Re: DDL result is lost by CREATE DATABASE with WAL_LOG strategy

On Fri, Feb 17, 2023 at 02:35:30PM -0800, Nathan Bossart wrote: > I'm happy to move this new test to wherever folks think it should go. I'll > look around to see if I can find a better place, too. I think that src/test/recovery/ is the best fit, because this stresses a code path for WAL replay on

Re: Missing free_var() at end of accum_sum_final()?

On Mon, Feb 20, 2023, at 08:38, Michael Paquier wrote: > Perhaps you should register this patch to the commit of March? Here > it is: > https://commitfest.postgresql.org/42/ Thanks, done. /Joel

Re: "out of relcache_callback_list slots" after multiple calls to pg_logical_slot_get_binary_changes

Good catch! At Sun, 19 Feb 2023 02:40:31 +, "shiy.f...@fujitsu.com" wrote in > init_rel_sync_cache()), but they are not unregistered when it shutdowns. So, > after multiple calls to the function, MAX_RELCACHE_CALLBACKS is exceeded. This > is mentioned in the following comment. > > /*

Re: Support logical replication of DDLs

On Fri, Feb 17, 2023 at 1:13 PM Zheng Li wrote: > > > > I've implemented a prototype to allow replicated objects to have the > > > same owner from the publisher in > > > v69-0008-Allow-replicated-objects-to-have-the-same-owner-from.patch. > > > > > > > I also think it would be a helpful addition f

Re: Use pg_pwritev_with_retry() instead of write() in dir_open_for_write() to avoid partial writes?

On Mon, Feb 20, 2023 at 11:03 AM Michael Paquier wrote: > > > I'm inclined to go with my version, with the argument order swapped to > > Bharath's order. > > Okay. That's fine by me. I ran some tests on my dev system [1] and I don't see much difference between v3 and v4. So, +1 for v3 patch (+ a

Re: bitscan forward/reverse on Windows

On Wed, Feb 8, 2023 at 3:14 PM John Naylor wrote: > > 0001 adds asserts to the existing coding. > > 0002 adds MSVC support. Tests pass on CI, but it's of course possible that there is some bug that prevents hitting the fastpath. I've mostly used the platform specific types, so some further cleanup

Re: Allow tailoring of ICU locales with custom rules

On 14.02.23 17:53, Laurenz Albe wrote: On Mon, 2023-02-06 at 22:16 +0100, Peter Eisentraut wrote: Right.  Here is a new patch with this fixed. Thanks. I played some more with it, and still are still some missing odds and ends: - There is a new option ICU_RULES to CREATE DATABASE, but it is n

Re: pg_walinspect memory leaks

On Sat, Feb 18, 2023 at 5:07 AM Jeff Davis wrote: > > On Thu, 2023-02-16 at 18:00 +0530, Bharath Rupireddy wrote: > > I'm attaching the patches here. For HEAD, I'd > > want to be a bit defensive and use the temporary memory context for > > pg_get_wal_fpi_info() too. > > I don't see why we shouldn'

Re: SQL/JSON revisited

Hi Amit and Andrew, Regarding not squashing [PATCH v3 11/11] Proposed reworking of SQL/JSON documentaion, here is exactly what Tom Lane wrote in the comment to commit 47046763c3: Use consistently for things that are in fact names of parameters (including OUT parameters), reserving for

Re: wrong query result due to wang plan

On Mon, Feb 20, 2023 at 3:01 AM Tom Lane wrote: > We still have to fix process_implied_equality though, and in that > context we do have all the SpecialJoinInfos, so the strip-the-outer-joins > fix seems to work. See attached. Yeah, process_implied_equality is also broken for variable-free cla

Re: Introduce list_reverse() to make lcons() usage less inefficient

On Fri, 17 Feb 2023 at 16:35, David Rowley wrote: > > On Fri, 17 Feb 2023 at 13:23, Andres Freund wrote: > > But wouldn't an even cheaper way here be to iterate over the children in > > reverse order when match_partition_order_desc? We can do that efficiently > > now. Looks like we don't have a r

Re: File API cleanup

On 23.12.22 09:33, Peter Eisentraut wrote: On 01.12.22 09:25, Peter Eisentraut wrote: Here are a couple of patches that clean up the internal File API and related things a bit: Here are two follow-up patches that clean up some stuff related to the earlier patch set.  I suspect these are all h

Re: Wrong query results caused by loss of join quals

On Mon, Feb 20, 2023 at 4:56 AM Tom Lane wrote: > * I suspect the other use of get_common_eclass_indexes, in > have_relevant_eclass_joinclause, is broken as well. It seems have_relevant_joinclause is broken for the presented case. It does not get a change to call have_relevant_eclass_joinclaus

RE: Allow logical replication to copy tables in binary format

On Thu, Feb 16, 2023 8:48 PM Amit Kapila wrote: > > On Mon, Jan 30, 2023 at 4:19 PM Melih Mutlu > wrote: > > > > Logical replication between different types like int and smallint is > > already not > working properly on HEAD too. > > Yes, the scenario you shared looks like working. But you didn

Re: Reduce timing overhead of EXPLAIN ANALYZE using rdtsc?

