Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Wed, Oct 12, 2016 at 10:06 AM, Kevin Grittner wrote: > The test in ExecCheckHeapTupleVisible() seems wrong to me. It's > not immediately obvious what the proper fix is. To identify what cases ExecCheckHeapTupleVisible() was meant to cover I commented out the body of the

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Wed, Oct 12, 2016 at 3:07 PM, Peter Geoghegan wrote: > On Wed, Oct 12, 2016 at 1:05 PM, Thomas Munro > wrote: >> Here's a patch that shows one way to fix it. I think it does make >> sense to change this, because otherwise automatic >>

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Wed, Oct 12, 2016 at 1:41 PM, Kevin Grittner wrote: > Aren't these two completely separate and independent bugs? Technically they are, but they are both isolated to the same small function. Surely it's better to fix them both at once? -- Peter Geoghegan -- Sent via

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Wed, Oct 12, 2016 at 3:55 PM, Peter Geoghegan wrote: > On Wed, Oct 12, 2016 at 1:41 PM, Kevin Grittner wrote: >> Aren't these two completely separate and independent bugs? > > Technically they are, but they are both isolated to the same small > function.

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Wed, Oct 12, 2016 at 3:02 PM, Peter Geoghegan wrote: > I agree that the multi-value case is a bug. > I think that it should be pretty obvious to you why the check exists > at all, Kevin. It exists because it would be improper to decide to > take the DO NOTHING path on the basis

Re: [GENERAL] confusion about user paring with pg_hba and pg_ident

On Tue, Oct 11, 2016 at 4:20 PM Tom Lane wrote: > arnaud gaboury writes: > > > I am a little confused about some of my settings when it comes to map > > > linux/psql users. > > > > I think you're misunderstanding what the user-mapping stuff does. >

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Wed, Oct 12, 2016 at 8:50 PM, Albe Laurenz wrote: > Kevin Grittner wrote: >> On Tue, Oct 11, 2016 at 2:29 PM, Jason Dusek wrote: >>> I notice the following oddity: >> >>> =# CREATE TABLE with_pk (i integer PRIMARY KEY); >>> CREATE TABLE >> >>>

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

Kevin Grittner wrote: > Sent: Tuesday, October 11, 2016 10:00 PM > To: Jason Dusek > Cc: pgsql-general@postgresql.org > Subject: Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES > > On Tue, Oct 11, 2016 at 2:29 PM, Jason Dusek wrote: > >> I notice the following

Re: [GENERAL] confusion about user paring with pg_hba and pg_ident

On Tue, Oct 11, 2016 at 4:20 PM Tom Lane wrote: > arnaud gaboury writes: > > > I am a little confused about some of my settings when it comes to map > > > linux/psql users. > > > > I think you're misunderstanding what the user-mapping stuff does. >

Re: [GENERAL] ANN: Upscene releases Database Workbench 5.2.4

On Wed, Oct 12, 2016 at 2:51 AM, Martijn Tonies (Upscene Productions) wrote: Upscene Productions is proud to announce the availability of the next version of the popular multi-DBMS development tool: " Database Workbench 5.2.4 " The 5.2

Re: [GENERAL] confusion about user paring with pg_hba and pg_ident

On 10/12/2016 02:07 AM, arnaud gaboury wrote: On Tue, Oct 11, 2016 at 4:20 PM Tom Lane > wrote: arnaud gaboury > writes: > I am a little confused about some of my settings

Re: [GENERAL] psql 9.3 automatic recovery in progress

Just to inform u that my log Yesterday didn't show any 'recovery' which is the main goal of this post. Looks like the script was causing the issue, I test and change it to the one from Scott I'. On odoo I switch user sessions from default value 1 week(:-/) to 1 hour. I found that option in the

Re: [GENERAL] LOG: munmap(0x7fff80000000) failed: Invalid argument

I have another application that consumes all of the huge pages; they aren't for pgsql. :) I've modified the configuration file from "try" to "off" and munmap is no more. Mischief managed. Thanks for your help. On Tue, Oct 11, 2016 at 6:15 PM, Tom Lane wrote: > Chris

