On Fri, Oct 07, 2022 at 02:48:24PM -0700, Nathan Bossart wrote:
> Here is an attempt at adding such a test module.
Using an ereport(NOTICE) to show the data reported in the callback is
fine by me. How about making the module a bit more modular, by
passing as argument a regclass and building a
On Fri, Oct 07, 2022 at 07:59:08AM -0400, Robert Haas wrote:
> I hadn't noticed this thread before.
>
> I'm not sure whether this is properly considered a privilege check. It
> could even be an anti-privilege, if the pg_hba.conf line in question
> is maked "reject".
>
> I'm not taking the
Ajay P S writes:
> I am pretty new to the Postgres code base. I would like to know the
> difference between HeapTupleData and TupleTableSlot structures.
HeapTupleData is just a pointer to a concrete tuple. It exists
mainly because it's often convenient to pass around the tuple's
logical
Hi,
On 2022-10-07 12:00:56 -0700, Andres Freund wrote:
> On 2022-10-07 15:30:43 +0900, Kyotaro Horiguchi wrote:
> > The key point of this is this:
> >
> > +* XXX: I think there cannot actually be data from an older slot
> > +* here. After a crash we throw away the old stats data and if
Hi,
I am pretty new to the Postgres code base. I would like to know the
difference between HeapTupleData and TupleTableSlot structures.
Basically I am trying to understand some of the table access methods like
heap_insert, heap_getnext, heap_getnextslot etc where some accepts
Heaptuple as input
Em sex., 7 de out. de 2022 às 15:45, Ranier Vilela
escreveu:
> Em sex., 7 de out. de 2022 às 14:44, Tom Lane
> escreveu:
>
>> Ranier Vilela writes:
>> > Anyway, it's not *true* that collectMatchesForHeapRow deals
>> > only "true" and "false", once that key->entryRes is initialized with
>> >
Andres Freund writes:
> On 2022-10-07 20:35:58 -0400, Tom Lane wrote:
>> Andres Freund writes:
>>> Why are we even tracking PM_CHILD_UNUSED / PM_CHILD_ASSIGNED in shared
>>> memory?
>> Because those flags are set by the child processes too, cf
>> MarkPostmasterChildActive and
Hi,
On 2022-10-07 20:35:58 -0400, Tom Lane wrote:
> Andres Freund writes:
> > Why are we even tracking PM_CHILD_UNUSED / PM_CHILD_ASSIGNED in shared
> > memory?
>
> Because those flags are set by the child processes too, cf
> MarkPostmasterChildActive and MarkPostmasterChildInactive.
Only
Andres Freund writes:
> Why are we even tracking PM_CHILD_UNUSED / PM_CHILD_ASSIGNED in shared memory?
Because those flags are set by the child processes too, cf
MarkPostmasterChildActive and MarkPostmasterChildInactive.
> Are you thinking these should be backpatched?
I am, but I'm not
Hi,
On 2022-10-07 19:57:35 -0400, Tom Lane wrote:
> As I mentioned in another thread, I came across a reproducible
> situation in which a memory clobber in a child backend crashes
> the postmaster too, at least on FreeBSD/arm64. Needless to say,
> this is Not Cool.
Ugh.
> I've now traced down
As I mentioned in another thread, I came across a reproducible
situation in which a memory clobber in a child backend crashes
the postmaster too, at least on FreeBSD/arm64. Needless to say,
this is Not Cool. I've now traced down what is happening,
and it's this:
1. Careless coding in aset.c
Hi,
I stumbled over:
https://about.gitlab.com/blog/2021/09/29/why-we-spent-the-last-month-eliminating-postgresql-subtransactions/
I wonder if SAVEPOINT / subtransaction performance has been boosted since
the blog was written.
Cheers
On Fri, Oct 07, 2022 at 04:18:59PM -0400, Tom Lane wrote:
> There's another problem there, which is that buildfarm animals
> using -DENFORCE_REGRESSION_TEST_NAME_RESTRICTIONS will complain
> about role names that don't start with "regress_".
