postgres-xc-developers Mailing List for Postgres-XC

On Thu, May 31, 2012 at 8:50 PM, Ashutosh Bapat <
ash...@en...> wrote:

> Hi All,
> If I run make installcheck successively on the same data cluster, I get
> GTM error "ERROR:  GTM error, could not create sequence", whenever a
> sequence is attempted to be created. It comes because open_sequence() and
> in turn CreateSequenceGTM return non-zero value. This code does not report
> why the GTM threw error.
>
> Is anybody else seeing this error? Can we improve code here, to report the
> exact cause of the failure. The error above is too vague for user to take
> corrective measure.

I have never seen it.
Yes this code could be improved, we could add an enum with the list of
possible errors and return that to Coordinator.
The possible errors I have in mind are:
- GTM internal error
- sequence already existing on GTM
-- 
Michael Paquier
http://michael.otacoo.com

Hi All,
If I run make installcheck successively on the same data cluster, I get GTM
error "ERROR:  GTM error, could not create sequence", whenever a sequence
is attempted to be created. It comes because open_sequence() and in turn
CreateSequenceGTM return non-zero value. This code does not report why the
GTM threw error.

Is anybody else seeing this error? Can we improve code here, to report the
exact cause of the failure. The error above is too vague for user to take
corrective measure.

-- 
Best Wishes,
Ashutosh Bapat
EntepriseDB Corporation
The Enterprise Postgres Company

Hi all,

There has been an important security fix in PostgreSQL repo regarding password management in pg_crypto.
https://github.com/postgres/postgres/commit/8db2e820c96a0ce33d3c15f476aa78bebce2c732
Here is the 9.1 stable commit.

We will need to incorporate this fix in branch 1.0 stable of Postgres-XC before releasing 1.0.0.
Hence, I am planning to merge xc's 1.0 stable with the latest PostgreSQL 9.1 stable point in the next couple of days.

Michael

Okay, let's do it.
----------
Koichi Suzuki


2012/5/31 Michael Paquier <mic...@gm...>:
>
>> Because DDL is automatically propagated to all the nodes, I think it
>> is reasonable to make pgxc_ddl as contrib module.   On the other hand,
>> pgxc_clean is AKO mandatory to cleanup the status after the crash so I
>> think it makes sense to make pgxc_clean as a core, I mean
>> automatically installed by make install and included in default client
>> installation.
>
> My point is that pgxc_clean has to be kicked by another utility (HA
> middleware, cron, etc.) and does not react directly with core code, so this
> tool looks like file_fdw or dblink where users can install it but it is not
> mandatory at all!
> Installing it with core means that we oblige people to clean up 2pc
> functionalities with pgxc_clean, which I believe we do not want to do.
> So for simplicity's sake, it should be moved to contrib.
> --
> Michael Paquier
> http://michael.otacoo.com

> Because DDL is automatically propagated to all the nodes, I think it
> is reasonable to make pgxc_ddl as contrib module.   On the other hand,
> pgxc_clean is AKO mandatory to cleanup the status after the crash so I
> think it makes sense to make pgxc_clean as a core, I mean
> automatically installed by make install and included in default client
> installation.
>
My point is that pgxc_clean has to be kicked by another utility (HA
middleware, cron, etc.) and does not react directly with core code, so this
tool looks like file_fdw or dblink where users can install it but it is not
mandatory at all!
Installing it with core means that we oblige people to clean up 2pc
functionalities with pgxc_clean, which I believe we do not want to do.
So for simplicity's sake, it should be moved to contrib.
-- 
Michael Paquier
http://michael.otacoo.com

Because DDL is automatically propagated to all the nodes, I think it
is reasonable to make pgxc_ddl as contrib module.   On the other hand,
pgxc_clean is AKO mandatory to cleanup the status after the crash so I
think it makes sense to make pgxc_clean as a core, I mean
automatically installed by make install and included in default client
installation.

