Обсуждение: Beta 1 Release Draft
All, So, I've taken another tack on this. I realized that the important thing about the beta was not to provide narrative marketing information about the beta features, but rather to get our users to test the new features. This means, among other things, listing as many of them as possible, rather than highlighting a few "big" ones. We'll do that for the final release. So, here's a draft based on that idea. Please provide feedback, improvements, and corrections as soon as you can; we may need to send this out as early as Thursday. ============================ PostgreSQL 9.3 Beta 1 Released ============================= The first beta release of PostgreSQL 9.3, the latest version of the world's best open source database, is now available. This beta contains previews of all of the features which will be available in version 9.3, and is ready for testing by the worldwide PostgreSQL community. Please download, test, and report what you find. Major Features -------------- The major features available for testing in this beta include: * Writeable Foreign Tables * pgsql_fdw driver for federation of PostgreSQL databases * Automatically updatable VIEWs * MATERIALIZED VIEW declaration * LATERAL JOINs * Additional JSON constructor and extractor functions * Indexed regular expression search * Disk page checksums to detect filesystem failures In 9.3, PostgreSQL has switched from using SysV shared memory to using Posix shared memory and mmap for memory management. This allows easier installation and configuration of PostgreSQL, but means that we need our users to rigorously test and ensure that no memory management issues have been introduced by the change. Additional Features ------------------- Additional features included in this release are: * Streaming-only remastering of replicas * Performance and locking improvements for Foreign Key locks * Parallel pg_dump for faster backups * Directories for configuration files * pg_isready database connection checker * 64-bit Large Object API * COPY FREEZE for reduced IO bulk loading * User-defined background workers for automating database tasks * Recursive view declaration For a full listing of the features in version 9.3 Beta, please see the [release notes](http://www.postgresql.org/docs/devel/static/release-9-3.html). Additional descriptions and notes on some of -the new features are available on the [9.3 Features Wiki Page](http://wiki.postgresql.org/wiki/PostgreSQL_9.3_Features). Test 9.3 Beta 1 Now ------------------- We depend on our community to help test the next version in order to guarantee that it is high-performance and bug-free. Please download PostgreSQL 9.3 Beta 1 and try it with your workloads and applications as soon as you can, and give feedback to the PostgreSQL developers. Features and APIs in Beta 1 will not change substantially before final release, so it is now safe to start building applications against the new features. [More information on how to test and report issues](http://www.postgresql.org/developer/beta) Get the PostgreSQL 9.3 Beta 1, including binaries and installers for Windows, Linux and Mac from [our download page](http://www.postgresql.org/download). Full documentation of the new version [is available online](http://www.postgresql.org/docs/devel/static), and also installs with PostgreSQL. -- Josh Berkus PostgreSQL Experts Inc. http://pgexperts.com
On Apr 27, 2013, at 7:53 PM, Josh Berkus wrote: > All, > > So, I've taken another tack on this. I realized that the important > thing about the beta was not to provide narrative marketing information > about the beta features, but rather to get our users to test the new > features. This means, among other things, listing as many of them as > possible, rather than highlighting a few "big" ones. We'll do that for > the final release. > > So, here's a draft based on that idea. Please provide feedback, > improvements, and corrections as soon as you can; we may need to send > this out as early as Thursday. > > ============================ > > PostgreSQL 9.3 Beta 1 Released > ============================= > > The first beta release of PostgreSQL 9.3, the latest version of the > world's best open source database, is now available. This beta contains > previews of all of the features which will be available in version 9.3, > and is ready for testing by the worldwide PostgreSQL community. Please > download, test, and report what you find. > > Major Features > -------------- > > The major features available for testing in this beta include: > > * Writeable Foreign Tables > * pgsql_fdw driver for federation of PostgreSQL databases > * Automatically updatable VIEWs > * MATERIALIZED VIEW declaration > * LATERAL JOINs > * Additional JSON constructor and extractor functions > * Indexed regular expression search > * Disk page checksums to detect filesystem failures > > In 9.3, PostgreSQL has switched from using SysV shared memory to using > Posix shared memory and mmap for memory management. This allows easier > installation and configuration of PostgreSQL, but means that we need our > users to rigorously test and ensure that no memory management issues > have been introduced by the change. > > Additional Features > ------------------- > > Additional