Huh, I hadn't noticed that one before. It looks like
On Fri, Oct 07, 2022 at 03:49:31PM +0900, Michael Paquier wrote:
> Perhaps there should be a module in src/test/modules/ to provide a
> short, still useful, example of what this could achieve?
Here is an attempt at adding such a test module.
--
Nathan Bossart
Amazon Web Services:
Mikhail Gribkov writes:
> Usually it's not a good idea to exit PG_TRY() block via return statement.
> Otherwise it would leave PG_exception_stack global variable in a wrong
> state and next ereport() will jump to some junk address.
Yeah, you can't return or goto out of the PG_TRY part.
>
Hello Zhihong,
> For the last question, please take a look at:
>
> #define MemSetAligned(start, val, len) \
>
> which is called by palloc0().
Oh, I totally missed that. Thanks for the heads up!
I'm attaching the new patch version, which contains both the fix
to the problem reported by Andres,
Nathan Bossart writes:
> On Fri, Oct 07, 2022 at 03:34:51PM +0900, Michael Paquier wrote:
>> Thanks. I would perhaps use names less generic than role{1,2,3} for
>> the roles or "role1" for the database name, but the logic looks
>> sound.
> Here is a new version with more descriptive role names.
On Fri, Oct 07, 2022 at 03:34:51PM +0900, Michael Paquier wrote:
> Thanks. I would perhaps use names less generic than role{1,2,3} for
> the roles or "role1" for the database name, but the logic looks
> sound.
Here is a new version with more descriptive role names.
--
Nathan Bossart
Amazon Web
I wrote:
>> Attached is a patch series that attempts to modernize our GUC
>> infrastructure, in particular removing the performance bottlenecks
>> it has when there are lots of GUC variables.
> Rebased over 0a20ff54f.
Here's a v3 rebased up to HEAD. The only real change is that I added
a couple
Hello Andres,
> cfbot shows that tests started failing with this version:
>
https://cirrus-ci.com/github/postgresql-cfbot/postgresql/commitfest/39/3822
> A full backtrace is at
https://api.cirrus-ci.com/v1/task/5354378189078528/logs/cores.log
Thanks for pointing this out. I had initially missed
On Fri, Oct 7, 2022 at 12:09 PM Israel Barth Rubio
wrote:
> Hello Zhihong,
>
> > + /* attribute is NOT to be copied from input */
> >
> > I think saying `is NOT copied from input` should suffice.
> >
> > + /* fieldno is 0-index and attnum is 1-index */
> >
> > 0-index ->
Hi,
On 2022-10-05 13:07:10 -0700, Andres Freund wrote:
> 0003: aix: Build SUBSYS.o using $(CC) -r instead of $(LD) -r
>
> This is the only direct use of $(LD), and xlc -r and gcc -r end up with the
> same set of symbols and similar performance (noise is high, so hard to say
> if
>
Hello Zhihong,
> + /* attribute is NOT to be copied from input */
>
> I think saying `is NOT copied from input` should suffice.
>
> + /* fieldno is 0-index and attnum is 1-index */
>
> 0-index -> 0-indexed
I have applied both suggestions, thanks! I'll submit a 4th version
Hi,
On 2022-10-07 15:30:43 +0900, Kyotaro Horiguchi wrote:
> At Fri, 7 Oct 2022 12:14:40 +0900, Masahiko Sawada
> wrote in
> > > What about if we go the other direction - simply remove the name from the
> > > stats entry at all. I don't actually think we need it anymore. Unless I am
> > >
Em sex., 7 de out. de 2022 às 14:44, Tom Lane escreveu:
> Ranier Vilela writes:
> > Anyway, it's not *true* that collectMatchesForHeapRow deals
> > only "true" and "false", once that key->entryRes is initialized with
> > GIN_FALSE not false.