Regards;
----------
Koichi Suzuki


2012/5/31 Michael Paquier <mic...@gm...>:
> Hi all,
>
> pgxc_clean and pgxc_ddl are installed with core features, but they are not
> mandatory things.
> It would make sense to move them to contrib modules for clarity.
> Opinions?
> --
> Michael Paquier
> http://michael.otacoo.com
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Postgres-xc-developers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers
>

Hi all,

pgxc_clean and pgxc_ddl are installed with core features, but they are not
mandatory things.
It would make sense to move them to contrib modules for clarity.
Opinions?
-- 
Michael Paquier
http://michael.otacoo.com

Michael-san,

Thank you for your advising how to report problems and to place
questions. We understood your points; writing in English and in a plain
text (not dedicated format like as I told).

Hemmi-san, who is a member of XC peripheral tools, has writen some (bug)
reports to the Postgres-xc-bugs mailing list in English and in plain texts.

best regards,

Tetsuo Sakata.


(2012/05/26 12:47), Michael Paquier wrote:
> 
> 
> On Fri, May 25, 2012 at 6:04 PM, 坂田　哲夫 <sak...@la... 
> <mailto:sak...@la...>> wrote:
> 
>     Suzuki-san,
> 
>     Now we are developing XC's peripheral tools like as back-up, performance
>     observation, fail over cluster. While developing, the tool development
>     team would ask some questions to XC core developing team.
> 
> People are free to ask questions to developers on this mailing list.
> So it can of course freely be used as long as those questions are really 
> related to core functionalities, features, patches or performance. Well 
> everything really related with the core of XC.
> 
> In case the request is more general, I would recommend to use the 
> general mailing list pos...@li... 
> <mailto:pos...@li...>.
> For questions regarding tools, it depends but you should orientate your 
> question to the general mailing list... If there are cases where the 
> hackers ML (developer ML) is necessary you can also use it as well if 
> you think so.
> 
>     In such a cases, we (NTT people) used to placing questions with 'Qustion
>     application form' which includes version of the system in question,
>     environment information (OS name and its version, hardware
>     configuration), operation sequence and its results and so on. (In short,
>     I mean something like NTT's "Bug-hyo").
>     Because it is convenient for us to inform a reciever of things needed
>     completely and to manage each questions easily.
> 
> I believe that people using Postgres-XC mailing lists are free to use 
> the format they wish. However to keep simplicity, and I personally love 
> simple things, people answering such formatted requests are not obliged 
> to follow a special format themselves.
> All the responses are kept on the same mail thread, so this is enough I 
> believe. Keeping a certain freedom when writing emails here is kind of 
> really important as the project is community-based, and basically should 
> stay independent from any external formatting.
> 
> If you report a bug, I strongly recommend to use the bug mailing list 
> pos...@li... 
> <mailto:pos...@li...> with the bug template 
> located in doc-xc/bug.template in source. This bug template is already 
> formatted and has all the fields already related to OS, version, etc. So 
> you should definitely use that. The bug template is based on the same 
> one as PostgreSQL, so as I am sure NTT guys are familiar with Postgres 
> it won't be a huge effort to adapt to it.
> 
>     I was wondering if you could accept the questions placed with such an
>     application form and write answers with it.
> 
> Once again, people willing to response to mails on this hackers ML are 
> not obliged to follow a special format. If they wish to answer without 
> any application form, they can do so. If they wish to answer with a 
> given application form, well they are free to follow anything they wish. 
> Personally I don't think an application form will permit to gain time, 
> and email threads are enough. We have also email archives available with 
> SourceForge services.
> So, if it is a bug, please respect the template in doc-xc/bug.template 
> and send it to pos...@li....If it is anything 
> else, at least in my opinion people are free to answer and interact with 
> XC's mailing lists as they wish.
> -- 
> Michael Paquier
> http://michael.otacoo.com


-- 
坂田 哲夫＠NTTオープンソースソフトウェアセンタ
電話：○三・五八六○・五一一五(代)
sakata.tetsuo _at_ lab.ntt.co.jp ☆メアド変わりました oss→lab☆
SAKATA, Tetsuo. Shinagawa Tokyo JAPAN.

