postgresql.org/message-id/11927.1384199...@sss.pgh.pa.us
What I'm hoping will happen next is that the
On Mon, Nov 10, 2014 at 10:48 AM, Josh Berkus j...@agliodbs.com wrote:
On 12/31/2013 09:55 AM, Tom Lane wrote:
Josh Berkus j...@agliodbs.com writes:
Tom,
There's an abbreviated version of this argument in the comments in
my proposed patch at
On 11/10/2014 10:59 AM, Jeff Janes wrote:
On Mon, Nov 10, 2014 at 10:48 AM, Josh Berkus j...@agliodbs.com wrote:
On 12/31/2013 09:55 AM, Tom Lane wrote:
Josh Berkus j...@agliodbs.com writes:
Tom,
There's an abbreviated version of this argument in the comments in
my proposed patch at
On 2014-11-10 10:48:24 -0800, Josh Berkus wrote:
On 12/31/2013 09:55 AM, Tom Lane wrote:
Josh Berkus j...@agliodbs.com writes:
Tom,
There's an abbreviated version of this argument in the comments in
my proposed patch at
http://www.postgresql.org/message-id/11927.1384199...@sss.pgh.pa.us
Josh Berkus j...@agliodbs.com writes:
On 11/10/2014 10:59 AM, Jeff Janes wrote:
On Mon, Nov 10, 2014 at 10:48 AM, Josh Berkus j...@agliodbs.com wrote:
Did this patch every make it in? Or did it hang waiting for verification?
It made it in:
commit 4162a55c77cbb54acb4ac442ef3565b813b9d07a
On 11/10/2014 11:11 AM, Tom Lane wrote:
Josh Berkus j...@agliodbs.com writes:
On 11/10/2014 10:59 AM, Jeff Janes wrote:
On Mon, Nov 10, 2014 at 10:48 AM, Josh Berkus j...@agliodbs.com wrote:
Did this patch every make it in? Or did it hang waiting for verification?
It made it in:
commit
On Mon, Nov 10, 2014 at 11:04 AM, Josh Berkus j...@agliodbs.com wrote:
On 11/10/2014 10:59 AM, Jeff Janes wrote:
On Mon, Nov 10, 2014 at 10:48 AM, Josh Berkus j...@agliodbs.com wrote:
Did this patch every make it in? Or did it hang waiting for
verification?
It made it in:
On 11/10/2014 12:13 PM, Jeff Janes wrote:
The related problem where the end rows are actually needed (e.g. ORDER
BY...LIMIT) has not been fixed.
My idea to fix that was to check if the row's creation-transaction was in
the MVCC snapshot (which just uses local memory) before checking if that
Josh Berkus j...@agliodbs.com writes:
Tom,
There's an abbreviated version of this argument in the comments in
my proposed patch at
http://www.postgresql.org/message-id/11927.1384199...@sss.pgh.pa.us
What I'm hoping will happen next is that the complainants will hot-patch
that and see if it
Tom,
There's an abbreviated version of this argument in the comments in
my proposed patch at
http://www.postgresql.org/message-id/11927.1384199...@sss.pgh.pa.us
What I'm hoping will happen next is that the complainants will hot-patch
that and see if it fixes their problems. We can't really
Josh Berkus j...@agliodbs.com writes:
What I'm hoping will happen next is that the complainants will hot-patch
that and see if it fixes their problems. We can't really determine
what to do without that information.
Unfortunately, the original reporter of this issue will not be available
for
Tom Lane t...@sss.pgh.pa.us wrote:
I wonder if we could ameliorate this problem by making
get_actual_variable_range() use SnapshotDirty
In that thread I claimed that a current MVCC snapshot was the
most appropriate thing, which it probably is;
If it reads from the end of the index, won't it
Kevin Grittner kgri...@ymail.com writes:
Tom Lane t...@sss.pgh.pa.us wrote:
I wonder if we could ameliorate this problem by making
get_actual_variable_range() use SnapshotDirty
In that thread I claimed that a current MVCC snapshot was the
most appropriate thing, which it probably is;
If it
Josh Berkus j...@agliodbs.com writes:
The explanation is in
http://archives.postgresql.org/message-id/20130910132133.GJ1024477%40alap2.anarazel.de
The referenced commit introduced a planner feature. Funnily you seem to
have been the trigger for it's introduction ;)
Oh, crap, the off the
On Wed, Sep 11, 2013 at 2:10 PM, Andres Freund and...@2ndquadrant.comwrote:
On 2013-09-11 15:06:23 -0400, Andrew Dunstan wrote:
One thing that this made me wonder is why we don't have
transaction_timeout,
or maybe transaction_idle_timeout.
