On Tue, Feb 2, 2021 at 9:03 PM Ajin Cherian wrote:
> I am sorry, my above steps were not correct. I think the reason for
> the failure I was seeing were some other steps I did prior to this. I
> will recreate this and update you with the appropriate steps.
The correct steps are as follows:
Publ
On Wed, Feb 3, 2021 at 6:13 PM Michael Paquier wrote:
>
> On Tue, Feb 02, 2021 at 07:23:37PM +0530, Amit Kapila wrote:
> > I don't mind changing to your proposed text but I think the current
> > wording is also okay and seems clear to me.
>
> Reading that again, I still find the word "transient" t
vignesh C writes:
> On Mon, Feb 1, 2021 at 11:04 AM Bharath Rupireddy
> wrote:
>> Are these superuser and permission checks enough from a security
>> standpoint that we don't expose some sensitive information to the
>> user?
> This will just print the backtrace of the current backend. Users
> ca
On Tue, Feb 02, 2021 at 07:23:37PM +0530, Amit Kapila wrote:
> I don't mind changing to your proposed text but I think the current
> wording is also okay and seems clear to me.
Reading that again, I still find the word "transient" to be misleading
in this context. Any extra opinions?
--
Michael
On Mon, Feb 1, 2021 at 11:04 AM Bharath Rupireddy
wrote:
>
> On Mon, Feb 1, 2021 at 6:14 AM Bharath Rupireddy
> wrote:
> > On Fri, Jan 29, 2021 at 7:10 PM vignesh C wrote:
> > > > 4) How about following
> > > > + errmsg("must be a superuser to print backtrace
> > > > of backe
On Wed, Feb 03, 2021 at 03:37:39PM +0900, Michael Paquier wrote:
> index 627b36300c..4ee3951ca0 100644
> --- a/doc/src/sgml/ref/reindex.sgml
> +++ b/doc/src/sgml/ref/reindex.sgml
> @@ -293,8 +311,30 @@ REINDEX [ ( class="parameter">option [, ...] ) ] { IN
> respectively. Each partition of the
Thanks Bharath for your comments.
On Mon, Feb 1, 2021 at 6:14 AM Bharath Rupireddy
wrote:
>
> On Fri, Jan 29, 2021 at 7:10 PM vignesh C wrote:
> > > 4) How about following
> > > + errmsg("must be a superuser to print backtrace
> > > of backend process")));
> > > instead of
>
On Tue, Feb 02, 2021 at 10:32:19AM +0900, Michael Paquier wrote:
> On Mon, Feb 01, 2021 at 06:28:57PM +0300, Alexey Kondratov wrote:
> > Hm, IIUC, REINDEX CONCURRENTLY doesn't use either of them. It directly uses
> > index_create() with a proper tablespaceOid instead of
> > SetRelationTableSpace().
Hi,
Can we have a new function, say pg_postgres_pid(), to return
postmaster PID similar to pg_backend_pid()? At times, it will be
difficult to use OS level commands to get the postmaster pid of a
backend to which it is connected. It's even worse if we have multiple
postgres server instances runnin
On Thu, Jan 28, 2021 at 10:07 AM japin wrote:
> Attaching v3 patches, please consider these for further review.
I think we can add a commitfest entry for this feature, so that the
patches will be tested on cfbot. Ignore if done already.
With Regards,
Bharath Rupireddy.
EnterpriseDB: http://www.e
On Mon, Jan 25, 2021 at 10:32 PM vignesh C wrote:
> > I mean it doesn’t seem right to disallow to create the subscription if
> > the publisher doesn't exist, and my reasoning was even though the
> > publisher exists while creating the subscription you might drop it
> > later right?. So basically,
On Tue, Feb 2, 2021 at 9:45 AM Fujii Masao wrote:
> One merit of keep_connections that I found is that we can use it even
> when connecting to the older PostgreSQL that doesn't support
> idle_session_timeout. Also it seems simpler to use keep_connections
> rather than setting idle_session_timeout
On Wed, Feb 3, 2021 at 2:07 AM Robert Haas wrote:
>
> Even more review comments, still looking mostly at 0001:
>
> If there's a reason why parallel_schedule is arranging to run the
> compression test in parallel with nothing else, the comment in that
> file should explain the reason. If there isn'
On Wed, 03 Feb 2021 at 02:13, Tom Lane wrote:
> japin writes:
>> Here's my analysis:
>> 1) In the explain_ExecutorEnd(), it will create a ExplainState on SQL
>> function
>> memory context, which is a long-lived, cause the memory grow up.