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Wed, Oct 12, 2016 at 2:50 AM, Albe Laurenz wrote: > Kevin Grittner wrote: >> I don't see that on development HEAD. What version are you >> running? What is your setting for default_transaction_isolation? > > The subject says SERIALIZABLE, and I can see it on my

[GENERAL] OmniDB 1.6

Guys, We are developing a web tool called OmniDB: http://www.omnidb.com.br/en_index.aspx OmniDB supports Firebird, MariaDB, MySQL, Oracle, PostgreSQL, SQLite, Microsoft SQL Server and Microsoft Access. OmniDB provides a unified workspace to manage databases from different DBMS in a comfortable

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

[adding Peter Geoghegan to the email addresses] On Wed, Oct 12, 2016 at 7:11 AM, Vitaly Burovoy wrote: > On 10/12/16, Thomas Munro wrote: >> This happens in both SERIALIZABLE and REPEATABLE READ when a single >> command inserts

Re: [GENERAL] ANN: Upscene releases Database Workbench 5.2.4

On Wed, Oct 12, 2016 at 2:51 AM, Martijn Tonies (Upscene Productions) < m.ton...@upscene.com> wrote: > Upscene Productions is proud to announce the availability of the next version of the popular multi-DBMS development tool: " Database Workbench 5.2.4 " The 5.2 release includes

Re: [GENERAL] My DB has has 5TB, many operations are very slow (on Google Cloud Compute)

Melvin: On Tue, Oct 11, 2016 at 9:50 PM, Melvin Davidson wrote: > >Requiring and exclusive table lock does not imply slownes. Just try > >'lock table x in exclusive mode' on an idle system. Pretty fast. > Sure on an idle system, you will get a table lock right away, but

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On 10/12/16, Thomas Munro wrote: > On Wed, Oct 12, 2016 at 8:50 PM, Albe Laurenz > wrote: >> Kevin Grittner wrote: >>> On Tue, Oct 11, 2016 at 2:29 PM, Jason Dusek >>> wrote: I notice the following oddity: >>>

Re: [GENERAL] confusion about user paring with pg_hba and pg_ident

On 10/12/2016 01:30 AM, arnaud gaboury wrote: On Tue, Oct 11, 2016 at 4:20 PM Tom Lane > wrote: arnaud gaboury > writes: > I am a little confused about some of my settings

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Wed, Oct 12, 2016 at 12:45 PM, Kevin Grittner wrote: >> The test in ExecCheckHeapTupleVisible() seems wrong to me. It's >> not immediately obvious what the proper fix is. (prune old e-mail address from cc list) I agree that the multi-value case is a bug. > To identify

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Thu, Oct 13, 2016 at 8:45 AM, Kevin Grittner wrote: > On Wed, Oct 12, 2016 at 10:06 AM, Kevin Grittner wrote: > >> The test in ExecCheckHeapTupleVisible() seems wrong to me. It's >> not immediately obvious what the proper fix is. > > To identify what

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Wed, Oct 12, 2016 at 1:05 PM, Thomas Munro wrote: > Here's a patch that shows one way to fix it. I think it does make > sense to change this, because otherwise automatic > retry-on-serialization-failure strategies will be befuddle by this > doomed transaction.

Re: [GENERAL] ANN: Upscene releases Database Workbench 5.2.4

Upscene Productions is proud to announce the availability of the next version of the popular multi-DBMS development tool: " Database Workbench 5.2.4 " The 5.2 release includes support for PostgreSQL and adds several other features and bugfixes. Database Workbench 5 comes in multiple editions with

Re: [GENERAL] confusion about user paring with pg_hba and pg_ident

On Wed, Oct 12, 2016 at 3:41 PM Adrian Klaver wrote: > On 10/12/2016 02:07 AM, arnaud gaboury wrote: > > > > > > > > > On Tue, Oct 11, 2016 at 4:20 PM Tom Lane > > > wrote: > > > > > > arnaud gaboury

[GENERAL] MultiXact member wraparound protections are disabled

Hi, We run postgres 9.4.5. Starting this morning, we started seeing messages like the below: Oct 12 14:07:15 site-db01a postgres[11253]: [106430-1] app=,user=,db=,ip=LOG: MultiXact member wraparound protections are disabled because oldest checkpointed MultiXact 1 does not exist on disk Oct 12