Hi! On 2/14/23 13:48, David Geier wrote: It still fails. I'll get Cirrus-CI working on my own Github fork so I can make sure it really compiles on all platforms before I submit a new version. It took some time until Cirrus CI allowed me to run tests against my new GitHub account (there's a

Re: Killing off removed rels properly

Hello Tom, 13.02.2023 21:39, Tom Lane wrote: Richard Guo writes: On Sat, Feb 11, 2023 at 4:50 AM Tom Lane wrote: I think it's time to clean this up by removing the rel from the planner data structures altogether. The attached passes check-world, and if it does trigger any problems I would s

Re: Missing free_var() at end of accum_sum_final()?

On Mon, 20 Feb 2023 at 08:03, Joel Jacobson wrote: > > On Mon, Feb 20, 2023, at 08:38, Michael Paquier wrote: > > Perhaps you should register this patch to the commit of March? Here > > it is: > > https://commitfest.postgresql.org/42/ > > Thanks, done. > I have been testing this a bit, and I get

Re: File* argument order, argument types

On Sat, Feb 18, 2023 at 6:23 AM Andres Freund wrote: > > Hi, > > While trying to add additional File* functions (FileZero, FileFallocat) I went > back and forth about the argument order between "amount" and "offset". > > We have: > > extern int FilePrefetch(File file, off_t offset, off_t amount, u

Re: Allow logical replication to copy tables in binary format

Hi, Hayato Kuroda (Fujitsu) , 20 Şub 2023 Pzt, 10:12 tarihinde şunu yazdı: > Dear Melih, > > Thank you for updating the patch. Before reviewing, I found that > cfbot have not accepted v8 patch [1]. > Thanks for letting me know. Attached the fixed version of the patch. Best, -- Melih Mutlu Micr

Re: Allow logical replication to copy tables in binary format

Amit Kapila , 16 Şub 2023 Per, 15:47 tarihinde şunu yazdı: > On Mon, Jan 30, 2023 at 4:19 PM Melih Mutlu > wrote: > >> 8. Note that the COPY binary format isn't portable across platforms > >> (Windows to Linux for instance) or major versions > >> https://www.postgresql.org/docs/devel/sql-copy.htm

Re: File descriptors in exec'd subprocesses

I had missed one: the "watch" end of the postmaster pipe also needs FD_CLOEXEC. From 36f8ed2406307f8b1578ae85510c5a07718e1ea8 Mon Sep 17 00:00:00 2001 From: Thomas Munro Date: Mon, 20 Feb 2023 23:26:36 +1300 Subject: [PATCH v3 1/3] Don't leak descriptors into subprograms. Open data and WAL files

I wish libpq add get_unresolv_host() and set_unresolv_host_ip() function

I want to resolve the domain name asynchronously before connecting to the target; one way is parse sql connect parameters string; like  if ("host=" in conn_params) and ("hostaddr=" not in conn_params):    get_domain_name_and_resolv_and_add_hostaddr_to_conn_params_string() Another way is to add i

Re: meson: Non-feature feature options

Hi, I added SSL and UUID patches. UUID patch has two different fixes: 1 - v1-0002-meson-Refactor-UUID-option.patch: Adding 'auto' choice to 'uuid' combo option. 2 - v1-0002-meson-Refactor-UUID-option-with-uuid_library.patch: Making 'uuid' feature option and adding new 'uuid_library' combo op

Re: dynamic result sets support in extended query protocol

On 31.01.23 12:07, Peter Eisentraut wrote: Applying this patch, your test queries seem to work correctly. Great! This is quite WIP, especially because there's a couple of scenarios uncovered by tests that I'd like to ensure correctness about, but if you would like to continue adding tests for

Re: File* argument order, argument types

On 18.02.23 01:52, Andres Freund wrote: I don't know what to actually propose. I guess the least bad I can see is to pick one type & argument order that we document to be the default, with a caveat placed above the functions not following the argument order. Order wise, I think we should choose

Re: some namespace.c refactoring

On 15.02.23 19:04, Alvaro Herrera wrote: That said, I think most of this code is invoked for DDL, where performance is not so critical; probably just fixing get_object_property_data to not be so naïve would suffice. Ok, I'll look into that. Queries are another matter. I can't think of a way

Re: some namespace.c refactoring

On 15.02.23 06:04, Tom Lane wrote: I have very serious concerns first about whether it even preserves the existing semantics, and second about whether there is a performance penalty. We can work out the performance issues, but what are your concerns about semantics?