>
> Yeah, agreed. I think people looking at this have a
On Wed, Feb 3, 2021 at 1:34 PM Amit Kapila wrote:
>
> On Wed, Feb 3, 2021 at 6:38 AM Peter Smith wrote:
> >
> > On Wed, Feb 3, 2021 at 12:26 AM Amit Kapila wrote:
> > >
> > > On Tue, Feb 2, 2021 at 3:31 PM Peter Smith wrote:
> > > >
> > > > After seeing Ajin's test [ac0202] which did a DROP TAB
On Wed, Feb 3, 2021 at 6:38 AM Peter Smith wrote:
>
> On Wed, Feb 3, 2021 at 12:26 AM Amit Kapila wrote:
> >
> > On Tue, Feb 2, 2021 at 3:31 PM Peter Smith wrote:
> > >
> > > After seeing Ajin's test [ac0202] which did a DROP TABLE, I have also
> > > tried a simple test where I do a DROP TABLE w
At Wed, 3 Feb 2021 01:26:41 +, "tsunakawa.ta...@fujitsu.com"
wrote in
> (41)
> +void
> +PQtraceEx(PGconn *conn, FILE *debug_port, int flags)
> +{
> + if (conn == NULL)
> + return;
> ...
> + if (!debug_port)
> + return;
>
> The if should better be as follows t
From: 'Alvaro Herrera'
> > + conn->fe_msg->num_fields != DEF_FE_MSGFIELDS)
> The rationale for that second condition is this: if the memory allocated
> is the initial size, we don't free memory, because it would just be
> allocated of the same size next time, and that size is not very b
On 2021/02/03 1:49, Fujii Masao wrote:
On 2021/02/02 22:00, torikoshia wrote:
On 2021-01-25 23:44, Fujii Masao wrote:
Another comment is; Doesn't the change of MyProc->waitStart need the
lock table's partition lock? If yes, we can do that by moving
LWLockRelease(partitionLock) just after t
On Wed, Feb 3, 2021 at 12:24 AM Amit Kapila wrote:
> The problem here is that we are allowing to drop the table when table
> synchronization is still in progress and then we don't have any way to
> know the corresponding slot or origin. I think we can try to drop the
> slot and origin as well but
(39)
+ of tracing. If (flags &
PQTRACE_OUTPUT_TIMESTAMPS) is
+ true, then timestamp is not printed with each message.
The flag name (OUTPUT) and its description (not printed) doesn't match.
I think you can use less programmatical expression like "If
flags contains PQTRACE_OUTPUT_TIME
On Wed, Feb 3, 2021 at 12:26 AM Amit Kapila wrote:
>
> On Tue, Feb 2, 2021 at 3:31 PM Peter Smith wrote:
> >
> > After seeing Ajin's test [ac0202] which did a DROP TABLE, I have also
> > tried a simple test where I do a DROP TABLE with very bad timing for
> > the tablesync worker. It seems that d
> For v3_0003-reloption-parallel_dml-src.patch :
> + table_close(rel, NoLock);
> Since the rel would always be closed, it seems the return value from
> RelationGetParallelDML() can be assigned to a variable, followed by call to
> table_close(), then the return statement.
Thanks for the commen
Hi,
Attaching v5 patches with the changes:
* rebase the code on the greg's latest parallel insert patch
* fix some code style.
Please consider it for further review.