>
> Look: GinTernaryValue is carefully designed so
On 2022-Oct-05, Alvaro Herrera wrote:
> Backpatching this to 12 shows yet another problem -- the topmost
> relation acquires additional FK constraints, not yet sure why. I think
> we must have fixed something in 13 that wasn't backpatched, but I can't
> remember what it is and whether it was
Em sex., 7 de out. de 2022 às 13:32, Robert Haas
escreveu:
> On Fri, Oct 7, 2022 at 11:40 AM Tom Lane wrote:
> > FWIW, I did take a look at this code, and I don't see any bug.
> > The entryRes[] array entries are indeed GinTernaryValue, but it's
> > obvious by inspection that
Ranier Vilela writes:
> Anyway, it's not *true* that collectMatchesForHeapRow deals
> only "true" and "false", once that key->entryRes is initialized with
> GIN_FALSE not false.
Look: GinTernaryValue is carefully designed so that it is
representation-compatible with bool, including that
Nikita Glukhov writes:
> On 30.09.2022, Tom Lane wrote:
>> I strongly recommend against having a new pg_proc column at all.
> I think the only way to avoid catalog modification (adding new columns
> to pg_proc or pg_type, introducing new function signatures etc.) and
> to avoid adding some
Hi hackers.
Andres Freund , 22 Eyl 2022 Per, 20:03 tarihinde şunu
yazdı:
> Now that meson has been merged, could you try to adjust this patch so it
> builds with meson?
>
Attached patch is the adjusted version to build with meson.
Add a commented-out trigger-type: manual and note in the commit
Em sex., 7 de out. de 2022 às 12:40, Tom Lane escreveu:
> Robert Haas writes:
> > On Thu, Oct 6, 2022 at 8:35 PM Ranier Vilela
> wrote:
> >> No Tom, unfortunately I don't have the knowledge to create a test with
> GIN_MAYBE values.
>
> > Well then don't post.
>
> > Basically what you're saying
pá 7. 10. 2022 v 16:03 odesílatel Julien Rouhaud
napsal:
> On Fri, Oct 07, 2022 at 07:26:08AM +0200, Pavel Stehule wrote:
> >
> > I checked this code again, and I don't think some refactoring is easy.
> > getFiltersFromFile is not duplicated. It is just probably badly named.
> >
> > These
On Fri, Oct 7, 2022 at 8:10 AM Peter Eisentraut
wrote:
> The only drawback in terms of code robustness is that someone adding a
> new shared object type would have to remember to add it to
> EventTriggerSupportsObjectType().
This doesn't seem like a good idea to me. If the function names give
On Fri, Oct 7, 2022 at 11:40 AM Tom Lane wrote:
> FWIW, I did take a look at this code, and I don't see any bug.
> The entryRes[] array entries are indeed GinTernaryValue, but it's
> obvious by inspection that matchPartialInPendingList only returns
> true or false, therefore
Robert Haas writes:
> On Thu, Oct 6, 2022 at 8:35 PM Ranier Vilela wrote:
>> No Tom, unfortunately I don't have the knowledge to create a test with
>> GIN_MAYBE values.
> Well then don't post.
> Basically what you're saying is that you suspect there might be a
> problem with this code but you
Peter Smith writes:
> WARNING: tables were not subscribed, you will have to run ALTER
> SUBSCRIPTION ... REFRESH PUBLICATION to subscribe the tables
> When I first encountered the above CREATE SUBSCRIPTION warning message
> I thought it was dubious-looking English...
> On closer inspection I
On 04.10.22 08:57, Amul Sul wrote:
On Tue, Oct 4, 2022 at 12:00 PM Peter Eisentraut
wrote:
On 16.05.22 10:27, Peter Eisentraut wrote:
Inspired by [0], I looked to convert more macros to inline functions.
Here is another one from the same batch of work that I somehow didn't
send in last
On Fri, Oct 07, 2022 at 07:26:08AM +0200, Pavel Stehule wrote:
>
> I checked this code again, and I don't think some refactoring is easy.
> getFiltersFromFile is not duplicated. It is just probably badly named.