On Wed, May 30, 2012 at 6:58 AM, Niekas <gi...@li...> wrote:

> Hi,
> I have submitted a pull request to the Homebrew project for adding a
> formula that builds Postgres-XC 1.0beta2 on OS X.
> The pull request is at https://github.com/mxcl/homebrew/pull/12494.
>
> For those of you who are not familiar with it, Homebrew is a popular
> package manager for OS X. To install Postgres-XC you just have to type:
>
> brew pull https://github.com/mxcl/homebrew/pull/12494
> brew install postgres-xc
>
> (The first command is needed as long as the pull request is pending.)
>
> If you use a Mac, I invite you to test it and provide feedback (if you
> provide positive feedback on GitHub, maybe they’ll merge the request
> quickly).
>
This looks nice.
But you shouldn't add the version beta2, they will have problems to compile
the code. We fixed the problem since beta2, and it will be OK when 1.0.0 is
out.
So you should wait a little bit...
-- 
Michael Paquier
http://michael.otacoo.com

Hi,
I have submitted a pull request to the Homebrew project for adding a formula that builds Postgres-XC 1.0beta2 on OS X.
The pull request is at https://github.com/mxcl/homebrew/pull/12494.

For those of you who are not familiar with it, Homebrew is a popular package manager for OS X. To install Postgres-XC you just have to type:

brew pull https://github.com/mxcl/homebrew/pull/12494
brew install postgres-xc

(The first command is needed as long as the pull request is pending.)

If you use a Mac, I invite you to test it and provide feedback (if you provide positive feedback on GitHub, maybe they’ll merge the request quickly).

Niekas

Each Postgres-XC node has to use a different data folder.
Please refer to the document start_pgxc_in_10_commands.txt that you can
find in here:
https://sourceforge.net/projects/postgres-xc/files/Publication/

By looking at this file, you can notice that the data folder of each node
is *different*:
2) initgtm -Z gtm -D gtm
3) initdb -D datanode1 --nodename dn1 #Initialize Datanode 1
4) initdb -D datanode2 --nodename dn2 #Initialize Datanode 2
5) initdb -D coord1 --nodename co1 # Initialize Coordinator 1

Similar to PostgreSQL, you need to specify different data folders for slave
and master nodes if they are on the same servers.

On Tue, May 29, 2012 at 4:33 PM, MOHD ARSHAD SALEEM <
mar...@ta...> wrote:

>  i tried with similar option but again it showing same error
>
> following command i run ->
> > postgres -X  -p 5433 -D /usr/local/pgsql/data -i
>
> Error which i was getting->
> > FATAL:  lock file "postmaster.pid" already exists
> > HINT:  Is another postmaster (PID 2700) running in data directory
> "/usr/local/pgsql/data"?
>
> Regards
> Arshad
>  ------------------------------
> *From:* Michael Paquier [mic...@gm...]
> *Sent:* Tuesday, May 29, 2012 12:50 PM
> *To:* MOHD ARSHAD SALEEM
> *Cc:* pos...@li...
> *Subject:* Re: [Postgres-xc-developers] Query Related to Postgresql-XC
>
>  On Mon, May 28, 2012 at 10:51 PM, MOHD ARSHAD SALEEM <
> mar...@ta...> wrote:
>
>>  Hi Team,
>>
>> I was working on postgresql setup and try to run coordinate and datanode
>> services on same machine.As per the postgresql.conf file default port
>> value  is using for first service(coordinator).whenever  i try to start
>> other service(Datanode) it showing error.how to give one more port number
>> value to start datanode service.
>>
> You can specify a port number by changing the value of parameter port in
> postgresql.conf in the data folder of each node.
> You can also specify a value when starting a node with binary postgres
> with its option "-p". For example:
> postgres -C -p 5433 -D $DATA
> Are you familiar with PostgreSQL?
>  --
> Michael Paquier
> http://michael.otacoo.com
>



-- 
Michael Paquier
http://michael.otacoo.com

i tried with similar option but again it showing same error

following command i run ->
> postgres -X  -p 5433 -D /usr/local/pgsql/data -i

Error which i was getting->
> FATAL:  lock file "postmaster.pid" already exists
> HINT:  Is another postmaster (PID 2700) running in data directory "/usr/local/pgsql/data"?