Re: psql \watch 2nd argument: iteration count

On 17.02.23 00:33, Andrey Borodin wrote: From time to time I want to collect some stats from locks, activity and other stat views into one table from different time points. In this case the \watch psql command is very handy. However, it's not currently possible to specify the number of times a q

Re: Add WAL read stats to pg_stat_wal

On Fri, Feb 17, 2023 at 12:41 AM Andres Freund wrote: > > On 2023-02-16 23:39:00 +0530, Bharath Rupireddy wrote: > > While working on [1], I was in need to know WAL read stats (number of > > times and amount of WAL data read from disk, time taken to read) to > > measure the benefit. I had to write

Re: Add support for unit "B" to pg_size_pretty()

On Mon, Feb 20, 2023 at 07:44:15AM +0100, Peter Eisentraut wrote: > This patch adds support for the unit "B" to pg_size_pretty(). This makes it It seems like what it actually does is to support "B" in pg_size_bytes() - is that what you meant ? pg_size_pretty() already supports "bytes", so this d

Re: psql \watch 2nd argument: iteration count

Greetings, * Peter Eisentraut (peter.eisentr...@enterprisedb.com) wrote: > On 17.02.23 00:33, Andrey Borodin wrote: > > From time to time I want to collect some stats from locks, activity > > and other stat views into one table from different time points. In > > this case the \watch psql command

Re: SQL/JSON revisited

Op 20-02-2023 om 08:35 schreef Amit Langote: Rebased again over queryjumble overhaul. Hi, But the following statement is a problem. It does not crash but it goes off, half-freezing the machine, and only comes back after fanatic Ctrl-C'ing. select json_query(jsonb '[3,4]', '$[*]' retur

Re: ICU locale validation / canonicalization

On 10.02.23 18:53, Jeff Davis wrote: To represent ICU locale strings in the catalog consistently, we have two choices, which as far as I can tell are equivalent: 1. ICU format Locale IDs. These are more readable, and still specified (albeit non-standard). 2. BCP47 language tags. These are stand

Re: Add WAL read stats to pg_stat_wal

On Mon, Feb 20, 2023 at 10:51 AM Kyotaro Horiguchi wrote: > > At Thu, 16 Feb 2023 11:11:38 -0800, Andres Freund wrote > in > > I wonder if we should keep the checkpointer around for longer. If we have > > checkpointer signal postmaster after it wrote the shutdown checkpoint, > > postmaster could

Re: Move defaults toward ICU in 16?

On 17.02.23 21:43, Jeff Davis wrote: select row_number() over (order by nazev collate "cs-x-icu"), nazev from obce except select row_number() over (order by nazev collate "cs_CZ"), nazev from obce; returns a not empty set. So minimally for Czech collate, an index rebuild is necessary Yes, that'

Re: windows CI failing PMSignalState->PMChildFlags[slot] == PM_CHILD_ASSIGNED

18.02.2023 23:09, Andres Freund wrote: Hi, On 2023-02-18 18:00:00 +0300, Alexander Lakhin wrote: 18.02.2023 04:06, Andres Freund wrote: Maybe it's just the backend started for the money test has got the same PID (5948) that the backend for the name test had? I somehow mashed name and money int

Re: Improving inferred query column names

On 11.02.23 20:24, Andres Freund wrote: Not useful column names: SELECT SUM(reads), SUM(writes) FROM pg_stat_io; column names: sum, sum So i often end up manually writing: SELECT SUM(reads) AS sum_reads, SUM(writes) AS sum_writes, ... FROM pg_stat_io; Of course we can't infer useful column name

Re: pg_init_privs corruption.