Best regards,
Houzj
v5_0004-reloption-parallel_dml-test-and-doc.patch
Description: v5_0004-reloption-parallel_dml-tes
Hi,
On 2021-02-02 20:19:19 -0300, Alvaro Herrera wrote:
> > > @@ -1169,6 +1169,8 @@ XLogInsertRecord(XLogRecData *rdata,
> > > {
> > > /* advance global request to include new block(s)
> > > */
> > > pg_atomic_monotonic_advance_u64(&XLogCtl->LogwrtRqst.Write,
> > >
I wrote:
> * It would now be possible to remove the PGNSP and PGUID kluges
> entirely in favor of plain BKI_LOOKUP references to pg_namespace
> and pg_authid. The catalog header usage would get a little
> more verbose: BKI_DEFAULT(PGNSP) becomes BKI_DEFAULT(pg_catalog)
> and BKI_DEFAULT(PGUID) bec
Hello,
So I addressed about half of your comments in this version merely by
fixing silly bugs. The problem I had which I described as
"synchronization fails" was one of those silly bugs.
So in further thinking, it seems simpler to make only LogwrtResult
atomic, and leave LogwrtRqst as currently,
On Thu, 2021-01-28 at 18:22 +, Jacob Champion wrote:
> = Proposal =
>
> I propose that every auth method should store the string it uses to
> identify a user -- what I'll call an "authenticated identity" -- into
> one central location in Port, after authentication succeeds but before
> any pg_
"Joel Jacobson" writes:
> On Tue, Feb 2, 2021, at 17:00, Tom Lane wrote:
>> Appreciate the review! Please confirm if you agree with above
>> analysis.
> Yes, I agree with the analysis.
Cool, thanks. I've pushed it now.
regards, tom lane
On Sat, Aug 29, 2020 at 8:13 PM Thomas Munro wrote:
> Currently we don't run the bgwriter process during crash recovery.
> I've CCed Simon and Heikki who established this in commit cdd46c76.
> Based on that commit message, I think the bar to clear to change the
> policy is to show that it's useful
On Mon, Feb 1, 2021 at 12:15 PM Heikki Linnakangas wrote:
> Thanks. I fixed it slightly differently, and also changed LocalToUtf()
> to follow the same pattern, even though LocalToUtf() did not have the
> same bug.
Looks good to me.
> I added a bunch of tests for various built-in conversions.
N
"Joel Jacobson" writes:
> Doc: improve documentation of oid columns that can be zero.
Since this is pretty closely tied to the catalog-foreign-key work,
I went ahead and reviewed/pushed it. The zero notations now match
up with what we'd found in the other thread.
regards
Even more review comments, still looking mostly at 0001:
If there's a reason why parallel_schedule is arranging to run the
compression test in parallel with nothing else, the comment in that
file should explain the reason. If there isn't, it should be added to
a parallel group that doesn't have th
On Tue, Feb 2, 2021 at 6:28 AM Victor Yegorov wrote:
> I really like this idea!
Cool!
> It resembles the approach used in bottom-up index deletion, block-based
> accounting provides a better estimate for the usefulness of the operation.
It does resemble bottom-up index deletion, in one importan
Hello, Peter.
> AFAICT that's not true, at least not in any practical sense. See the
> comment in the middle of MarkBufferDirtyHint() that begins with "If we
> must not write WAL, due to a relfilenode-specific...", and see the
> "Checksums" section at the end of src/backend/storage/page/README. Th
On Tue, Feb 2, 2021, at 17:00, Tom Lane wrote:
>No, I think it's correct as-is (and this is one reason that I chose to
>have two separate FK entries for cases like this). confrelid can be
>zero in rows that are not FK constraints. However, such a row must
>also have empty confkey. The above entr
út 2. 2. 2021 v 20:36 odesílatel Tom Lane napsal:
> Pavel Stehule writes:
> > When I fixed one plpgsql_check issue, I found another plpgsql issue. Now
> we
> > have field nstatements that hold a number of plpgsql statements in
> > function. Unfortunately I made an error when I wrote this functio
Pavel Stehule writes:
> When I fixed one plpgsql_check issue, I found another plpgsql issue. Now we
> have field nstatements that hold a number of plpgsql statements in
> function. Unfortunately I made an error when I wrote this functionality and
> for FOR statements, this counter is incremented 2
japin writes:
> Here's my analysis:
> 1) In the explain_ExecutorEnd(), it will create a ExplainState on SQL function
> memory context, which is a long-lived, cause the memory grow up.