Regards
Arshad
________________________________
From: Michael Paquier [mic...@gm...]
Sent: Tuesday, May 29, 2012 12:50 PM
To: MOHD ARSHAD SALEEM
Cc: pos...@li...
Subject: Re: [Postgres-xc-developers] Query Related to Postgresql-XC

On Mon, May 28, 2012 at 10:51 PM, MOHD ARSHAD SALEEM <mar...@ta...<mailto:mar...@ta...>> wrote:
Hi Team,

I was working on postgresql setup and try to run coordinate and datanode services on same machine.As per the postgresql.conf file default port value  is using for first service(coordinator).whenever  i try to start other service(Datanode) it showing error.how to give one more port number value to start datanode service.
You can specify a port number by changing the value of parameter port in postgresql.conf in the data folder of each node.
You can also specify a value when starting a node with binary postgres with its option "-p". For example:
postgres -C -p 5433 -D $DATA
Are you familiar with PostgreSQL?
--
Michael Paquier
http://michael.otacoo.com

On Mon, May 28, 2012 at 10:51 PM, MOHD ARSHAD SALEEM <
mar...@ta...> wrote:

>  Hi Team,
>
> I was working on postgresql setup and try to run coordinate and datanode
> services on same machine.As per the postgresql.conf file default port
> value  is using for first service(coordinator).whenever  i try to start
> other service(Datanode) it showing error.how to give one more port number
> value to start datanode service.
>
You can specify a port number by changing the value of parameter port in
postgresql.conf in the data folder of each node.
You can also specify a value when starting a node with binary postgres with
its option "-p". For example:
postgres -C -p 5433 -D $DATA
Are you familiar with PostgreSQL?
-- 
Michael Paquier
http://michael.otacoo.com

> The error is related to port no only in postgresql.conf the port no(5432) gives to start one of the service(coordinator or datanode).
> how to start another service.because port no is already in use by the service.
> In postgres-XC document they wrote you have to use different port no for coordinator node and datanode 
________________________________________




From: mas...@gm... [mas...@gm...] on behalf of Mason [ma...@us...]
Sent: Monday, May 28, 2012 10:42 PM
To: MOHD ARSHAD SALEEM
Cc: pos...@li...
Subject: Re: [Postgres-xc-developers] Query Related to Postgresql-XC

On Mon, May 28, 2012 at 9:51 AM, MOHD ARSHAD SALEEM
<mar...@ta...> wrote:
> Hi Team,
>
> I was working on postgresql setup and try to run coordinate and datanode
> services on same machine.As per the postgresql.conf file default port value
> is using for first service(coordinator).whenever  i try to start other
> service(Datanode) it showing error.how to give one more port number value to
> start datanode service.
>

What is the particular error?

Just make sure that you have unique ports configured in the
postgresql.conf file if they are running on the same system.  Note
that it if you are using 5432 and have a regular PostgreSQL instance
running, it could conflict with that.

We still need to review XC documentation with better examples than this one
which is wrong.
http://postgres-xc.sourceforge.net/docs/1_0/install-short.html

On Tue, May 29, 2012 at 10:03 AM, Koichi Suzuki <koi...@gm...>wrote:

> Please take a look at XC tutorial material available at
>
> https://sourceforge.net/projects/postgres-xc/files/Presentation/20120516_PGConTutorial/
>
> Also, Michael's material will be found at
> https://sourceforge.net/projects/postgres-xc/files/Publication/
>
> Please try start_pgxc_in_10_commands.txt
>
> Hope they help.
>
> Regards;
> ----------
> Koichi Suzuki
>
>
> 2012/5/28 MOHD ARSHAD SALEEM <mar...@ta...>:
> > Hi Team,
> >
> > I was working on postgresql setup and try to run coordinate and datanode
> > services on same machine.As per the postgresql.conf file default port
> value
> > is using for first service(coordinator).whenever  i try to start other
> > service(Datanode) it showing error.how to give one more port number
> value to
> > start datanode service.
> >
> > Regards
> > Arshad
> >
> >
> ------------------------------------------------------------------------------
> > Live Security Virtual Conference
> > Exclusive live event will cover all the ways today's security and
> > threat landscape has changed and how IT managers can respond. Discussions
> > will include endpoint security, mobile security and the latest in malware
> > threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> > _______________________________________________
> > Postgres-xc-developers mailing list
> > Pos...@li...
> > https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers
> >
>
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Postgres-xc-developers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers
>



-- 
Michael Paquier
http://michael.otacoo.com

Please take a look at XC tutorial material available at
https://sourceforge.net/projects/postgres-xc/files/Presentation/20120516_PGConTutorial/

Also, Michael's material will be found at
https://sourceforge.net/projects/postgres-xc/files/Publication/

Please try start_pgxc_in_10_commands.txt

Hope they help.

Regards;
----------
Koichi Suzuki


2012/5/28 MOHD ARSHAD SALEEM <mar...@ta...>:
> Hi Team,
>
> I was working on postgresql setup and try to run coordinate and datanode
> services on same machine.As per the postgresql.conf file default port value
> is using for first service(coordinator).whenever  i try to start other
> service(Datanode) it showing error.how to give one more port number value to
> start datanode service.
>
> Regards
> Arshad
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Postgres-xc-developers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers
>

On Mon, May 28, 2012 at 9:51 AM, MOHD ARSHAD SALEEM
<mar...@ta...> wrote:
> Hi Team,
>
> I was working on postgresql setup and try to run coordinate and datanode
> services on same machine.As per the postgresql.conf file default port value
> is using for first service(coordinator).whenever  i try to start other
> service(Datanode) it showing error.how to give one more port number value to
> start datanode service.
>

What is the particular error?

Just make sure that you have unique ports configured in the
postgresql.conf file if they are running on the same system.  Note
that it if you are using 5432 and have a regular PostgreSQL instance
running, it could conflict with that.

Regards,

Mason

Hi Team,

I was working on postgresql setup and try to run coordinate and datanode services on same machine.As per the postgresql.conf file default port value  is using for first service(coordinator).whenever  i try to start other service(Datanode) it showing error.how to give one more port number value to start datanode service.

Regards
Arshad

On Mon, May 28, 2012 at 4:06 PM, Koichi Suzuki <koi...@gm...> wrote:

> ML in Japanese may consist of very few people and will not benefit.
> For example, I'm subscribing JPUG ML but it is not active at all.
>
> I think it is better to have all in the same ML and encourage to post
> in English.   We can expect someone to volunteer to help
> communication.
>
OK. Let's simply avoid communication in Japanese.
-- 
Michael Paquier
http://michael.otacoo.com

ML in Japanese may consist of very few people and will not benefit.
For example, I'm subscribing JPUG ML but it is not active at all.

I think it is better to have all in the same ML and encourage to post
in English.   We can expect someone to volunteer to help
communication.

Regards;
----------
Koichi Suzuki


2012/5/28 Michael Paquier <mic...@gm...>:
>
>
> On Mon, May 28, 2012 at 9:56 AM, Koichi Suzuki <koi...@gm...> wrote:
>>
>> One more thing:
>>
>> At least, subject should be in English to make it easier to keep track
>> of each issue.
>
> Why not creating a new mailing list only for Japanese speakers?
> Most of the users here won't understand anything if the content of the
> message is in Japanese and they will could feel it as a spam.
> We should at least treat messages on separate mailing lists depending on
> language to avoid any inconvenience to any users that cannot speak Japanese.
> --
> Michael Paquier
> http://michael.otacoo.com

On Mon, May 28, 2012 at 9:56 AM, Koichi Suzuki <koi...@gm...> wrote:

> One more thing:
>
> At least, subject should be in English to make it easier to keep track
> of each issue.
>
Why not creating a new mailing list only for Japanese speakers?
Most of the users here won't understand anything if the content of the
message is in Japanese and they will could feel it as a spam.
We should at least treat messages on separate mailing lists depending on
language to avoid any inconvenience to any users that cannot speak Japanese.
-- 
Michael Paquier
http://michael.otacoo.com

One more thing:

At least, subject should be in English to make it easier to keep track
of each issue.