Greetings, * Tom Lane (t...@sss.pgh.pa.us) wrote: > Floris Van Nee writes: > > This is as far as I can see the same case as what I reported a few years > > ago here: > > https://www.postgresql.org/message-id/flat/1574068566573.13088%40Optiver.com#488bd647ce6f5d2c92764673a7c58289 > > There was a

Re: Improving inferred query column names

On Mon, Feb 20, 2023 at 8:08 AM Peter Eisentraut < peter.eisentr...@enterprisedb.com> wrote: > On 11.02.23 20:24, Andres Freund wrote: > > > > I think on a green field it'd be clearly better to do something like the > > above. What does give me pause is that it seems quite likely to break > > exi

Re: ExecRTCheckPerms() and many prunable partitions (checkAsUser)

On 2023-Feb-20, Amit Langote wrote: > On Fri, Feb 17, 2023 at 9:02 PM Alvaro Herrera > wrote: > > I tried a few things for a new test case, but I was unable to find > > anything useful. Maybe an intermediate view, I thought; no dice. > > Maybe one with a security barrier would do? Anyway, for

pg_crc32c_armv8.c:35:9: error: implicit declaration of function '__crc32cb' is invalid in C99

I suddenly (?, IIRC a few days ago this worked fine) started getting the following error while trying to following while building from source on Apple M1 This is on REL_13_9 but happens to all other releases too. In file included from pg_crc32c_armv8.c:17: /Library/Developer/CommandLineTools/usr

Re: "out of relcache_callback_list slots" after multiple calls to pg_logical_slot_get_binary_changes

Kyotaro Horiguchi writes: > I'm pretty sure that everytime an output plugin is initialized on a > process, it installs the same set of syscache/relcache callbacks each > time. Do you think we could simply stop duplicate registration of > those callbacks by using a static boolean? It would be far

Re: XID formatting and SLRU refactorings (was: Add 64-bit XIDs into PostgreSQL 15)

Hi, > OK, here is the patchset v53 where I mostly modified the commit > messages. It is explicitly said that 0001 modifies the WAL records and > why we decided to do it in this patch. Additionally any mention of > 64-bit XIDs is removed since it is not guaranteed that the rest of the > patches are

Re: ExecRTCheckPerms() and many prunable partitions (checkAsUser)

Alvaro Herrera writes: > On 2023-Feb-20, Amit Langote wrote: >> One more thing we could try is come up with a postgres_fdw test case, >> because it uses the RelOptInfo.userid value for remote-costs-based >> path size estimation. But adding a test case to contrib module's >> suite test a core plan

Re: pg_crc32c_armv8.c:35:9: error: implicit declaration of function '__crc32cb' is invalid in C99

Markur Sens writes: > I suddenly (?, IIRC a few days ago this worked fine) started getting the > following error while trying to following while building from source on Apple > M1 > This is on REL_13_9 but happens to all other releases too. > In file included from pg_crc32c_armv8.c:17: > /Libra

Incorrect command tag row count for MERGE with a cross-partition update

Playing around with MERGE some more, I noticed that the command tag row count is wrong if it does a cross-partition update: CREATE TABLE target (a int, b int) PARTITION BY LIST (b); CREATE TABLE target_p1 PARTITION OF target FOR VALUES IN (1); CREATE TABLE target_p2 PARTITION OF target FOR VALUES

Re: pg_crc32c_armv8.c:35:9: error: implicit declaration of function '__crc32cb' is invalid in C99

> On 20 Feb 2023, at 5:47 PM, Tom Lane wrote: > > Markur Sens mailto:markurs...@gmail.com>> writes: >> I suddenly (?, IIRC a few days ago this worked fine) started getting the >> following error while trying to following while building from source on >> Apple M1 >> This is on REL_13_9 but hap

Assert failure with MERGE into partitioned table with RLS

As part of the MERGE RETURNING patch I noticed a suspicious Assert() in ExecInitPartitionInfo() that looked like it needed updating for MERGE. After more testing, I can confirm that this is indeed a pre-existing bug, that can be triggered using MERGE into a partitioned table that has RLS enabled (

Re: Add WAL read stats to pg_stat_wal

Hi, On 2023-02-20 14:21:39 +0900, Kyotaro Horiguchi wrote: > I totally agree that it will be useful, but I'm not quite sure how > checkpointer would be able to let postmaster know about that state > without requiring access to shared memory. SendPostmasterSignal(PMSIGNAL_SHUTDOWN_CHECKPOINT_COMPL

Re: Allow tailoring of ICU locales with custom rules

Peter Eisentraut wrote: [patch v5] Two quick comments: - pg_dump support need to be added for CREATE COLLATION / DATABASE - there doesn't seem to be a way to add rules to template1. If someone wants to have icu rules and initial contents to their new databases, I think they need to crea