Yeah, agreed. I think people looking at this have assumed that the
ExecutorEnd hook would automatically be runni
Hi
When I fixed one plpgsql_check issue, I found another plpgsql issue. Now we
have field nstatements that hold a number of plpgsql statements in
function. Unfortunately I made an error when I wrote this functionality and
for FOR statements, this counter is incremented 2x. Higher number than a
rea
On 2021/02/02 22:00, torikoshia wrote:
On 2021-01-25 23:44, Fujii Masao wrote:
Another comment is; Doesn't the change of MyProc->waitStart need the
lock table's partition lock? If yes, we can do that by moving
LWLockRelease(partitionLock) just after the change of
MyProc->waitStart, but which
"Joel Jacobson" writes:
> I could only find one minor error,
> found by running the regression-tests,
> and then using the query below to compare "is_opt"
> with my own "zero_values" in my tool
> that derives its value from pg_catalog content.
> ...
> I therefore think is_opt should be changed to
On Mon, Feb 1, 2021 at 07:47:57PM -0500, Bruce Momjian wrote:
> On Mon, Feb 1, 2021 at 06:31:32PM -0500, Stephen Frost wrote:
> > * Bruce Momjian (br...@momjian.us) wrote:
> > > The purpose of cluster file encryption is to prevent users with read
> > > access to the directories used to store
On Tue, Feb 2, 2021 at 11:17 PM Tom Lane wrote:
>
> Dunno about refobjversion; I have my doubts that putting that info in
> pg_depend was a sane design choice at all. But from what I understand
> of it, wouldn't all deps on a given object necessarily have the same
> version?
Correct, assuming of
Peter Eisentraut writes:
> I do wonder, however, under what circumstances code would be put into a
> situation where it would add the exact same dependency again, and also
> under what circumstances it would add a dependency between the same
> objects but a different deptype, and how that would
A race with KeepFileRestoredFromArchive() can cause a restartpoint to fail, as
seen once on the buildfarm[1]. The attached patch adds a test case; it
applies atop the "stop events" patch[2]. We have two systems for adding
long-term pg_wal directory entries. KeepFileRestoredFromArchive() adds the
On Tue, Feb 2, 2021 at 6:49 PM Peter Eisentraut
wrote:
>
> I do wonder, however, under what circumstances code would be put into a
> situation where it would add the exact same dependency again, and also
> under what circumstances it would add a dependency between the same
> objects but a differen
вт, 2 февр. 2021 г. в 05:27, Peter Geoghegan :
> And now here is the second thing I thought of, which is much better:
>
> Sometimes 1% of the dead tuples in a heap relation will be spread
> across 90%+ of the pages. With other workloads 1% of dead tuples might
> be highly concentrated, and appear
On 2021-Jan-29, tsunakawa.ta...@fujitsu.com wrote:
> (30)
> +/*
> + * Deallocate FE/BE message tracking memory. We only do this because
> + * FE message can grow arbitrarily large, and skip it in the initial state,
> + * because it's likely pointless.
> + */
> +void
> +pqTraceUninit(PGconn *conn)
On Mon, Nov 30, 2020 at 02:09:10PM -0500, Tom Lane wrote:
> Justin Pryzby writes:
> > I think this is waiting on me to provide a patch for the contrib/ modules
> > with
> > update script removing the SQL operators, and documentating their
> > deprecation.
>
> Right. We can remove the SQL opera
On Tue, Feb 2, 2021 at 10:49 AM Michael Paquier wrote:
>
> On Tue, Feb 02, 2021 at 10:38:31AM +1100, Peter Smith wrote:
> > PSA a trivial patch to correct what seems like a typo in the tablesync
> > comment.
>
> - * subscribed tables and their state. Some transient state during data
> - *
Hi all,
Thank you Kirk san for creating the v14 patch.
I update the patch. I fixed all of Tsunakawa san's review comments.
I am trying to solve three bugs. Two bags were pointed out by Alvaro san
in a previous e-mail. And I found one bug.