Regards;
----------
Koichi Suzuki


2012/5/26 Koichi Suzuki <koi...@gm...>:
> Thanks Michael for the reply.
>
> I understand NTT how NTT people are doing inside and I think you can
> keep using it inside.   On the other hand, to have response from many
> people, I have to say that Michael is right.    It will not be
> reasonable to convince the people to use this.   Instead, I advise to
> past your questions and reports to ML and paste the response to your
> form again.    I think this will keep things move forward quickly.
>
> If you need an answer to a question "Is it all?" or "Am I all set?",
> we will be more than happy to response.   I think this helps to close
> each of your format.
>
> What about the language?   Of course, English is better.   But if you
> feel it's not practical for the people to open an issue, Japanese may
> be okay.    We can translate into English to help people to respond if
> the number of issue is not too many.    I don't think you need
> responses translated into Japanese.
>
> Please let me know your idea.
>
> Regards;
> ----------
> Koichi Suzuki
>
>
> 2012/5/26 Michael Paquier <mic...@gm...>:
>>
>>
>> On Fri, May 25, 2012 at 6:04 PM, 坂田　哲夫 <sak...@la...> wrote:
>>>
>>> Suzuki-san,
>>>
>>> Now we are developing XC's peripheral tools like as back-up, performance
>>> observation, fail over cluster. While developing, the tool development
>>> team would ask some questions to XC core developing team.
>>
>> People are free to ask questions to developers on this mailing list.
>> So it can of course freely be used as long as those questions are really
>> related to core functionalities, features, patches or performance. Well
>> everything really related with the core of XC.
>>
>> In case the request is more general, I would recommend to use the general
>> mailing list pos...@li....
>> For questions regarding tools, it depends but you should orientate your
>> question to the general mailing list... If there are cases where the hackers
>> ML (developer ML) is necessary you can also use it as well if you think so.
>>
>>> In such a cases, we (NTT people) used to placing questions with 'Qustion
>>> application form' which includes version of the system in question,
>>> environment information (OS name and its version, hardware
>>> configuration), operation sequence and its results and so on. (In short,
>>> I mean something like NTT's "Bug-hyo").
>>> Because it is convenient for us to inform a reciever of things needed
>>> completely and to manage each questions easily.
>>
>> I believe that people using Postgres-XC mailing lists are free to use the
>> format they wish. However to keep simplicity, and I personally love simple
>> things, people answering such formatted requests are not obliged to follow a
>> special format themselves.
>> All the responses are kept on the same mail thread, so this is enough I
>> believe. Keeping a certain freedom when writing emails here is kind of
>> really important as the project is community-based, and basically should
>> stay independent from any external formatting.
>>
>> If you report a bug, I strongly recommend to use the bug mailing
>> list pos...@li... with the bug template located in
>> doc-xc/bug.template in source. This bug template is already formatted and
>> has all the fields already related to OS, version, etc. So you should
>> definitely use that. The bug template is based on the same one as
>> PostgreSQL, so as I am sure NTT guys are familiar with Postgres it won't be
>> a huge effort to adapt to it.
>>
>>>
>>> I was wondering if you could accept the questions placed with such an
>>> application form and write answers with it.
>>
>> Once again, people willing to response to mails on this hackers ML are not
>> obliged to follow a special format. If they wish to answer without any
>> application form, they can do so. If they wish to answer with a given
>> application form, well they are free to follow anything they wish.
>> Personally I don't think an application form will permit to gain time, and
>> email threads are enough. We have also email archives available with
>> SourceForge services.
>> So, if it is a bug, please respect the template in doc-xc/bug.template and
>> send it to pos...@li....If it is anything else, at
>> least in my opinion people are free to answer and interact with XC's mailing
>> lists as they wish.
>> --
>> Michael Paquier
>> http://michael.otacoo.com
>>
>> ------------------------------------------------------------------------------
>> Live Security Virtual Conference
>> Exclusive live event will cover all the ways today's security and
>> threat landscape has changed and how IT managers can respond. Discussions
>> will include endpoint security, mobile security and the latest in malware
>> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
>> _______________________________________________
>> Postgres-xc-bugs mailing list
>> Pos...@li...
>> https://lists.sourceforge.net/lists/listinfo/postgres-xc-bugs
>>