>
> These routines are used from pg_dump, pg_dumpall and pg_restore. There are
>
On Thu, Oct 6, 2022 at 8:35 PM Ranier Vilela wrote:
> No Tom, unfortunately I don't have the knowledge to create a test with
> GIN_MAYBE values.
Well then don't post.
Basically what you're saying is that you suspect there might be a
problem with this code but you don't really know that and you
On 2022/10/07 18:43, bt22nakamorit wrote:
2022-10-07 16:59 Fujii Masao wrote:
s/subscriber/publisher ?
I did not understand what this means.
Sorry for confusing wording... I was just trying to say that walsender
is connected to a database of the publisher instead of subscriber.
Tom Lane wrote:
> > Currently, test/regress/sql/psql.sql doesn't AFAICS write anything
> > outside of stdout, but \g, \o, \copy need to write to external
> > files to be tested properly.
>
> Yeah, I don't think we can usefully test these in psql.sql, because
> file-system side effects
Hello,
I'm trying to make single-row mode and pipeline mode work together in
Psycopg using libpq. I think there is something wrong with respect to
the single-row mode flag, not being correctly reset, in some situations.
The minimal case I'm considering is (in a pipeline):
* send query 1,
*
We have in event_trigger.c two functions
EventTriggerSupportsObjectType() and EventTriggerSupportsObjectClass()
that return whether a given object type/class supports event triggers.
Maybe there was a real differentiation there once, but right now it
seems to me that *all* object types/classes
On Fri, Oct 7, 2022 at 2:03 PM Vik Fearing wrote:
> On 10/4/22 09:41, Peter Eisentraut wrote:
> > In PostgreSQL 10, we added identity columns, as an alternative to serial
> > columns (since 6.something). They mostly work the same. Identity
> > columns are SQL-conforming, have some more
On 10/4/22 09:41, Peter Eisentraut wrote:
In PostgreSQL 10, we added identity columns, as an alternative to serial
columns (since 6.something). They mostly work the same. Identity
columns are SQL-conforming, have some more features (e.g., overriding
clause), and are a bit more robust in
On Fri, Apr 1, 2022 at 6:07 PM Nathan Bossart wrote:
>
> 6198420 ensured that has_privs_of_role() is used for predefined roles,
> which means that the role inheritance hierarchy is checked instead of mere
> role membership. However, inheritance is still not respected for
> pg_hba.conf.
Hi Kurado Hayato,
> In your patch:
>
> ```
> + /* Simple case, we already have a primary key or a replica
> identity index */
> + idxoid = GetRelationIdentityOrPK(localrel);
> + if (OidIsValid(idxoid))
> + return idxoid;
> +
> + /*
> +* If index
>
> This is the wrong thread / CF entry. Please see
>
Yep, my fault. Sorry about that.
--
Best regards,
Maxim Orlov.
2022-10-07 17:16 Fujii Masao wrote:
The patch failed to be applied into the master cleanly. Could you
rebase it?
patching file src/bin/psql/common.c
Hunk #1 succeeded at 94 (offset 4 lines).
Hunk #2 succeeded at 104 (offset 4 lines).
Hunk #3 succeeded at 133 (offset 4 lines).
Hunk #4 succeeded
On Fri, Oct 7, 2022 at 19:15 Alvaro Herrera wrote:
> On 2022-Oct-07, Amit Langote wrote:
>
> > > Thanks for the heads up. Hmm, this I am not sure how to reproduce on
> > > my own, so I am currently left with second-guessing what may be going
> > > wrong on 32 bit machines with whichever of the
On 2022-Oct-07, Amit Langote wrote:
> > Thanks for the heads up. Hmm, this I am not sure how to reproduce on
> > my own, so I am currently left with second-guessing what may be going
> > wrong on 32 bit machines with whichever of the 4 patches.