> From: Alvaro Herrera
> Sent: Friday, January 22, 202
On Tue, Feb 2, 2021 at 3:31 PM Peter Smith wrote:
>
> After seeing Ajin's test [ac0202] which did a DROP TABLE, I have also
> tried a simple test where I do a DROP TABLE with very bad timing for
> the tablesync worker. It seems that doing this can cause the sync
> worker's MyLogicalRepWorker->reli
On Tue, Feb 2, 2021 at 11:35 AM Ajin Cherian wrote:
>
> Another failure I see in my testing
>
The problem here is that we are allowing to drop the table when table
synchronization is still in progress and then we don't have any way to
know the corresponding slot or origin. I think we can try to d
On 2021-01-25 23:44, Fujii Masao wrote:
Another comment is; Doesn't the change of MyProc->waitStart need the
lock table's partition lock? If yes, we can do that by moving
LWLockRelease(partitionLock) just after the change of
MyProc->waitStart, but which causes the time that lwlock is being held
t
On Tue, Feb 2, 2021, at 12:34, Euler Taveira wrote:
>You should probably email: webmaster (at) postgresql (dot) org
Thanks, done.
/Joel
On Tue, Feb 2, 2021, at 8:38 AM, japin wrote:
> In 0003 patch, function GetPublicationRelationQuals() has been defined, but it
> never used. So why should we define it?
Thanks for taking a look again. It is an oversight. It was introduced in an
attempt to refactor ALTER PUBLICATION SET TABLE. In A
On 02/02/2021 12:35, Peter Eisentraut wrote:
I had a bit of trouble parsing the error message "every hash partition
modulus must be a factor of the next larger modulus", so I went into the
code, added some comments and added errdetail messages for each case. I
think it's a bit clearer now.
Yea
On Tue, Feb 2, 2021 at 7:36 PM Peter Eisentraut
wrote:
> I had a bit of trouble parsing the error message "every hash partition
> modulus must be a factor of the next larger modulus", so I went into the
> code, added some comments and added errdetail messages for each case. I
> think it's a bit c
On Tue, 02 Feb 2021 at 19:16, japin wrote:
> On Tue, 02 Feb 2021 at 13:02, Michael Paquier wrote:
>> On Mon, Feb 01, 2021 at 04:11:50PM -0300, Euler Taveira wrote:
>>> After the commit 3696a600e2, the last patch does not apply cleanly. I'm
>>> attaching another version to address the documentat
On Tue, Feb 2, 2021, at 5:13 AM, Joel Jacobson wrote:
> I've tried to login to the CF interface a couple of times now, but seems to
> have lost my password.
> I've tried to use the "Password reset" form [1], but I don't get any email.
> The email is correct, because when I try to re-register it sa
On Tue, Feb 2, 2021, at 2:19 AM, Michael Paquier wrote:
> On Tue, Feb 02, 2021 at 10:38:31AM +1100, Peter Smith wrote:
> > PSA a trivial patch to correct what seems like a typo in the tablesync
> > comment.
>
> - * subscribed tables and their state. Some transient state during data
> - *
On Tue, 02 Feb 2021 at 13:02, Michael Paquier wrote:
> On Mon, Feb 01, 2021 at 04:11:50PM -0300, Euler Taveira wrote:
>> After the commit 3696a600e2, the last patch does not apply cleanly. I'm
>> attaching another version to address the documentation issues.
>
> I have bumped into this thread, a
On 2021-01-22 16:42, Tom Lane wrote:
pg_depend
pg_shdepend
Yeah, this is noted in the patch's own regression tests.
Wouldn't it be possible to add primary keys to these two as well?
Neither of the existing indexes is suitable, not being unique.
We could imagine adding a unique index across t
On 2021-02-01 15:24, Tom Lane wrote:
Peter Eisentraut writes:
On 2021-01-30 22:56, Tom Lane wrote:
Hmm, shouldn't there have been a catversion bump in there?
I suppose yes on the grounds that it introduces something new in a
freshly initdb-ed database. But I thought it wasn't necessary bec
I had a bit of trouble parsing the error message "every hash partition
modulus must be a factor of the next larger modulus", so I went into the
code, added some comments and added errdetail messages for each case. I
think it's a bit clearer now.
