Re: [HACKERS] Accessing non catalog table in backend

2016-01-04 Thread Atri Sharma
Thanks! I was wary to use SPI inside the executor for node evaluation functions. Does it seem safe? Regards, Atri On 5 Jan 2016 12:20 am, "Jim Nasby" wrote: > On 1/4/16 12:07 PM, Atri Sharma wrote: > >> Hi All, >> >> I wanted to check if it is possible to query a non

Re: [HACKERS] Additional role attributes && superuser review

2016-01-04 Thread Stephen Frost
* Michael Paquier (michael.paqu...@gmail.com) wrote: > On Thu, Dec 31, 2015 at 4:26 PM, Noah Misch wrote: > > The proposed pg_replication role introduces abstraction that could, as you > > hope, spare a DBA from studying sets of functions to grant together. The > >

Re: [HACKERS] 9.5 BLOCKER: regrole and regnamespace and quotes

2016-01-04 Thread Jim Nasby
On 1/3/16 10:37 PM, Tom Lane wrote: It's not a release stopper, but I plan to fix it in HEAD whenever I have an idle hour. Since I'm sure there's much better things you can be working on I was going to just do this myself. Then it occurred to me that this should be a great item for a new

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Alvaro Herrera
Robert Haas wrote: > But you could also write SELECT relname FROM pg_class WHERE > pg_relation_size(oid) > 100 * 1024^3, which is actually fewer > characters. Maybe pg_size_bytes('100 GB') is easier for some people > to remember than 100 * 1024^3, but I'm probably not one of those > people.

Re: [HACKERS] Accessing non catalog table in backend

2016-01-04 Thread Jim Nasby
On 1/4/16 12:07 PM, Atri Sharma wrote: Hi All, I wanted to check if it is possible to query a non catalog table in backend. I understand that cql interface is only for catalog querying hence it is not usable for this purpose per se. AFAIK it's possible to do it with low level routines, but I

Re: [HACKERS] remove wal_level archive

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 11:04 AM, Alvaro Herrera wrote: > Peter Eisentraut wrote: >> So we've had several rounds of discussions about simplifying replication >> configuration in general and the wal_level setting in particular. [0][1] >> Let's get something going. > > I

Re: [HACKERS] Additional role attributes && superuser review

2016-01-04 Thread Stephen Frost
* Robert Haas (robertmh...@gmail.com) wrote: > On Tue, Dec 29, 2015 at 5:35 AM, Stephen Frost wrote: > > * Noah Misch (n...@leadboat.com) wrote: > >> > Updated patch attached. I'll give it another good look and then commit > >> > it, barring objections. > >> > >> This thread

Re: [HACKERS] Broken lock management in policy.c.

2016-01-04 Thread Stephen Frost
* Peter Geoghegan (p...@heroku.com) wrote: > On Mon, Jan 4, 2016 at 12:13 PM, Tom Lane wrote: > > I've pushed an example based on your original test case. Feel free > > to suggest improvements, although at this point they'll probably > > land in 9.5.1. > > I think that

Re: [HACKERS] Broken lock management in policy.c.

2016-01-04 Thread Tom Lane
[ getting back to this now that there's a little time ] Peter Geoghegan writes: > On Sun, Jan 3, 2016 at 7:01 PM, Peter Geoghegan wrote: >> I would also advise only referencing a single relation within the >> SELECT FOR UPDATE. > To state what may be obvious:

Re: [HACKERS] pgsql: Further tweaking of print_aligned_vertical().

2016-01-04 Thread Greg Stark
On Mon, Jan 4, 2016 at 6:18 PM, Tom Lane wrote: >> VAX (simh): >> https://docs.google.com/presentation/d/1PG-bMU4WiS1pjtBwRvH4cE-nN9y5gj8ZLCO7KMrlvYg/view > >> Fuzzer: >> https://docs.google.com/presentation/d/12Dd9Bhcugkdi2w0ye4T1fy9ccjconJEz9cNthdeyH7k/view > > Very cool,

Re: [HACKERS] remove wal_level archive

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 3:05 PM, Alvaro Herrera wrote: > Robert Haas wrote: >> On Mon, Jan 4, 2016 at 11:04 AM, Alvaro Herrera >> wrote: >> > Peter Eisentraut wrote: >> >> So we've had several rounds of discussions about simplifying replication

Re: [HACKERS] ALTER TABLE behind-the-scenes effects' CONTEXT

2016-01-04 Thread Pavel Stehule
2016-01-04 21:44 GMT+01:00 Alvaro Herrera : > Pavel Stehule wrote: > > 2015-10-05 0:08 GMT+02:00 Marko Tiikkaja : > > > > > In the past I've found the error message in cases such as this somewhat > > > less helpful than it could be: > > > > > > =# CREATE