On Sat, May 26, 2012 at 6:33 PM, Koichi Suzuki <koi...@gm...> wrote:

> Thanks Michael for the reply.
>
> I understand NTT how NTT people are doing inside and I think you can
> keep using it inside.   On the other hand, to have response from many
> people, I have to say that Michael is right.    It will not be
> reasonable to convince the people to use this.   Instead, I advise to
> past your questions and reports to ML and paste the response to your
> form again.    I think this will keep things move forward quickly.
>
> If you need an answer to a question "Is it all?" or "Am I all set?",
> we will be more than happy to response.   I think this helps to close
> each of your format.
>
> What about the language?   Of course, English is better.   But if you
> feel it's not practical for the people to open an issue, Japanese may
> be okay.    We can translate into English to help people to respond if
> the number of issue is not too many.    I don't think you need
> responses translated into Japanese.
>
English is mandatory on this mailing list.
Without giving precise numbers about registered people, less than 10% of
the people registered to this mailing list can read and write Japanese.
AFAIK, I am the only non-national able to do that. So by sending messages
in Japanese to this mailing list, 90% of the people won't be able to get
what is spoken about because of the language barrier, That is honestly too
much. If some people feel more confortable in Japanese what we could do is
create a special mailing list only for Japanese speakers, this will avoid
any language inconvenience to the other users, and people willing to speak
in Japanese could do it freely there.

However you have to consider that you will really get an answer faster if
you send your messages in English here.
-- 
Michael Paquier
http://michael.otacoo.com

Thanks Michael for the reply.

I understand NTT how NTT people are doing inside and I think you can
keep using it inside.   On the other hand, to have response from many
people, I have to say that Michael is right.    It will not be
reasonable to convince the people to use this.   Instead, I advise to
past your questions and reports to ML and paste the response to your
form again.    I think this will keep things move forward quickly.

If you need an answer to a question "Is it all?" or "Am I all set?",
we will be more than happy to response.   I think this helps to close
each of your format.

What about the language?   Of course, English is better.   But if you
feel it's not practical for the people to open an issue, Japanese may
be okay.    We can translate into English to help people to respond if
the number of issue is not too many.    I don't think you need
responses translated into Japanese.

Please let me know your idea.