Because it's harder than it sounds, at least if
All,
We've confirmed that this issue is caused by having long-running idle
transactions on the server. When we disabled their queueing system
(which prodiced hour-long idle txns), the progressive slowness went away.
Why that should affect 9.X far more strongly than 8.4, I'm not sure
about.
On 09/11/2013 02:35 PM, Josh Berkus wrote:
All,
We've confirmed that this issue is caused by having long-running idle
transactions on the server. When we disabled their queueing system
(which prodiced hour-long idle txns), the progressive slowness went away.
Why that should affect 9.X far
On 2013-09-11 15:06:23 -0400, Andrew Dunstan wrote:
On 09/11/2013 02:35 PM, Josh Berkus wrote:
All,
We've confirmed that this issue is caused by having long-running idle
transactions on the server. When we disabled their queueing system
(which prodiced hour-long idle txns), the
On 2013-09-11 11:35:45 -0700, Josh Berkus wrote:
All,
We've confirmed that this issue is caused by having long-running idle
transactions on the server. When we disabled their queueing system
(which prodiced hour-long idle txns), the progressive slowness went away.
Why that should affect
The explanation is in
http://archives.postgresql.org/message-id/20130910132133.GJ1024477%40alap2.anarazel.de
The referenced commit introduced a planner feature. Funnily you seem to
have been the trigger for it's introduction ;)
Oh, crap, the off the end of the index optimization?
It's the
On Tue, Sep 10, 2013 at 05:43:16AM +, Mel Llaguno wrote:
Bruce,
First of all, I'd like to thank you for taking some interest in this
issue. We'd love to migrate to the latest PG version, but this issue
is currently preventing us from doing so.
Regardless of the DB used (base application
On 2013-09-09 20:38:09 -0400, Andrew Dunstan wrote:
On 08/01/2013 03:20 PM, Jeff Janes wrote:
On Thu, Aug 1, 2013 at 10:58 AM, Josh Berkus j...@agliodbs.com wrote:
Amit, All:
So we just retested this on 9.3b2. The performance is the same as 9.1
and 9.2; that is, progressively worse as
Hi,
On 2013-09-09 20:38:09 -0400, Andrew Dunstan wrote:
On 08/01/2013 03:20 PM, Jeff Janes wrote:
On Thu, Aug 1, 2013 at 10:58 AM, Josh Berkus j...@agliodbs.com wrote:
Amit, All:
So we just retested this on 9.3b2. The performance is the same as 9.1
and 9.2; that is, progressively worse
On 09/10/2013 08:20 AM, Andres Freund wrote:
A backtrace for this would be useful. Alternatively you could recompile
postgres using -fno-omit-frame-pointer in CFLAGS and use perf record -g.
It's using a custom build, so this should be doable.
Any chance you have older prepared xacts,
On 2013-09-10 08:45:33 -0400, Andrew Dunstan wrote:
On 09/10/2013 08:20 AM, Andres Freund wrote:
A backtrace for this would be useful. Alternatively you could recompile
postgres using -fno-omit-frame-pointer in CFLAGS and use perf record -g.
It's using a custom build, so this should be
On 2013-09-10 15:21:33 +0200, Andres Freund wrote:
If I interpret things correctly you're using serializable? I guess there
is no chance to use repeatable read instead?
Err, that wouldn't help much. Read committed. That lets PGXACT-xmin advance
these days and thus might help to reduce the
On 09/10/2013 09:23 AM, Andres Freund wrote:
On 2013-09-10 15:21:33 +0200, Andres Freund wrote:
If I interpret things correctly you're using serializable? I guess there
is no chance to use repeatable read instead?
Err, that wouldn't help much. Read committed. That lets PGXACT-xmin advance
On 08/01/2013 03:20 PM, Jeff Janes wrote:
On Thu, Aug 1, 2013 at 10:58 AM, Josh Berkus j...@agliodbs.com wrote:
Amit, All:
So we just retested this on 9.3b2. The performance is the same as 9.1
and 9.2; that is, progressively worse as the test cycles go on, and
unacceptably slow compared to
On Mon, Sep 9, 2013 at 08:38:09PM -0400, Andrew Dunstan wrote:
On 08/01/2013 03:20 PM, Jeff Janes wrote:
On Thu, Aug 1, 2013 at 10:58 AM, Josh Berkus j...@agliodbs.com wrote:
Amit, All:
So we just retested this on 9.3b2. The performance is the same as 9.1
and 9.2; that is, progressively
: Monday, September 09, 2013 6:38 PM
To: Jeff Janes
Cc: Josh Berkus; Amit Kapila; pgsql-performance@postgresql.org
Subject: Re: [PERFORM] Performance bug in prepared statement binding in
9.2?