features included in this release are: > > * Streaming-only remastering of replicas > * Performance and locking improvements for Foreign Key locks > * Parallel pg_dump for faster backups > * Directories for configuration files > * pg_isready database connection checker > * 64-bit Large Object API > * COPY FREEZE for reduced IO bulk loading > * User-defined background workers for automating database tasks > * Recursive view declaration > > For a full listing of the features in version 9.3 Beta, please see the > [release > notes](http://www.postgresql.org/docs/devel/static/release-9-3.html). > Additional descriptions and notes on some of -the new features are > available on the [9.3 Features Wiki > Page](http://wiki.postgresql.org/wiki/PostgreSQL_9.3_Features). > > Test 9.3 Beta 1 Now > ------------------- > > We depend on our community to help test the next version in order to > guarantee that it is high-performance and bug-free. Please download > PostgreSQL 9.3 Beta 1 and try it with your workloads and applications as > soon as you can, and give feedback to the PostgreSQL developers. > Features and APIs in Beta 1 will not change substantially before final > release, so it is now safe to start building applications against the > new features. [More information on how to test and report > issues](http://www.postgresql.org/developer/beta) > > Get the PostgreSQL 9.3 Beta 1, including binaries and installers for > Windows, Linux and Mac from [our download > page](http://www.postgresql.org/download). > > Full documentation of the new version [is available > online](http://www.postgresql.org/docs/devel/static), and also installs > with PostgreSQL. +1 I think this is very good. Jonathan
On 28 April 2013 00:53, Josh Berkus <josh@agliodbs.com> wrote: > The major features available for testing in this beta include: > > * Writeable Foreign Tables Can we describe that using a phrase/sentence that the outside world will understand? Something about Distributed Database or similar? > Additional features included in this release are: Fast failover is an important feature as well. -- Simon Riggs http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services
On 04/28/2013 12:57 PM, Simon Riggs wrote: > On 28 April 2013 00:53, Josh Berkus <josh@agliodbs.com> wrote: > >> The major features available for testing in this beta include: >> >> * Writeable Foreign Tables > > Can we describe that using a phrase/sentence that the outside world > will understand? > Something about Distributed Database or similar? Ah, I originally had a longer sentence. Thoughts on some way to sum it up in 10 words or less? >> Additional features included in this release are: > > > Fast failover is an important feature as well. Right. Do we just call that "fast failover"? -- Josh Berkus PostgreSQL Experts Inc. http://pgexperts.com
On Sat, Apr 27, 2013 at 04:53:24PM -0700, Josh Berkus wrote: > In 9.3, PostgreSQL has switched from using SysV shared memory to using > Posix shared memory and mmap for memory management. This allows easier > installation and configuration of PostgreSQL, but means that we need our > users to rigorously test and ensure that no memory management issues > have been introduced by the change. I am not aware of any additional memory testing available, unless you are talkinga about mmap() perhaps being slower than SysV shared memory on some operating systems. Is that it? -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On 05/02/2013 05:52 PM, Bruce Momjian wrote: > On Sat, Apr 27, 2013 at 04:53:24PM -0700, Josh Berkus wrote: >> In 9.3, PostgreSQL has switched from using SysV shared memory to using >> Posix shared memory and mmap for memory management. This allows easier >> installation and configuration of PostgreSQL, but means that we need our >> users to rigorously test and ensure that no memory management issues >> have been introduced by the change. > > I am not aware of any additional memory testing available, unless you > are talkinga about mmap() perhaps being slower than SysV shared memory > on some operating systems. Is that it? Yeah, or something we didn't anticipate, like Linux swapping the RAM out when we're still using it. -- Josh Berkus PostgreSQL Experts Inc. http://pgexperts.com
On Thu, May 2, 2013 at 06:27:03PM -0700, Josh Berkus wrote: > On 05/02/2013 05:52 PM, Bruce Momjian wrote: > > On Sat, Apr 27, 2013 at 04:53:24PM -0700, Josh Berkus wrote: > >> In 9.3, PostgreSQL has switched from using SysV shared memory to using > >> Posix shared memory and mmap for memory management. This allows easier > >> installation and configuration of PostgreSQL, but means that we need our > >> users to rigorously test and ensure that no memory management issues > >> have been introduced by the change. > > > > I am not aware of any additional memory testing available, unless you > > are talkinga about mmap() perhaps being slower than SysV shared memory > > on some operating systems. Is that it? > > Yeah, or something we didn't anticipate, like Linux swapping the RAM out > when we're still using it. OK, my question is whether it deserves special mention as a testing area. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