Re: [HACKERS] remove wal_level archive

2016-01-04 Thread Alvaro Herrera
Robert Haas wrote: > On Mon, Jan 4, 2016 at 11:04 AM, Alvaro Herrera > wrote: > > Peter Eisentraut wrote: > >> So we've had several rounds of discussions about simplifying replication > >> configuration in general and the wal_level setting in particular. [0][1] > >>

Re: [HACKERS] Accessing non catalog table in backend

2016-01-04 Thread Jim Nasby
On 1/4/16 12:53 PM, Atri Sharma wrote: Please don't top-post. On 5 Jan 2016 12:20 am, "Jim Nasby" > wrote: On 1/4/16 12:07 PM, Atri Sharma wrote: Hi All, I wanted to check if it is possible to query a non catalog

Re: [HACKERS] Broken lock management in policy.c.

2016-01-04 Thread Peter Geoghegan
On Mon, Jan 4, 2016 at 12:13 PM, Tom Lane wrote: > I've pushed an example based on your original test case. Feel free > to suggest improvements, although at this point they'll probably > land in 9.5.1. I think that that's a vast improvement. I probably should have pushed for

Re: [HACKERS] Additional role attributes && superuser review

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 12:55 PM, Stephen Frost wrote: > I'd like to be able to include, in both of those, a simple set of > instructions for granting the necessary rights to the user who is > running those processes. A set of rights which an administrator can go > look up and

Re: [HACKERS] [PATCH] we have added support for box type in SP-GiST index

2016-01-04 Thread Alvaro Herrera
CC'ing Teodor because he's author of one of the patches. Alexander Lebedev wrote: > Hello, Hacker. So, can we have a more thorough explanation of what is the point of this patch? If it is supposed to improve the performance of searching for boxes, can we see measurements from some benchmark?

Re: [HACKERS] bootstrap pg_shseclabel in relcache initialization

2016-01-04 Thread Alvaro Herrera
Adam Brightwell wrote: > While working on an auth hook, I found that I was unable to access the > pg_shseclabel system table while processing the hook. [ ... ] > Given that the shared relations currently exposed can also have > security labels that can be used for auth purposes, I believe it

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Pavel Stehule
2016-01-04 21:29 GMT+01:00 Robert Haas : > On Mon, Jan 4, 2016 at 12:17 PM, Shulgin, Oleksandr > wrote: > >> I'm also kind of wondering what the intended use case for this > >> function is. Why do we want it? Do we want it? > > > > As

[HACKERS] Beginner hacker item: Fix to_reg*() input type

2016-01-04 Thread Jim Nasby
All the to_reg* functions in src/backend/utils/adt/regproc.c currently accept a cstring. Per [1], they should accept text. This should be a fairly simple change to make: - Modify the functions in regproc.c. Take a look at how other text input functions work to see what needs to happen here

Re: [HACKERS] parallel joins, and better parallel explain

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 4:50 AM, Dilip Kumar wrote: > I tried to create a inner table such that, inner table data don't fit in RAM > (I created VM with 1GB Ram). > Purpose of this is to make Disk scan dominant, > and since parallel join is repeating the Disk Scan and hash

Re: [HACKERS] ALTER TABLE behind-the-scenes effects' CONTEXT

2016-01-04 Thread Alvaro Herrera
Pavel Stehule wrote: > 2015-10-05 0:08 GMT+02:00 Marko Tiikkaja : > > > In the past I've found the error message in cases such as this somewhat > > less helpful than it could be: > > > > =# CREATE TABLE qqq (a int); > > =# CREATE UNIQUE INDEX IF NOT EXISTS qqq_a_idx ON qqq(a); > >

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 3:48 PM, Alvaro Herrera wrote: > Robert Haas wrote: > >> But you could also write SELECT relname FROM pg_class WHERE >> pg_relation_size(oid) > 100 * 1024^3, which is actually fewer >> characters. Maybe pg_size_bytes('100 GB') is easier for some

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 12:17 PM, Shulgin, Oleksandr wrote: >> I'm also kind of wondering what the intended use case for this >> function is. Why do we want it? Do we want it? > > As suggested above a usecase could be like the following: > > SELECT relname FROM

Re: [HACKERS] Extracting fields from 'infinity'::TIMESTAMP[TZ]

2016-01-04 Thread Alvaro Herrera
Vitaly Burovoy wrote: > Majority of the votes for NULL for "other things" except epoch. > Nobody answers about differences between monotonic and oscillating values. > > I suppose behavior of monotonic values (julian, century, decade, > isoyear, millennium and year) should be the same as for

Re: [HACKERS] Broken lock management in policy.c.