Re: add PROCESS_MAIN to VACUUM

On Fri, Jan 13, 2023 at 03:30:15PM -0800, Nathan Bossart wrote: > On Fri, Jan 13, 2023 at 03:24:09PM -0800, Jeff Davis wrote: > > For completeness, did you consider CLUSTER and REINDEX options as well? > > I have not, but I can put together patches for those as well. Are you planning to do that h

Re: Killing off removed rels properly

Alexander Lakhin writes: > After this change the following query triggers an assert: > CREATE TABLE tt (tid integer PRIMARY KEY) PARTITION BY LIST (tid); > CREATE TABLE ttp PARTITION OF tt DEFAULT; > CREATE TABLE st (sid integer); > MERGE INTO tt USING st ON tt.tid = st.sid WHEN NOT MATCHED THEN

Re: Add index scan progress to pg_stat_progress_vacuum

Thanks! > I think PROGRESS_VACUUM_INDEXES_TOTAL and > PROGRESS_VACUUM_INDEXES_PROCESSED are better for consistency. The rest > looks good to me. Took care of that in v25. Regards -- Sami Imseih Amazon Web Services v25-0001-Add-2-new-columns-to-pg_stat_progress_vacuum.-Th.patch Descript

Re: Killing off removed rels properly

Hmm, there's something else going on here. After getting rid of the assertion failure, I see that the plan looks like # explain MERGE INTO tt USING st ON tt.tid = st.sid WHEN NOT MATCHED THEN INSERT VALUES (st.sid); QUERY PLAN -

Re: add PROCESS_MAIN to VACUUM

On Mon, Feb 20, 2023 at 10:31:11AM -0600, Justin Pryzby wrote: > On Fri, Jan 13, 2023 at 03:30:15PM -0800, Nathan Bossart wrote: >> On Fri, Jan 13, 2023 at 03:24:09PM -0800, Jeff Davis wrote: >> > For completeness, did you consider CLUSTER and REINDEX options as well? >> >> I have not, but I can p

Re: SQL/JSON revisited

Hi, On 2023-02-20 16:35:52 +0900, Amit Langote wrote: > Subject: [PATCH v4 03/10] SQL/JSON query functions > +/* > + * Evaluate a JSON error/empty behavior result. > + */ > +static Datum > +ExecEvalJsonBehavior(JsonBehavior *behavior, bool *is_null) > +{ > + *is_null = false; > + > + switc

Re: Killing off removed rels properly

I wrote: > Hmm, there's something else going on here. After getting rid of the > assertion failure, I see that the plan looks like > # explain MERGE INTO tt USING st ON tt.tid = st.sid WHEN NOT MATCHED THEN > INSERT > VALUES (st.sid); > QUERY PLAN

Re: Missing free_var() at end of accum_sum_final()?

Hi, I found another small but significant improvement of the previous patch: else if (ndigits < var->buf_len) { -memset(var->buf, 0, var->buf_len); +var->buf[0] = 0; var->digits = var->buf + 1; var->ndigits = ndigits; } We don't need to set all buf elements to zero, only the fir

Re: Ignoring BRIN for HOT updates (was: -udpates seems broken)

Hi, On Sun, 19 Feb 2023 at 16:04, Tomas Vondra wrote: > > Hi, > > On 2/19/23 02:03, Matthias van de Meent wrote: > > On Thu, 16 Jun 2022 at 15:05, Tomas Vondra > > wrote: > >> > >> I've pushed the revert. Let's try again for PG16. > > > > As we discussed in person at the developer meeting, here'

Re: Performance issues with parallelism and LIMIT

Hi, On 2/8/23 11:42, Tomas Vondra wrote: On 2/1/23 14:41, David Geier wrote: Yeah, this is a pretty annoying regression. We already can hit poor behavior when matching rows are not distributed uniformly in the tables (which is what LIMIT costing assumes), and this makes it more likely to hit si

Re: Reducing System Allocator Thrashing of ExecutorState to Alleviate FDW-related Performance Degradations

Hi, On 2023-02-17 09:52:01 -0800, Andres Freund wrote: > On 2023-02-17 17:26:20 +1300, David Rowley wrote: > Random note: > > I wonder if we should having a bitmap (in an int) in front of aset's > freelist. In a lot of cases we incur plenty cache misses, just to find the > freelist bucket empty.