Regards;
----------
Koichi Suzuki


2012/5/26 Michael Paquier <mic...@gm...>:
>
>
> On Fri, May 25, 2012 at 6:04 PM, 坂田　哲夫 <sak...@la...> wrote:
>>
>> Suzuki-san,
>>
>> Now we are developing XC's peripheral tools like as back-up, performance
>> observation, fail over cluster. While developing, the tool development
>> team would ask some questions to XC core developing team.
>
> People are free to ask questions to developers on this mailing list.
> So it can of course freely be used as long as those questions are really
> related to core functionalities, features, patches or performance. Well
> everything really related with the core of XC.
>
> In case the request is more general, I would recommend to use the general
> mailing list pos...@li....
> For questions regarding tools, it depends but you should orientate your
> question to the general mailing list... If there are cases where the hackers
> ML (developer ML) is necessary you can also use it as well if you think so.
>
>> In such a cases, we (NTT people) used to placing questions with 'Qustion
>> application form' which includes version of the system in question,
>> environment information (OS name and its version, hardware
>> configuration), operation sequence and its results and so on. (In short,
>> I mean something like NTT's "Bug-hyo").
>> Because it is convenient for us to inform a reciever of things needed
>> completely and to manage each questions easily.
>
> I believe that people using Postgres-XC mailing lists are free to use the
> format they wish. However to keep simplicity, and I personally love simple
> things, people answering such formatted requests are not obliged to follow a
> special format themselves.
> All the responses are kept on the same mail thread, so this is enough I
> believe. Keeping a certain freedom when writing emails here is kind of
> really important as the project is community-based, and basically should
> stay independent from any external formatting.
>
> If you report a bug, I strongly recommend to use the bug mailing
> list pos...@li... with the bug template located in
> doc-xc/bug.template in source. This bug template is already formatted and
> has all the fields already related to OS, version, etc. So you should
> definitely use that. The bug template is based on the same one as
> PostgreSQL, so as I am sure NTT guys are familiar with Postgres it won't be
> a huge effort to adapt to it.
>
>>
>> I was wondering if you could accept the questions placed with such an
>> application form and write answers with it.
>
> Once again, people willing to response to mails on this hackers ML are not
> obliged to follow a special format. If they wish to answer without any
> application form, they can do so. If they wish to answer with a given
> application form, well they are free to follow anything they wish.
> Personally I don't think an application form will permit to gain time, and
> email threads are enough. We have also email archives available with
> SourceForge services.
> So, if it is a bug, please respect the template in doc-xc/bug.template and
> send it to pos...@li....If it is anything else, at
> least in my opinion people are free to answer and interact with XC's mailing
> lists as they wish.
> --
> Michael Paquier
> http://michael.otacoo.com
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Postgres-xc-bugs mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-bugs
>

On Fri, May 25, 2012 at 6:04 PM, 坂田 哲夫 <sak...@la...> wrote:

> Suzuki-san,
>
> Now we are developing XC's peripheral tools like as back-up, performance
> observation, fail over cluster. While developing, the tool development
> team would ask some questions to XC core developing team.
>
People are free to ask questions to developers on this mailing list.
So it can of course freely be used as long as those questions are really
related to core functionalities, features, patches or performance. Well
everything really related with the core of XC.

In case the request is more general, I would recommend to use the general
mailing list pos...@li....
For questions regarding tools, it depends but you should orientate your
question to the general mailing list... If there are cases where the
hackers ML (developer ML) is necessary you can also use it as well if you
think so.

In such a cases, we (NTT people) used to placing questions with 'Qustion
> application form' which includes version of the system in question,
> environment information (OS name and its version, hardware
> configuration), operation sequence and its results and so on. (In short,
> I mean something like NTT's "Bug-hyo").
> Because it is convenient for us to inform a reciever of things needed
> completely and to manage each questions easily.
>
I believe that people using Postgres-XC mailing lists are free to use the
format they wish. However to keep simplicity, and I personally love simple
things, people answering such formatted requests are not obliged to follow
a special format themselves.
All the responses are kept on the same mail thread, so this is enough I
believe. Keeping a certain freedom when writing emails here is kind of
really important as the project is community-based, and basically should
stay independent from any external formatting.

If you report a bug, I strongly recommend to use the bug mailing list
pos...@li... with the bug template located in
doc-xc/bug.template in source. This bug template is already formatted and
has all the fields already related to OS, version, etc. So you should
definitely use that. The bug template is based on the same one as
PostgreSQL, so as I am sure NTT guys are familiar with Postgres it won't be
a huge effort to adapt to it.


> I was wondering if you could accept the questions placed with such an
> application form and write answers with it.
>
Once again, people willing to response to mails on this hackers ML are not
obliged to follow a special format. If they wish to answer without any
application form, they can do so. If they wish to answer with a given
application form, well they are free to follow anything they wish.
Personally I don't think an application form will permit to gain time, and
email threads are enough. We have also email archives available with
SourceForge services.
So, if it is a bug, please respect the template in doc-xc/bug.template and
send it to pos...@li....If it is anything else,
at least in my opinion people are free to answer and interact with XC's
mailing lists as they wish.
-- 
Michael Paquier
http://michael.otacoo.com