2016-01-04 Thread Tom Lane
I wrote: > I'll go draft something up ... I've pushed an example based on your original test case. Feel free to suggest improvements, although at this point they'll probably land in 9.5.1. regards, tom lane -- Sent via pgsql-hackers mailing list

Re: [HACKERS] count_nulls(VARIADIC "any")

2016-01-04 Thread Pavel Stehule
Hi 2016-01-04 5:49 GMT+01:00 Jim Nasby : > On 1/3/16 10:23 PM, Pavel Stehule wrote: > >> Hi >> >> 2016-01-03 22:49 GMT+01:00 Jim Nasby > >: >> >> On 1/3/16 2:37 PM, Pavel Stehule wrote: >> >> +

Re: [HACKERS] Some 9.5beta2 backend processes not terminating properly?

2016-01-04 Thread Andres Freund
On 2016-01-03 13:57:18 -0500, Tom Lane wrote: > Done, we'll soon see what the buildfarm thinks. Thanks. I wonder if we ought to backport this further: e.g. walsender continously uses nonblocking sockets via pq_getbyte_if_available(). On the other hand I can't immediately see a problem with that,

Re: [HACKERS] parallel joins, and better parallel explain

2016-01-04 Thread Dilip Kumar
On Thu, Dec 24, 2015 at 4:45 AM, Robert Haas wrote: > On Wed, Dec 23, 2015 at 2:34 AM, Dilip Kumar > wrote: > > Yeah right, After applying all three patches this problem is fixed, now > > parallel hash join is faster than normal hash join. > > > > I

Re: [HACKERS] Multi-tenancy with RLS

2016-01-04 Thread Haribabu Kommi
On Mon, Jan 4, 2016 at 8:34 PM, Amit Langote wrote: > On 2016/01/04 14:43, Haribabu Kommi wrote: >>> >>> Here I attached new series of patches with a slightly different approach. >>> Instead of creating the policies on the system catalog tables whenever >>> the

Re: [HACKERS] \x auto and EXPLAIN

2016-01-04 Thread Shulgin, Oleksandr
On Sun, Jan 3, 2016 at 6:43 PM, Tom Lane wrote: > Andreas Karlsson writes: > > psql's "\x auto" is a nice feature, but it is made much less useful in > > my opinion due to the expanded output format making query plans > > unreadable (and query plans often

Re: [HACKERS] Proposal: SET ROLE hook

2016-01-04 Thread Pavel Stehule
Hi 2015-10-16 18:20 GMT+02:00 Joe Conway : > In many environments there is a policy requiring users to login using > unprivileged accounts, and then escalate their privileges if and when > they require it. In PostgreSQL this could be done by granting the > superuser role to

[HACKERS] Bug in MergeAttributesIntoExisting() function.

2016-01-04 Thread amul sul
Hi, In inheritance, child column's pg_attribute.attislocal flag not getting updated, if it is inherited using ALTER TABLE INHERIT . Due to this, if we try to drop column(s) from parent table, which are not getting drop from child. Attached herewith is quick patch fixing this issue.

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Shulgin, Oleksandr
On Wed, Dec 30, 2015 at 8:28 PM, Pavel Stehule wrote: > > > 2015-12-30 17:33 GMT+01:00 Robert Haas : > >> On Mon, Dec 28, 2015 at 8:45 AM, Shulgin, Oleksandr >> wrote: >> > I didn't check out earlier versions of this

Re: [HACKERS] Remove Windows crash dump support?

2016-01-04 Thread Andres Freund
On 2016-01-04 16:39:22 +0900, Michael Paquier wrote: > So I think that it is still useful for debugging code > paths running custom code, even if we consider Postgres as rock-solid > on Windows. Given the state of e.g. the socket using code for windows I personally certainly don't consider it

Re: [HACKERS] Avoiding pin scan during btree vacuum

2016-01-04 Thread Andres Freund
On 2016-01-03 15:40:01 +, Simon Riggs wrote: > I'm happy with this being a simple patch now, not least because I would > like to backpatch this to 9.4 where catalog scans became MVCC. > > A backpatch is warranted because it is a severe performance issue with > replication and we can fix that

Re: [HACKERS] Keyword classifications

2016-01-04 Thread Simon Riggs
On 1 January 2016 at 19:06, Tom Lane wrote: > I wrote: > > Now, one line of thought here is that flatten_reloptions() is out of its > > mind to not be worrying about quoting the reloption values. And perhaps > > it is, but I think if we go that direction, we may be fighting

Re: [HACKERS] Multi-tenancy with RLS

2016-01-04 Thread Amit Langote
On 2016/01/04 14:43, Haribabu Kommi wrote: >> >> Here I attached new series of patches with a slightly different approach. >> Instead of creating the policies on the system catalog tables whenever >> the catalog security command is executed, just enable row level security >> on the system catalog

Re: [HACKERS]WIP: Covering + unique indexes.