Re: meson: Optionally disable installation of test modules

On 09.02.23 16:30, Nazir Bilal Yavuz wrote: On 2/8/23 13:30, Peter Eisentraut wrote: If you feel that your patch is ready, please add it to the commit fest. I look forward to reviewing it. Thanks! Commit fest entry link: https://commitfest.postgresql.org/42/4173/ I tested this a bit. It

Re: psql \watch 2nd argument: iteration count

> > > Also, if we do not provide a timespan, 2 seconds are selected. But if > > > we provide an incorrect argument - 1 second is selected. > > > PFA the patch that adds iteration count argument and makes timespan > > > argument more consistent. > > > > That should probably be fixed. > > And should

Missing cases from SPI_result_code_string()

Another one noticed from the MERGE RETURNING patch -- the switch statement in SPI_result_code_string() is missing cases for SPI_OK_TD_REGISTER and SPI_OK_MERGE. The SPI_OK_TD_REGISTER case goes back all the way, so I suppose it should be back-patched to all supported branches, though evidently thi

Re: meson: Non-feature feature options

On 20.02.23 13:33, Nazir Bilal Yavuz wrote: I added SSL and UUID patches. UUID patch has two different fixes: 1 - v1-0002-meson-Refactor-UUID-option.patch: Adding 'auto' choice to 'uuid' combo option. 2 - v1-0002-meson-Refactor-UUID-option-with-uuid_library.patch: Making 'uuid' feature optio

Re: improving user.c error messages

On Mon, Feb 20, 2023 at 08:54:48AM +0100, Peter Eisentraut wrote: > I'm concerned about the loose use of "privilege" here. A privilege is > something I can grant. So if someone doesn't have the "REPLICATION > privilege", as in the above example, I would expect to be able to do "GRANT > REPLICATIO

Re: meson: Non-feature feature options

Hi, On Mon, 20 Feb 2023 at 21:53, Peter Eisentraut wrote: > > But what does uuid_library=auto do? Which one does it pick? This is > not a behavior we currently have, is it? Yes, we didn't have that behavior before. It checks uuid libs by the order of 'e2fs', 'bsd' and 'ossp'. It uses the first

Re: Killing off removed rels properly

I wrote: >> It looks like we're somehow triggering the elide-a-left-join code >> when we shouldn't? So the reason why we see this with a partitioned target table and not with a regular target table reduces to this bit in preprocess_targetlist: /* * For non-inherited UPDATE/DELETE/MERGE,

Re: Reducing System Allocator Thrashing of ExecutorState to Alleviate FDW-related Performance Degradations

On Tue, 21 Feb 2023 at 07:30, Andres Freund wrote: > 2) We should introduce an API mcxt.c API to perform allocations that the >caller promises not to individually free. It's not just pfree. Offhand, there's also repalloc, GetMemoryChunkSpace and GetMemoryChunkContext too. I am interested in

Re: pg_walinspect memory leaks

On Mon, 2023-02-20 at 15:17 +0530, Bharath Rupireddy wrote: > Similarly, the loops in GetXLogSummaryStats() too > don't palloc any memory, so no memory leak. Break on palloc in gdb in that loop and you'll see a palloc in CStringGetTextDatum(name). In general, you should expect *GetDatum() to pall

Re: meson: Non-feature feature options

> On 20 Feb 2023, at 19:53, Peter Eisentraut > wrote: > I would rename the ssl_type variable to ssl_library, so that if we ever > expose that as an option, it would be consistent with uuid_library. +1, ssl_library is a better name. -- Daniel Gustafsson

Re: Missing cases from SPI_result_code_string()

Dean Rasheed writes: > Another one noticed from the MERGE RETURNING patch -- the switch > statement in SPI_result_code_string() is missing cases for > SPI_OK_TD_REGISTER and SPI_OK_MERGE. Ugh. Grepping around, it looks like pltcl_process_SPI_result is missing a case for SPI_OK_MERGE as well.

Re: Move defaults toward ICU in 16?

On Mon, 2023-02-20 at 15:55 +0100, Peter Eisentraut wrote: > I'm confused.  We are not going to try to change existing databases > to a > different collation provider during pg_upgrade, are we? No, certainly not. I interpreted Pavel's comments as a comparison of ICU and libc in general and not s

Re: meson: Non-feature feature options

Hi, On 2023-02-20 19:53:53 +0100, Peter Eisentraut wrote: > On 20.02.23 13:33, Nazir Bilal Yavuz wrote: > > I added SSL and UUID patches. UUID patch has two different fixes: > > > > 1 - v1-0002-meson-Refactor-UUID-option.patch: Adding 'auto' choice to > > 'uuid' combo option. > > > > 2 - v1-0002

Re: Reducing System Allocator Thrashing of ExecutorState to Alleviate FDW-related Performance Degradations