On 08/01/2013 03:20 PM, Jeff Janes wrote:
On Thu, Aug 1, 2013 at 10:58 AM, Josh Berkus j...@agliodbs.com wrote
On Tue, Sep 10, 2013 at 01:36:27AM +, Mel Llaguno wrote:
Let me clarify further - when we reconstruct our schema (without the
upgrade step) via a sql script, the problem still persists. Restoring
an upgraded DB which contains existing data into exactly the same
instance will correct the
] Performance bug in prepared statement binding in
9.2?
On Tue, Sep 10, 2013 at 01:36:27AM +, Mel Llaguno wrote:
Let me clarify further - when we reconstruct our schema (without the
upgrade step) via a sql script, the problem still persists. Restoring
an upgraded DB which contains existing data
On Tue, Sep 10, 2013 at 02:06:08AM +, Mel Llaguno wrote:
We're currently using an embedded PG 8.4.17 for our application. Our
PG 9.x tests consists of one of the following :
- Take a 8.4.17 DB which contains only our application schema and
required seed data and use pg_upgrade to create a
]
Performance bug in prepared statement binding in 9.2?
On Tue, Sep 10, 2013
at 02:06:08AM +, Mel Llaguno wrote:
We're currently using an embedded
PG 8.4.17 for our application. Our
PG 9.x tests consists of one of the
following :
- Take a 8.4.17 DB which contains only our application
schema
Amit, All:
So we just retested this on 9.3b2. The performance is the same as 9.1
and 9.2; that is, progressively worse as the test cycles go on, and
unacceptably slow compared to 8.4.
Some issue introduced in 9.1 is causing BINDs to get progressively
slower as the PARSEs BINDs get run
On Thu, Aug 1, 2013 at 10:58 AM, Josh Berkus j...@agliodbs.com wrote:
Amit, All:
So we just retested this on 9.3b2. The performance is the same as 9.1
and 9.2; that is, progressively worse as the test cycles go on, and
unacceptably slow compared to 8.4.
Some issue introduced in 9.1 is
On Thursday, May 30, 2013 11:36 PM Josh Berkus wrote:
Amit,
I think it might be because of choosing custom plan option due to
which it might be generating new plan during exec_bind_message().
exec_bind_message()-GetCachedPlan()-choose_custom_plan(). If it
chooses custom plan, then it will
When you say interactively, does it mean that you are using psql to test the
same?
Yes. Of course, on the psql command line, there's no separate BIND
step, just PREPARE and EXECUTE.
From the changes, it doesn't seem that any of such changes can cause the
problem you have seen.
No, but
On Thursday, May 30, 2013 5:45 AM Josh Berkus wrote:
Folks,
I'm seeing what may be a major performance bug in BIND in 9.2.4.
We have a client who has an application which uses
Tomcat+Hibernate+JDBC. They are in the process of upgrading this
application from 8.4.17 to 9.2.4. As part of
Amit,
I think it might be because of choosing custom plan option due to which it
might be generating new plan during exec_bind_message().
exec_bind_message()-GetCachedPlan()-choose_custom_plan(). If it chooses
custom plan, then it will regenerate the plan which can cause extra cost
This is why I'm asking them to run tests on 9.1. If 9.1 doesn't exhibit
this behavior, then customplan is liable to be at fault.
9.1 shows the same performance as 9.2. So it's not the custom plan thing.
--
Josh Berkus
PostgreSQL Experts Inc.
http://pgexperts.com
--
Sent via
Folks,
I'm seeing what may be a major performance bug in BIND in 9.2.4.
We have a client who has an application which uses
Tomcat+Hibernate+JDBC. They are in the process of upgrading this
application from 8.4.17 to 9.2.4. As part of this, they have been doing
performance testing, and 9.2 is
On 05/29/2013 05:14 PM, Josh Berkus wrote:
Here's timings overall for 29 test cycles (cycle 1 has been omitted).
As you can see, parse+execute times are pretty much constant, as are
application think times, but bind times vary quite a lot. In 8.4, the
29 cycles are constantly 4.5min to
43 matches
Mail list logo