2016-01-04 Thread David Rowley
On 2 December 2015 at 01:53, Anastasia Lubennikova < a.lubennik...@postgrespro.ru> wrote: > Finally, completed patch "covering_unique_3.0.patch" is here. > It includes the functionality discussed above in the thread, regression > tests and docs update. > I think it's quite ready for review. > Hi

Re: [HACKERS] POC: Cache data in GetSnapshotData()

2016-01-04 Thread Andres Freund
On 2015-12-19 22:47:30 -0800, Mithun Cy wrote: > After some analysis I saw writing to shared memory to store shared snapshot > is not protected under exclusive write lock, this leads to memory > corruptions. > I think until this is fixed measuring the performance will not be much > useful. I

Re: [HACKERS] Additional role attributes && superuser review

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 4:56 PM, Stephen Frost wrote: >> First, it's not really going to matter to users very much whether the >> command to enable one of these features is a single GRANT command or a >> short sequence of GRANT commands executed one after another. So even >>

Re: [HACKERS] Additional role attributes && superuser review

2016-01-04 Thread Stephen Frost
* Robert Haas (robertmh...@gmail.com) wrote: > On Mon, Jan 4, 2016 at 12:55 PM, Stephen Frost wrote: > > I'd like to be able to include, in both of those, a simple set of > > instructions for granting the necessary rights to the user who is > > running those processes. A set

Re: [HACKERS] Additional role attributes && superuser review

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 5:22 PM, Stephen Frost wrote: >> So, is this another case where the support is all in off-list fora and >> thus invisible, or can you point to specific on-list discussions where >> it was supported, and to the opinions offered in support? I don't >>

Re: [HACKERS] Beginner hacker item: Fix to_reg*() input type

2016-01-04 Thread Petr Korobeinikov
> - Modify the functions in regproc.c. Take a look at how other text input > functions work to see what needs to happen here (you'll want to use > text_to_cstring() as part of that.) > > - Modify the appropriate entries in src/include/catalog/pg_proc.h Let me try. `make check` says "All 160 tests

Re: [HACKERS] parallel joins, and better parallel explain

2016-01-04 Thread Dilip Kumar
On Tue, Jan 5, 2016 at 1:52 AM, Robert Haas wrote: > On Mon, Jan 4, 2016 at 4:50 AM, Dilip Kumar wrote: > > I tried to create a inner table such that, inner table data don't fit in > RAM > > (I created VM with 1GB Ram). > > Purpose of this is to

Re: [HACKERS] Accessing non catalog table in backend

2016-01-04 Thread Amit Langote
On 2016/01/05 3:53, Atri Sharma wrote: > I was wary to use SPI inside the executor for node evaluation functions. > Does it seem safe? What is "node evaluation functions"? Is it "Plan" nodes or "Expr" nodes that you are talking about? I guess you'd know to use ExecProcNode() or ExecEvalExpr() for

Re: [HACKERS] Making tab-complete.c easier to maintain

2016-01-04 Thread Tom Lane
Michael Paquier writes: > So, here are the commands that still remain with TailMatches to cover > this case, per gram.y: > - CREATE TABLE > - CREATE INDEX > - CREATE VIEW > - GRANT > - CREATE TRIGGER > - CREATE SEQUENCE > New patches are attached. I've reviewed and

Re: [HACKERS] remove wal_level archive

2016-01-04 Thread Michael Paquier
On Tue, Jan 5, 2016 at 1:04 AM, Alvaro Herrera wrote: > Peter Eisentraut wrote: >> So we've had several rounds of discussions about simplifying replication >> configuration in general and the wal_level setting in particular. [0][1] >> Let's get something going. > > I

[HACKERS] comment typo in RewindTest.pm

2016-01-04 Thread Masahiko Sawada
Hi all, Attached patch fixes the comment typo in RewindTest.pm Please find it. Regards, -- Masahiko Sawada diff --git a/src/bin/pg_rewind/RewindTest.pm b/src/bin/pg_rewind/RewindTest.pm index c1c7d1f..41d294e 100644 --- a/src/bin/pg_rewind/RewindTest.pm +++ b/src/bin/pg_rewind/RewindTest.pm @@

Re: [HACKERS] Keyword classifications

2016-01-04 Thread Alvaro Herrera
Tom Lane wrote: > But some experimentation suggests that we could > fix that by subdividing col_name_keyword into two categories, one being > the keywords that can also be type names (BIGINT, BIT, etc) and one > being those that can't (BETWEEN, COALESCE, etc). Everywhere > col_name_keyword is

Re: [HACKERS] \x auto and EXPLAIN

2016-01-04 Thread Andreas Karlsson
On 01/03/2016 06:43 PM, Tom Lane wrote: I see two ways to fix this. 1) Never use expanded display for the case where there is only one column. There seems to me like there is little value in using expanded display for when you only have one column, but I may be missing some use case here.