Hi, On 2023-02-21 08:33:22 +1300, David Rowley wrote: > On Tue, 21 Feb 2023 at 07:30, Andres Freund wrote: > > 2) We should introduce an API mcxt.c API to perform allocations that the > >caller promises not to individually free. > > It's not just pfree. Offhand, there's also repalloc, > Get

Re: meson: Optionally disable installation of test modules

On 2023-02-20 19:43:56 +0100, Peter Eisentraut wrote: > I don't think any callers try to copy a directory source, so the > shutil.copytree() stuff isn't necessary. I'd like to use it for installing docs outside of the normal install target. Of course we could add the ability at a later point, but

Re: Missing cases from SPI_result_code_string()

On Mon, 20 Feb 2023 at 19:39, Tom Lane wrote: > > Ugh. Grepping around, it looks like pltcl_process_SPI_result > is missing a case for SPI_OK_MERGE as well. > Yes, I was about to post a patch for that too. That's the last case that I found, looking around. Regards, Dean

Allow MERGE to be executed from PL/Tcl

Another one noticed in the MERGE RETURNING patch -- this allows PL/Tcl to execute MERGE (i.e., don't fail when SPI returns SPI_OK_MERGE). I'm not sure if anyone uses PL/Tcl anymore, but it's a trivial fix, probably not worth a regression test case. Regards, Dean diff --git a/src/pl/tcl/pltcl.c b/s

Re: pgbench: using prepared BEGIN statement in a pipeline could cause an error

Found one last problem: if you do "-f foobar.sql -M prepared" in that order, then the prepare fails because the statement names would not be assigned when the file is parsed. This coding only supported doing "-M prepared -f foobar.sql", which funnily enough is the only one that PostgreSQL/Cluster.

Re: Killing off removed rels properly

I wrote: > But while I'm looking at this, 3c569049b seems kind of broken on > its own terms. Why is it populating so little of the IndexOptInfo > for a partitioned index? I realize that we're not going to directly > plan anything using such an index, but not populating fields like > sortopfamily

Re: Time delayed LR (WAS Re: logical replication restrictions)

Here are some review comments for patch v3-0001. (I haven't looked at the test code yet) == Commit Message 1. If the subscription sets min_send_delay parameter, an apply worker passes the value to the publisher as an output plugin option. And then, the walsender will delay the transaction se

Re: Missing free_var() at end of accum_sum_final()?

Hi, My apologies, it seems my email didn't reach the list, probably due to the benchmark plot images being to large. Here is the email again, but with URLs to the images instead, and benchmark updated with results for the 0005-fixed-buf.patch. -- On Mon, Feb 20, 2023, at 12:32, Dean Rasheed wrot

Re: Silent overflow of interval type

Nick Babadzhanian writes: > Please find attached the v2 of the said patch with the tests added. Pushed with light editing (for instance, I don't think interval.sql is the place to test timestamp operators, even if the result is an interval). > I don't know how the decision on > backpatching is m

Re: [PoC] Federated Authn/z with OAUTHBEARER

Greetings, * mahendrakar s (mahendrakarfo...@gmail.com) wrote: > The "issuer" field has been removed to align with the RFC > implementation - https://www.rfc-editor.org/rfc/rfc7628. > This patch "v6" is a single patch to support the OAUTH BEARER token > through psql connection string. > Below fl

Re: Proposal: Support custom authentication methods using hooks

Greetings, * Andrey Chudnovsky (achudnovs...@gmail.com) wrote: > The thread link is here: > https://www.postgresql.org/message-id/flat/CABkiuWo4fJQ7dhqgYLtJh41kpCkT6iXOO8Eym3Rdh5tx2RJCJw%40mail.gmail.com#f94c36969a68a07c087fa9af0f5401e1 Thanks for pointing out the updates on that thread, I've res

Re: Weird failure with latches in curculio on v15[

On Sun, Feb 19, 2023 at 08:06:24PM +0530, Robert Haas wrote: > I mean, my idea was to basically just have one big callback: > ArchiverModuleMainLoopCB(). Which wouldn't return, or perhaps, would > only return when archiving was totally caught up and there was nothing > more to do right now. And the

Re: pg_walinspect memory leaks

On Mon, Feb 20, 2023 at 11:34:03AM -0800, Jeff Davis wrote: > Committed to 16 with the changes to GetXLogSummaryStats() as well. > Committed unmodified version of your 15 backport. Thank you! Thanks for taking care of the FPI code path, Jeff! -- Michael signature.asc Description: PGP signature