> >
> > For now, I'll just post 0001, which I am
On Fri, Oct 7, 2022 at 6:26 PM Amit Langote wrote:
> On Sun, Oct 2, 2022 at 10:24 AM Andres Freund wrote:
> > On 2022-10-01 18:21:15 -0700, Andres Freund wrote:
> > > On 2022-09-29 18:18:10 +0900, Amit Langote wrote:
> > > > So, here's a final revision for today. Sorry for the noise.
> > >
> >
2022-10-07 16:59 Fujii Masao wrote:
s/subscriber/publisher ?
I did not understand what this means.
Thanks for the patch!
-
+ printf("fork child process\n");
+ printf(" am_walsender: %d\n", am_walsender);
+ printf(" am_db_walsender: %d\n",
On Sun, Oct 2, 2022 at 10:24 AM Andres Freund wrote:
> On 2022-10-01 18:21:15 -0700, Andres Freund wrote:
> > On 2022-09-29 18:18:10 +0900, Amit Langote wrote:
> > > So, here's a final revision for today. Sorry for the noise.
> >
> > This appears to fail on 32bit systems. Seems the new test is
On 2022/09/30 16:54, bt22nakamorit wrote:
2022-09-21 11:45 に Fujii Masao wrote:
We can execute the shell commands via psql in various ways
other than \! meta-command. For example,
* `command`
* \g | command
* \gx | command
* \o | command
* \w | command
* \copy ... program 'command'
On Fri, Oct 7, 2022 at 2:29 PM John Naylor wrote:
>
> On Fri, Sep 16, 2022 at 1:01 PM Masahiko Sawada wrote:
> > In addition to two patches, I've attached the third patch. It's not
> > part of radix tree implementation but introduces a contrib module
> > bench_radix_tree, a tool for radix tree
On 2022/10/06 22:30, bt22nakamorit wrote:
Hi,
When walsender process is evoked for logical replication, walsender is
connected to a database of the subscriber.
Naturally, ones would want the name of the connected database to show in the
entry of ps command for walsender.
In detail, running
Hi hackers,
I've found some odd lines in plpython-related code. These look to me like a
potential source of problems, but maybe I'm not fully aware of some nuances.
Usually it's not a good idea to exit PG_TRY() block via return statement.
Otherwise it would leave PG_exception_stack global
On Tue, Aug 02, 2022 at 04:49:19PM -0700, Nathan Bossart wrote:
> I've rebased this patch and will register it in the next commitfest
> shortly.
Perhaps there should be a module in src/test/modules/ to provide a
short, still useful, example of what this could achieve?
--
Michael
signature.asc
Hi,
I found that tag files generated by src/tools/make_ctags
doesn't include some keywords, that were field names of node
structs, for example norm_select in RestrictInfo. Such fields
are defined with pg_node_attr macro that was introduced
recently, like:
Selectivity norm_selec
On Thu, Oct 06, 2022 at 08:27:11PM -0700, Nathan Bossart wrote:
> On Fri, Oct 07, 2022 at 11:06:47AM +0900, Michael Paquier wrote:
>> Rather than putting that in a separate script, which means
>> initializing a new node, etc. could it be better to put that in
>> 001_password.pl instead? It would
At Fri, 7 Oct 2022 12:14:40 +0900, Masahiko Sawada
wrote in
> > What about if we go the other direction - simply remove the name from the
> > stats entry at all. I don't actually think we need it anymore. Unless I am
> > missing something right now - entirely possible! - the danger that
> >
On Tue, Sep 27, 2022 at 6:03 PM Etsuro Fujita wrote:
> I will review the patch a bit more, but I feel that it is
> in good shape.
One thing I noticed is this bit added to CopyMultiInsertBufferFlush()
to run triggers on the foreign table.
+ /* Run AFTER ROW INSERT triggers */
+
On Thursday, October 6, 2022 8:40 PM Kuroda, Hayato/黒田 隼人
wrote:
>
> Dear Hou,
>
> I put comments for v35-0001.
Thanks for the comments.
> 01. catalog.sgml
>
> ```
> + Controls how to handle the streaming of in-progress transactions:
> + f = disallow streaming of in-progress
64 matches
Mail list logo