Re: [HACKERS] Bug in MergeAttributesIntoExisting() function.

2016-01-04 Thread Rushabh Lathia
On Mon, Jan 4, 2016 at 4:41 PM, amul sul wrote: > Hi, > > In inheritance, child column's pg_attribute.attislocal flag not getting > updated, if it is inherited using ALTER TABLE INHERIT . > > Due to this, if we try to drop column(s) from parent table, which are not >

Re: [HACKERS] dynloader.h missing in prebuilt package for Windows?

2016-01-04 Thread Tom Lane
Michael Paquier writes: > The patch would put the buildfarm in red as it is incomplete anyway, > with MSVC what is used instead of dynloader.h is > port/dynloader/win32.h. Instead of this patch I would be incline to > remove the #define stuff with dynloader.h that use

Re: [HACKERS] dynloader.h missing in prebuilt package for Windows?

2016-01-04 Thread Michael Paquier
On Tue, Jan 5, 2016 at 2:27 PM, Tom Lane wrote: > Michael Paquier writes: >> The patch would put the buildfarm in red as it is incomplete anyway, >> with MSVC what is used instead of dynloader.h is >> port/dynloader/win32.h. Instead of this patch I

Re: [HACKERS] Accessing non catalog table in backend

2016-01-04 Thread Amit Langote
On 2016/01/05 14:30, Atri Sharma wrote: > On Tue, Jan 5, 2016 at 9:54 AM, Amit Langote >> On 2016/01/05 3:53, Atri Sharma wrote: >>> I was wary to use SPI inside the executor for node evaluation functions. >>> Does it seem safe? >> >> What is "node evaluation

Re: [HACKERS] Making tab-complete.c easier to maintain

2016-01-04 Thread Michael Paquier
On Tue, Jan 5, 2016 at 10:13 AM, Tom Lane wrote: > Michael Paquier writes: >> So, here are the commands that still remain with TailMatches to cover >> this case, per gram.y: >> - CREATE TABLE >> - CREATE INDEX >> - CREATE VIEW >> - GRANT >> - CREATE

Re: [HACKERS] dynloader.h missing in prebuilt package for Windows?

2016-01-04 Thread Chapman Flack
On 01/05/16 00:18, Michael Paquier wrote: > with MSVC what is used instead of dynloader.h is > port/dynloader/win32.h. Seems like a good catch - AFAICS, what happens with port/dynloader is that for 12 different OSes, there's an .h file there to be copied _renamed to dynloader.h_ into the build

Re: [HACKERS] dynloader.h missing in prebuilt package for Windows?

2016-01-04 Thread Michael Paquier
On Mon, Jan 4, 2016 at 9:37 PM, Chapman Flack wrote: > On 01/05/16 00:18, Michael Paquier wrote: > >> with MSVC what is used instead of dynloader.h is >> port/dynloader/win32.h. > > Seems like a good catch - AFAICS, what happens with port/dynloader > is that for 12

Re: [HACKERS] Keyword classifications

2016-01-04 Thread Tom Lane
Michael Paquier writes: > On Sat, Jan 2, 2016 at 4:06 AM, Tom Lane wrote: >> The grammar fixes seem like a good thing to do in the long run, too, >> but there's little need to risk back-patching it since accepting >> col_name_keywords without

Re: [HACKERS] ALTER TABLE behind-the-scenes effects' CONTEXT

2016-01-04 Thread Simon Riggs
On 4 January 2016 at 20:44, Alvaro Herrera wrote: > Maybe > there are more ALTER TABLE subcommands that should be setting something > up? In cases where multiple subcommands are being run, it might be > useful to see which one caused a certain error message. > I like

Re: [HACKERS] dynloader.h missing in prebuilt package for Windows?

2016-01-04 Thread Michael Paquier
On Mon, Jan 4, 2016 at 10:06 AM, Bruce Momjian wrote: > On Mon, Jan 4, 2016 at 12:59:26PM -0500, Tom Lane wrote: >> Bruce Momjian writes: >> > On Sun, Jan 3, 2016 at 12:35:02PM -0500, Tom Lane wrote: >> >> If we're willing to allow 9.4.6 to install different

Re: [HACKERS] Accessing non catalog table in backend

2016-01-04 Thread Tom Lane
Atri Sharma writes: > I fail to see the relevance of which node is getting evaluated (its a Plan > node BTW) for this question. The concern I had was around using SPI inside > executor and its fail safety. The code path executor -> PL function -> SPI certainly works, so

Re: [HACKERS] Accessing non catalog table in backend

2016-01-04 Thread Atri Sharma
On Tue, Jan 5, 2016 at 9:54 AM, Amit Langote wrote: > On 2016/01/05 3:53, Atri Sharma wrote: > > I was wary to use SPI inside the executor for node evaluation functions. > > Does it seem safe? > > What is "node evaluation functions"? Is it "Plan" nodes or "Expr"

Re: [HACKERS] Some 9.5beta2 backend processes not terminating properly?