Re: improving user.c error messages

On Mon, Feb 20, 2023 at 11:02:10AM -0800, Nathan Bossart wrote: > On Mon, Feb 20, 2023 at 08:54:48AM +0100, Peter Eisentraut wrote: >> I'm concerned about the loose use of "privilege" here. A privilege is >> something I can grant. So if someone doesn't have the "REPLICATION >> privilege", as in t

Re: [PATCH] Add pretty-printed XML output option

On 18.02.23 19:09, Peter Eisentraut wrote: On 17.02.23 23:24, Nikolay Samokhvalov wrote: My idea was to follow the SQL standard (part 14, SQL/XML); unfortunately, there is no free version, but there are drafts at http://www.wiscorp.com/SQLStandards.html

Re: Weird failure with latches in curculio on v15[

Greetings, * Michael Paquier (mich...@paquier.xyz) wrote: > On Sun, Feb 19, 2023 at 08:06:24PM +0530, Robert Haas wrote: > > I mean, my idea was to basically just have one big callback: > > ArchiverModuleMainLoopCB(). Which wouldn't return, or perhaps, would > > only return when archiving was tota

Re: ICU locale validation / canonicalization

New patch attached. The new patch also includes a GUC that (when enabled) validates that the collator is actually found. On Mon, 2023-02-20 at 15:46 +0100, Peter Eisentraut wrote: > a) BCP47 tags are preferred, and Agreed. > b) They don't work with ICU versions before 54. I tried in versions 5

Disable rdns for Kerberos tests

Greetings, The name canonicalization support for Kerberos is doing us more harm than good in the regression tests, so I propose we disable it. Patch attached. Thoughts? Thanks, Stephen From 992d946d17c79d240ac6587998e2f94b12a726de Mon Sep 17 00:00:00 2001 From: Stephen Frost Date: Mon, 20 Feb

Re: Support logical replication of DDLs

On 2/19/23 11:14 PM, Amit Kapila wrote: On Sun, Feb 19, 2023 at 7:50 AM Jonathan S. Katz wrote: On 2/17/23 4:15 AM, Amit Kapila wrote: This happens because of the function used in the index expression. Now, this is not the only thing, the replication can even fail during DDL replication when

meson and sslfiles.mk in src/test/ssl/

Hi all, While playing with the SSL tests last week, I have noticed that we don't have a way to regenerate the SSL files with meson for its TAP suite. It seems to me that we had better transfer the rules currently stored in sslfiles.mk into something that meson can use? Another approach may be to

Re: Weird failure with latches in curculio on v15

On Tue, Feb 14, 2023 at 09:47:55AM -0800, Nathan Bossart wrote: > Here is a new version of the stopgap/back-branch fix for restore_command. > This is more or less a rebased version of v4 with an added stderr message > as Andres suggested upthread. So, this thread has moved around many subjects, st

Re: Weird failure with latches in curculio on v15

On Tue, Feb 21, 2023 at 1:03 PM Michael Paquier wrote: > Perhaps beginning a new thread with a patch and a summary would be > better at this stage? Another thing I am wondering is if it could be > possible to test that rather reliably. I have been playing with a few > scenarios like holding the

Re: Rework of collation code, extensibility

On Mon, 2023-02-13 at 15:45 -0800, Jeff Davis wrote: > New version attached. Changes: These patches, especially 0001, have been around for a while, and they've received some review attention with no outstanding TODOs that I'm aware of. I plan to commit v10 (or something close to it) soon unless s

Re: DDL result is lost by CREATE DATABASE with WAL_LOG strategy

On Mon, Feb 20, 2023 at 05:02:01PM +0900, Michael Paquier wrote: > On Fri, Feb 17, 2023 at 02:35:30PM -0800, Nathan Bossart wrote: >> I'm happy to move this new test to wherever folks think it should go. I'll >> look around to see if I can find a better place, too. > > I think that src/test/recov

Re: Time delayed LR (WAS Re: logical replication restrictions)

Here are some review comments for the v3-0001 test code. == src/test/regress/sql/subscription.sql 1. +-- fail - utilizing streaming = parallel with time-delayed replication is not supported +CREATE SUBSCRIPTION regress_testsub CONNECTION 'dbname=regress_doesnotexist' PUBLICATION testpub WITH

meson logs environment

I've noticed that `meson test` logs the complete environment in meson_logs/testlog.txt. That seems unnecessary and probably undesirable for the buildfarm client. Is there any way to suppress that, or at least only print some relevant subset? (The buildfarm client itself only reports an approv

  1   2   >