On Sat, Jun 08, 2019 at 08:40:43PM -0400, Dave Cramer wrote:
On Sat, 8 Jun 2019 at 20:09, Andres Freund wrote:
Hi,
On 2019-06-08 19:41:34 -0400, Dave Cramer wrote:
> So the reason we are discussing using pgoutput plugin is because it is
part
> of core and guaranteed to be in cloud providers s
On Thu, Jun 06, 2019 at 06:01:21PM +0900, Michael Paquier wrote:
On Wed, Jun 05, 2019 at 10:31:54PM +0200, Peter Eisentraut wrote:
I think -r/--relfilenode was actually a good suggestion. Because it
doesn't actually check a *file* but potentially several files (forks,
segments). The -f naming
Dent John writes:
> I guess my question is, and I correctly understanding that reloptions are
> basically off-limits to EXTENSIONS?
IIRC that's basically true. There's a lot of dissatisfaction with the
current implementation of reloptions, although I think that it's been
mainly focused on the f
"Goel, Dhruv" writes:
> Yes, you are correct. The test case here was that if a tuple is inserted
> after the reference snapshot is taken in Phase 2 and before the index is
> marked ready. If this tuple is deleted before the reference snapshot of Phase
> 3, it will never make it to the index. I
Hi,
On June 9, 2019 8:36:37 AM PDT, Tom Lane wrote:
>"Goel, Dhruv" writes:
>I think you are mistaken that doing transactional updates in pg_index
>is OK. If memory serves, we rely on xmin of the pg_index row for
>purposes
>such as detecting whether a concurrently-created index is safe to use
>y
Thanks Fabien,
But the 2 direct questions i have are :
1. What is the structure of the Bloom Index ? Can you please let me know
what are the fields of a Bloom Index ? Is it just the Item Pointer
and BloomSignatureWord ?
When i describe my bloom index it looks like following.
postgres=# \d+ foo.i
I've been using cube extension recompiled with #define MAX_DIM 256.
But with a version 11.3 I'm getting the following error:failed to add item to
index page in
There's a regression unit test in contrib/cube/expected/cube.out:
CREATE TABLE test_cube (c cube);
\copy test_cube from 'data/test_cube.d
Hi,
(on postgres lists, please do not top-quote).
On 2019-06-08 04:06:39 -0500, Thierry Husson wrote:
> In fact, I suppose all temporary tables and their content could be
> completly ignored by MVCC principles as they are not subject to
> concurrency being unmodifiable/unreadable by other connect
On Sat, 8 Jun 2019 at 04:51, Andres Freund wrote:
>
> On 2019-06-07 09:48:29 -0400, Robert Haas wrote:
> > However, it looks to me as though copy.c can create a bunch of
> > BulkInsertStates but only call finish_bulk_insert() once, so unless
> > that's a bug in need of fixing I don't quite see how
Andres Freund writes:
> On June 9, 2019 8:36:37 AM PDT, Tom Lane wrote:
>> I think you are mistaken that doing transactional updates in pg_index
>> is OK. If memory serves, we rely on xmin of the pg_index row for
>> purposes such as detecting whether a concurrently-created index is safe
>> to us
Hi all
Memory leaks occur when the ecpg_update_declare_statement() is called the
second time.
FILE:postgresql\src\interfaces\ecpg\ecpglib\prepare.c
void
ecpg_update_declare_statement(const char *declared_name, const char
*cursor_name, const int lineno)
{
struct declared_statement *p = N
Hi all
In src\backend\utils\misc\guc.c, I found a potential memory leak.
make_absolute_path() return a malloc'd copy, we should free memory before the
function return false.
SelectConfigFiles(const char *userDoption, co
Hi
Currently the documentation for the default role "pg_signal_backend" states,
somewhat ambiguously, "Send signals to other backends (eg: cancel query,
terminate)",
giving the impression other signals (e.g. SIGHUP) can be sent too, which is
currently not the case.
Attached patch clarifies this
"Zhang, Jie" writes:
> In src\backend\utils\misc\guc.c, I found a potential memory leak.
> make_absolute_path() return a malloc'd copy, we should free memory before the
> function return false.
If SelectConfigFiles were executed more than once per postmaster
launch, this might be worth adding co
Hello.
In pg_upgrade, prep statuts is shown in English even if LANG is
set to other languages.
$ LANG=ja_JP.UTF8 pg_upgrade ...
<"Performing Consistency Checks on Old Live Server" in Japanese>
--
Checking cluster versions
Several TAP test suites have a need to translate from an msys path to a
Windows path. They currently use two ways to do that:
1. TestLib::real_dir, new in v11, is sound but works for directories only.
2. The $vfs_path approach is semi-private to PostgresNode.pm and 017_shm.pl,
and it does not
On Tue, 4 Jun 2019 at 21:28, Andres Freund wrote:
>
> Hi,
>
> On 2019-06-04 15:51:01 +0530, Amit Khandekar wrote:
> > After giving more thought on this, I think it might make sense to
> > arrange for the xl_running_xact record to be sent from master to the
> > standby, when a logical slot is to be
Hi,
I am reading the code that generating plan for `rowmarks` of Postgres
9.4 (
https://github.com/postgres/postgres/blob/REL9_4_STABLE/src/backend/optimizer/plan/planner.c#L2070
)
After emitting the `LockRows` plannode, the results cannot be considered
in order, and there are comments ther
But the 2 direct questions i have are :
1. What is the structure of the Bloom Index ? Can you please let me know
what are the fields of a Bloom Index ? Is it just the Item Pointer
and BloomSignatureWord ?
I'm not sure of Postgres actual implementation, I have just looked at the
underlying h
HEAPTUPLE_RECENTLY_DEAD, /* tuple is dead, but not deletable yet */
It is a tuple which has been deleted AND committed but before the delete
there is a transaction started but not committed. Let call this transaction
as Transaction A.
if we create index on this time, Let's call this index as In
Hello,
On Mon, Jun 10, 2019 at 11:31 AM Zhenghua Lyu wrote:
>
> 1. why after emitting `lockrows` plannode, the result can no longer be
> assumed sorted?
>
The plan corresponding to your select query is as following:
QUERY PLAN
---
Limit
-> LockRows
21 matches
Mail list logo