2016-01-04 Thread Andres Freund
On 2016-01-04 10:20:43 -0500, Tom Lane wrote: > I'm slightly worried about breaking 3rd-party code that might be using > recv() and somehow expecting the current behavior. However, it's equally > arguable that such code would have Windows-specific problems that would be > fixed by the patch. I

Re: [HACKERS] Some 9.5beta2 backend processes not terminating properly?

2016-01-04 Thread Magnus Hagander
On Mon, Jan 4, 2016 at 4:20 PM, Tom Lane wrote: > Andres Freund writes: > > I wonder if we ought to backport this further: e.g. walsender > > continously uses nonblocking sockets via pq_getbyte_if_available(). On > > the other hand I can't immediately see

Re: [HACKERS] Avoiding pin scan during btree vacuum

2016-01-04 Thread Tom Lane
Andres Freund writes: > On 2016-01-03 15:40:01 +, Simon Riggs wrote: >> I'm happy with this being a simple patch now, not least because I would >> like to backpatch this to 9.4 where catalog scans became MVCC. >> >> A backpatch is warranted because it is a severe

Re: [HACKERS] Some 9.5beta2 backend processes not terminating properly?

2016-01-04 Thread Tom Lane
Magnus Hagander writes: > On Mon, Jan 4, 2016 at 4:20 PM, Tom Lane wrote: >> I'm slightly worried about breaking 3rd-party code that might be using >> recv() and somehow expecting the current behavior. However, it's equally >> arguable that such code

Re: [HACKERS] comment typo in RewindTest.pm

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 8:08 AM, Masahiko Sawada wrote: > Attached patch fixes the comment typo in RewindTest.pm > Please find it. Committed and back-patched to 9.5. -- Robert Haas EnterpriseDB: http://www.enterprisedb.com The Enterprise PostgreSQL Company -- Sent via

Re: [HACKERS] Some 9.5beta2 backend processes not terminating properly?

2016-01-04 Thread Tom Lane
Robert Haas writes: > OK, well, if the consensus is in favor of a back-patch, so be it. It > seems a little strange to me to back-patch a commit that doesn't fix > anything, but I just work here. Well, it's true that we can't point to specific field reports and say that

Re: [HACKERS] Bug in MergeAttributesIntoExisting() function.

2016-01-04 Thread Amit Langote
Hi, On Mon, Jan 4, 2016 at 8:11 PM, amul sul wrote: > Hi, > > In inheritance, child column's pg_attribute.attislocal flag not getting > updated, if it is inherited using ALTER TABLE INHERIT . > > Due to this, if we try to drop column(s) from parent table, which are not >

Re: [HACKERS] Bug in MergeAttributesIntoExisting() function.

2016-01-04 Thread Tom Lane
amul sul writes: > In inheritance, child column's pg_attribute.attislocal flag not getting > updated, if it is inherited using ALTER TABLE INHERIT . I think this patch is wrong and you have broken the intended behavior. It's a bit hard to tell though because your example

Re: [HACKERS] Avoiding pin scan during btree vacuum

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 10:30 AM, Tom Lane wrote: >> This seems like a might subtle thing to backpatch. If we really want to >> go there, ISTM that the relevant code should stew in an unreleased >> branch for a while, before being backpatched. > > I'm definitely -1 on

Re: [HACKERS] Some 9.5beta2 backend processes not terminating properly?

2016-01-04 Thread Tom Lane
Andres Freund writes: > On 2016-01-04 10:35:12 -0500, Robert Haas wrote: >> If we don't know of a specific problem that would be fixed by >> back-patching this commit to pre-9.5 branches, and it seems like we >> don't, then I don't really see much upside to back-patching it.

Re: [HACKERS] pgsql: Further tweaking of print_aligned_vertical().

2016-01-04 Thread Greg Stark
On Mon, Jan 4, 2016 at 6:24 AM, Michael Paquier wrote: > Arg, actually it is not: > https://wiki.postgresql.org/wiki/PostgreSQL_Conference_Europe_Talks_2015 > Greg could you add it there? No, apparently that page is restricted to only some users. I used plenty of

Re: [HACKERS] Some 9.5beta2 backend processes not terminating properly?

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 10:22 AM, Magnus Hagander wrote: > On Mon, Jan 4, 2016 at 4:20 PM, Tom Lane wrote: >> Andres Freund writes: >> > I wonder if we ought to backport this further: e.g. walsender >> > continously uses nonblocking

Re: [HACKERS] Some 9.5beta2 backend processes not terminating properly?