>> Yeah, or something we didn't anticipate, like Linux swapping the RAM out >> when we're still using it. > > OK, my question is whether it deserves special mention as a testing > area. Well, I'll admit to having an ulterior motive here. The SHMMAX/SHMALL thing has been a constant irritation for both sysadmins and developers installing on their laptops (especially Mac laptops) for so long that I want to give it special mention, but it's not really a "major feature". This was the compromise we worked out on -advocacy. Also, we don't *know* that the switch to mmap() won't introduce new memory issues we haven't thought of, and if it does, they'll be on offbeat platforms/configurations we don't test. So I do think there's a call for extra testing around this area. BTW, please remember to credit Agent M along with Robert Haas for that feature. -- Josh Berkus PostgreSQL Experts Inc. http://pgexperts.com
On Fri, May 3, 2013 at 09:37:24AM -0700, Josh Berkus wrote: > > >> Yeah, or something we didn't anticipate, like Linux swapping the RAM out > >> when we're still using it. > > > > OK, my question is whether it deserves special mention as a testing > > area. > > Well, I'll admit to having an ulterior motive here. The SHMMAX/SHMALL > thing has been a constant irritation for both sysadmins and developers > installing on their laptops (especially Mac laptops) for so long that I > want to give it special mention, but it's not really a "major feature". > This was the compromise we worked out on -advocacy. > > Also, we don't *know* that the switch to mmap() won't introduce new > memory issues we haven't thought of, and if it does, they'll be on > offbeat platforms/configurations we don't test. So I do think there's a > call for extra testing around this area. > > BTW, please remember to credit Agent M along with Robert Haas for that > feature. Uh, I see no "Agent M" mentioned the commit logs. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
> Uh, I see no "Agent M" mentioned the commit logs. Well, the original patch for moving to mmap() was submitted by Agent M for PostgreSQL 9.2. I believe that Robert Haas used some of that code in his final patch; even if he didn't, the spur for him to work on it was Agent M's patch. Thread: http://www.postgresql.org/message-id/15869.1340753409@sss.pgh.pa.us -- Josh Berkus PostgreSQL Experts Inc. http://pgexperts.com
On Fri, May 3, 2013 at 10:00:46AM -0700, Josh Berkus wrote: > > > Uh, I see no "Agent M" mentioned the commit logs. > > Well, the original patch for moving to mmap() was submitted by Agent M > for PostgreSQL 9.2. I believe that Robert Haas used some of that code > in his final patch; even if he didn't, the spur for him to work on it > was Agent M's patch. > > Thread: http://www.postgresql.org/message-id/15869.1340753409@sss.pgh.pa.us Well, it is really Robert's call to decide if "Agent M" is supposed to be credited. -- Bruce Momjian <bruce@momjian.us> http://momjian.us EnterpriseDB http://enterprisedb.com + It's impossible for everything to be true. +
On 05/03/2013 10:11 AM, Bruce Momjian wrote: > On Fri, May 3, 2013 at 10:00:46AM -0700, Josh Berkus wrote: >> >>> Uh, I see no "Agent M" mentioned the commit logs. >> >> Well, the original patch for moving to mmap() was submitted by Agent M >> for PostgreSQL 9.2. I believe that Robert Haas used some of that code >> in his final patch; even if he didn't, the spur for him to work on it >> was Agent M's patch. >> >> Thread: http://www.postgresql.org/message-id/15869.1340753409@sss.pgh.pa.us > > Well, it is really Robert's call to decide if "Agent M" is supposed to > be credited. Yeah, that's why I cc'd him. -- Josh Berkus PostgreSQL Experts Inc. http://pgexperts.com
On 28 April 2013 20:57, Simon Riggs <simon@2ndquadrant.com> wrote: >> Additional features included in this release are: > > > Fast failover is an important feature as well. Can we add that to the list please? -- Simon Riggs http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services
>> Fast failover is an important feature as well. > > Can we add that to the list please? Yes, can you give me a < 12-word summary of the feature? Just "Fast Failover" doesn't quite describe it. It would also be helpful to have something on the "What's New" page about this. -- Josh Berkus PostgreSQL Experts Inc. http://pgexperts.com
On 7 May 2013 02:05, Josh Berkus <josh@agliodbs.com> wrote: > >>> Fast failover is an important feature as well. >> >> Can we add that to the list please? > > Yes, can you give me a < 12-word summary of the feature? Just "Fast > Failover" doesn't quite describe it. > > It would also be helpful to have something on the "What's New" page > about this. Standby promotion is almost instant, allowing 99.999% availability for a replicated cluster. -- Simon Riggs http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services
Josh Berkus wrote: > All, > > So, I've taken another tack on this. I realized that the important > thing about the beta was not to provide narrative marketing information > about the beta features, but rather to get our users to test the new > features. Great, so did you add a line about testing the new FK locking? I think it belongs in "major features". -- Álvaro Herrera http://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Training & Services