From e7f392e2f8950236a22f007cc3aed36729da22e1 Mon
On Tue, Feb 2, 2021 at 7:40 PM Amit Kapila wrote:
>
> On Tue, Feb 2, 2021 at 10:34 AM Ajin Cherian wrote:
> >
> > On Mon, Feb 1, 2021 at 11:26 PM Amit Kapila wrote:
> > > I have updated the patch to display WARNING for each of the tablesync
> > > slots during DropSubscription. As discussed, I ha
After seeing Ajin's test [ac0202] which did a DROP TABLE, I have also
tried a simple test where I do a DROP TABLE with very bad timing for
the tablesync worker. It seems that doing this can cause the sync
worker's MyLogicalRepWorker->relid to become invalid.
In my test this caused a stack trace wi
On Jan 28, 2021, at 3:31 PM, Michael Paquier
mailto:mich...@paquier.xyz>> wrote:
On Wed, Jan 27, 2021 at 09:18:48AM +, Paul Guo wrote:
Second one is use copy_file_range() for the local rewind case to replace
read()+write().
This introduces copy_file_range() check and HAVE_COPY_FILE_RANGE s
On Tue, 02 Feb 2021 at 07:12, Jeff Janes wrote:
> On Mon, Feb 1, 2021 at 6:09 PM Jeff Janes wrote:
>
>>
>>
>> create or replace function gibberish(int) returns text language SQL as $_$
>> select left(string_agg(md5(random()::text),),$1) from
>> generate_series(0,$1/32) $_$;
>>
>> create tabl
From: Andrey Lepikhov
> Of course, you can rebase it.
Thank you. I might modify the basic part to incorporate my past proposal about
improving the layering or modularity related to ri_useMultiInsert. (But I may
end up giving up due to lack of energy.)
Also, I might defer working on the exten
Hi,
At a customer we came across a curious plan (see attached testcase).
Given the testcase we see that the outer semi join tries to join the
outer with the inner table id columns, even though the middle table id
column is also there. Is this expected behavior?
The reason i'm asking is two-f
Hi
rebase and set PK for pg_variable table
Regards
Pavel
schema-variables-20210202.patch.gz
Description: application/gzip
On Tue, Feb 2, 2021 at 10:34 AM Ajin Cherian wrote:
>
> On Mon, Feb 1, 2021 at 11:26 PM Amit Kapila wrote:
> > I have updated the patch to display WARNING for each of the tablesync
> > slots during DropSubscription. As discussed, I have moved the drop
> > slot related code towards the end in Alte
On 2/2/21 11:57, tsunakawa.ta...@fujitsu.com wrote:
Hello, Andrey-san,
From: Tang, Haiying
Sometimes before i suggested additional optimization [1] which can
additionally speed up COPY by 2-4 times. Maybe you can perform the
benchmark for this solution too?
...
But the patch no longer appli
>
> I've had a further look at this, and this walker function is doing a lot
> of work recursing the parse tree, and I'm not sure that it reliably retrieves
> the information that we;re looking for, for all cases of different SQL
> queries. Unless it can be made much more efficient and specific to
On 2021/01/27 14:08, Masahiko Sawada wrote:
On Wed, Jan 27, 2021 at 10:29 AM Fujii Masao
wrote:
You fixed some issues. But maybe you forgot to attach the latest patches?
Yes, I've attached the updated patches.
Thanks for updating the patch! I tried to review 0001 and 0002 as the
self-
Hi Euler,
I've tried to login to the CF interface a couple of times now, but seems to
have lost my password.
I've tried to use the "Password reset" form [1], but I don't get any email.
The email is correct, because when I try to re-register it says it's taken.
Not sure who I should ask for help.
On Mon, Feb 1, 2021 at 8:19 PM Hou, Zhijie wrote:
>
> Hi,
>
> When developing the reloption patch, I noticed some issues in the patch.
>
> 1).
> > - Reduce Insert parallel-safety checks required for some SQL, by noting
> > that the subquery must operate on a relation (check for RTE_RELATION in
> >
79 matches
Mail list logo