2016-01-04 Thread Andres Freund
On 2016-01-04 10:35:12 -0500, Robert Haas wrote: > If we don't know of a specific problem that would be fixed by > back-patching this commit to pre-9.5 branches, and it seems like we > don't, then I don't really see much upside to back-patching it. I > mean, yeah, we think that this is wrong

Re: [HACKERS] Some 9.5beta2 backend processes not terminating properly?

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 10:49 AM, Tom Lane wrote: > Andres Freund writes: >> On 2016-01-04 10:35:12 -0500, Robert Haas wrote: >>> If we don't know of a specific problem that would be fixed by >>> back-patching this commit to pre-9.5 branches, and it seems

Re: [HACKERS] Bug in MergeAttributesIntoExisting() function.

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 9:28 AM, Rushabh Lathia wrote: > On Mon, Jan 4, 2016 at 4:41 PM, amul sul wrote: >> Hi, >> In inheritance, child column's pg_attribute.attislocal flag not getting >> updated, if it is inherited using ALTER TABLE INHERIT . >>

Re: [HACKERS] [PATCH] Refactoring of LWLock tranches

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 1:20 AM, Amit Kapila wrote: > On Mon, Jan 4, 2016 at 4:49 AM, Robert Haas wrote: >> On Thu, Dec 31, 2015 at 3:36 AM, Amit Kapila >> wrote: >> > LWLock *LWLockAssignFromTranche(const char

Re: [HACKERS] Rationalizing Query.withCheckOptions

2016-01-04 Thread Stephen Frost
Tom, * Tom Lane (t...@sss.pgh.pa.us) wrote: > The RLS patch added this to struct Query: > > List *withCheckOptions;/* a list of WithCheckOption's, which > * are only added during rewrite and > *

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Pavel Stehule
Hi 2016-01-04 12:46 GMT+01:00 Shulgin, Oleksandr : > On Wed, Dec 30, 2015 at 8:28 PM, Pavel Stehule > wrote: > >> >> >> 2015-12-30 17:33 GMT+01:00 Robert Haas : >> >>> On Mon, Dec 28, 2015 at 8:45 AM, Shulgin,

Re: [HACKERS] Some 9.5beta2 backend processes not terminating properly?

2016-01-04 Thread Tom Lane
Andres Freund writes: > I wonder if we ought to backport this further: e.g. walsender > continously uses nonblocking sockets via pq_getbyte_if_available(). On > the other hand I can't immediately see a problem with that, besides > differing messages on windows/the rest of the

Re: [HACKERS] Building pg_xlogdump reproducibly

2016-01-04 Thread Andres Freund
Hi, On 2016-01-04 15:59:46 +0100, Christoph Berg wrote: > The list of objects used to link pg_xlogdump is coming from > $(wildcard *desc.c) which returns them in filesystem order. This makes > the build result depend on this ordering, yielding different > compilation results. > -RMGRDESCSOURCES

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 10:17 AM, Pavel Stehule wrote: > [ new patch ] + case '-': + ereport(ERROR, + (errcode(ERRCODE_INVALID_PARAMETER_VALUE), + errmsg("size cannot be negative"))); Why not? I bet if you

[HACKERS] Building pg_xlogdump reproducibly

2016-01-04 Thread Christoph Berg
The list of objects used to link pg_xlogdump is coming from $(wildcard *desc.c) which returns them in filesystem order. This makes the build result depend on this ordering, yielding different compilation results. This patch fixes the reproducibility issue: --- a/src/bin/pg_xlogdump/Makefile +++

Re: [HACKERS] Building pg_xlogdump reproducibly

2016-01-04 Thread David Fetter
On Mon, Jan 04, 2016 at 04:51:25PM +0100, Andres Freund wrote: > Hi, > > On 2016-01-04 15:59:46 +0100, Christoph Berg wrote: > > The list of objects used to link pg_xlogdump is coming from > > $(wildcard *desc.c) which returns them in filesystem order. This makes > > the build result depend on

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Pavel Stehule
2016-01-04 16:51 GMT+01:00 Robert Haas : > On Mon, Jan 4, 2016 at 10:17 AM, Pavel Stehule > wrote: > > [ new patch ] > > + case '-': > + ereport(ERROR, > + (errcode(ERRCODE_INVALID_PARAMETER_VALUE), > +

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Shulgin, Oleksandr
On Mon, Jan 4, 2016 at 6:03 PM, Pavel Stehule wrote: > > 2016-01-04 17:48 GMT+01:00 Shulgin, Oleksandr < oleksandr.shul...@zalando.de>: >> >> On Mon, Jan 4, 2016 at 4:51 PM, Robert Haas wrote: >>> >>> On Mon, Jan 4, 2016 at 10:17 AM, Pavel Stehule

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 11:48 AM, Shulgin, Oleksandr wrote: > On Mon, Jan 4, 2016 at 4:51 PM, Robert Haas wrote: >> >> On Mon, Jan 4, 2016 at 10:17 AM, Pavel Stehule >> wrote: >> > [ new patch ] >> >> + case

Re: [HACKERS] Some 9.5beta2 backend processes not terminating properly?