Re: [GENERAL] MultiXact member wraparound protections are disabled

AnandKumar, Karthik wrote: > Hi, > > We run postgres 9.4.5. > > Starting this morning, we started seeing messages like the below: > Oct 12 14:07:15 site-db01a postgres[11253]: [106430-1] app=,user=,db=,ip=LOG: > MultiXact member wraparound protections are disabled because oldest >

Re: [GENERAL] MultiXact member wraparound protections are disabled

root@site-db01a:/var/lib/pgsql/cmates/data # ls pg_multixact/members 0001 0002 0003 0004 0005 0006 0007 0008 0009 000A 000B 000C 000D 000E 000F 0010 0011 0012 0013 0014 0015 0016 0017 0018 0019 001A 001B root@site-db01a:/var/lib/pgsql/cmates/data # ls

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Wed, Oct 12, 2016 at 2:06 PM, Kevin Grittner wrote: > If the "proper" fix is impossible (or just too freaking ugly) we > might fall back on the fix Thomas suggested, but I would like to > take advantage of the "special properties" of the INSERT/ON > CONFLICT DO NOTHING code

Re: [GENERAL] Postgres UPGRADE from 9.2 to 9.4

2016-09-09 1:09 GMT+12:00 Scott Marlowe : > On Tue, Sep 6, 2016 at 5:25 PM, John R Pierce wrote: > > On 9/6/2016 4:20 PM, Melvin Davidson wrote: > >> > >> If you use slony to replicate, you CAN have 9.2 on the master and 9.4 on > >> the slave. > > >

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Thu, Oct 13, 2016 at 10:06 AM, Kevin Grittner wrote: > On Wed, Oct 12, 2016 at 3:02 PM, Peter Geoghegan wrote: > >> I agree that the multi-value case is a bug. > >> I think that it should be pretty obvious to you why the check exists >> at all, Kevin. It

Re: [GENERAL] postgres failed to start from services manager on windows 2008 r2

Please let me know if any other details required, I'm still struck at same place. On Oct 6, 2016 12:07 PM, "PHANIKUMAR G" wrote: > hi Thomas, > > thanks for your reply. small correction for my previous reply, we use > postgresql zip and bundle with our products

Re: [GENERAL] MultiXact member wraparound protections are disabled

Sharing output postgres@site-db01a:~/cmates/data/pg_multixact/members $ ls 0002 0004 0006 0008 000A 000C 000E 0010 0012 0014 0016 0018 001A 0001 0003 0005 0007 0009 000B 000D 000F 0011 0013 0015 0017 0019 001B postgres@site-db01a:~/cmates/data/pg_multixact/offsets $

Re: [GENERAL] MultiXact member wraparound protections are disabled

We are also seeing this in our log file Oct 12 19:08:14 site-db01a postgres[6117]: [7589-1] app=,user=,db=,ip=LOG: MultiXact member wraparound protections are disabled because oldest checkpointed MultiXact 1 does not exist on disk On Wed, Oct 12, 2016 at 7:13 PM, avi Singh

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Wed, Oct 12, 2016 at 6:06 PM, Thomas Munro wrote: > But yeah, the existing code raises false positive serialization > failures under SERIALIZABLE, and that's visible in the isolation test > I posted: there is actually a serial order of those transactions with >

Re: [GENERAL] MultiXact member wraparound protections are disabled

Got the output of pg_control postgres@site-db01a:~/cmates/data/global $ /usr/pgsql-9.4/bin/pg_controldata /var/lib/pgsql/cmates/data pg_control version number:942 Catalog version number: 201409291 Database system identifier: 6228991221455883206 Database

Re: [GENERAL] SERIALIZABLE and INSERTs with multiple VALUES

On Thu, Oct 13, 2016 at 2:32 PM, Peter Geoghegan wrote: > On Wed, Oct 12, 2016 at 6:06 PM, Thomas Munro > wrote: >> But yeah, the existing code raises false positive serialization >> failures under SERIALIZABLE, and that's visible in the isolation