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 10:59 AM, Tom Lane wrote: > Robert Haas writes: >> OK, well, if the consensus is in favor of a back-patch, so be it. It >> seems a little strange to me to back-patch a commit that doesn't fix >> anything, but I just work here. >

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Shulgin, Oleksandr
On Mon, Jan 4, 2016 at 6:14 PM, Robert Haas wrote: > On Mon, Jan 4, 2016 at 11:48 AM, Shulgin, Oleksandr > > > > postgres=# select pg_size_bytes(''); > > ERROR: invalid input syntax for type numeric: "" > > I think that's a pretty bad error message. I mean, the user is >

Re: [HACKERS] 2016-01 Commitfest

2016-01-04 Thread Alvaro Herrera
Here are the current numbers, now that the commitfest has actually closed for business: Needs review: 79. Waiting on Author: 5. Ready for Committer: 6. Committed: 8. Total: 98. Of those RfC patches, one ("Default Roles") doesn't actually seem ready to commit, since there's been some

Re: [HACKERS] remove wal_level archive

2016-01-04 Thread Alvaro Herrera
Peter Eisentraut wrote: > So we've had several rounds of discussions about simplifying replication > configuration in general and the wal_level setting in particular. [0][1] > Let's get something going. I looked at this patch, which I think has got enough consensus that you should just push

Re: [HACKERS] Re: pglogical_output - a general purpose logical decoding output plugin

2016-01-04 Thread Shulgin, Oleksandr
On Sun, Jan 3, 2016 at 7:21 PM, Tomasz Rybak wrote: > The following review has been posted through the commitfest application: > make installcheck-world: tested, passed > Implements feature: not tested > Spec compliant: not tested > Documentation:

Re: [HACKERS] custom function for converting human readable sizes to bytes

2016-01-04 Thread Pavel Stehule
2016-01-04 18:14 GMT+01:00 Robert Haas : > On Mon, Jan 4, 2016 at 11:48 AM, Shulgin, Oleksandr > wrote: > > On Mon, Jan 4, 2016 at 4:51 PM, Robert Haas > wrote: > >> > >> On Mon, Jan 4, 2016 at 10:17 AM, Pavel Stehule

Re: [HACKERS] Additional role attributes && superuser review

2016-01-04 Thread Robert Haas
On Mon, Jan 4, 2016 at 3:07 PM, Stephen Frost wrote: > I'm not sure it's entirely relevant now- I've outlined the reasoning in > my email to Noah as a, hopefully, pretty comprehensive summary. If that > doesn't sway your minds then it seems unlikely that a reference to a >

Re: [HACKERS] Additional role attributes && superuser review

2016-01-04 Thread Stephen Frost
* Robert Haas (robertmh...@gmail.com) wrote: > On Mon, Jan 4, 2016 at 3:07 PM, Stephen Frost wrote: > > I'm not sure it's entirely relevant now- I've outlined the reasoning in > > my email to Noah as a, hopefully, pretty comprehensive summary. If that > > doesn't sway your

Re: [HACKERS] dynloader.h missing in prebuilt package for Windows?

2016-01-04 Thread Bruce Momjian
On Sun, Jan 3, 2016 at 12:35:02PM -0500, Tom Lane wrote: > Robert Haas writes: > > On Dec 31, 2015, at 1:04 PM, Bruce Momjian wrote: > >> Let's hold this for 9.5.1 and all minor releases will get it at the same > >> time. > > > I vote for going ahead

Re: [HACKERS] dynloader.h missing in prebuilt package for Windows?

2016-01-04 Thread Tom Lane
Bruce Momjian writes: > On Sun, Jan 3, 2016 at 12:35:02PM -0500, Tom Lane wrote: >> If we're willing to allow 9.4.6 to install different files than 9.4.5 >> does, I don't see why it's a problem for 9.5.1. But having said that, >> I agree that this seems pretty low-risk, and so

Re: [HACKERS] dynloader.h missing in prebuilt package for Windows?

2016-01-04 Thread Bruce Momjian
On Mon, Jan 4, 2016 at 12:59:26PM -0500, Tom Lane wrote: > Bruce Momjian writes: > > On Sun, Jan 3, 2016 at 12:35:02PM -0500, Tom Lane wrote: > >> If we're willing to allow 9.4.6 to install different files than 9.4.5 > >> does, I don't see why it's a problem for 9.5.1. But

  1   2   >