postgres-xc-developers Mailing List for Postgres-XC

Hello,

I wonder if it makes sense to only allow long options after a short 
option. For example:

-XC --coordinator

So the only way to use the XC options is to pass -XC and then a long 
option. That should keep us completely clear of PostgreSQL.

Sincerely,

Joshua D. Drake

On Mon, Jul 30, 2012 at 7:17 PM, Ashutosh Bapat <
ash...@en...> wrote:

> We need to look at this in broader perspective. Using long options would
> fix the -C problem for now.

This option has been chosen long ago... This reminds me of the problem with
node list...
 I think that long options are enough safe.
Like for initdb with --nodename, I can pretty assure that options like
--coordinator and --datanode will never be used in Postgres.
So we should try to maximize the use of long options with keywords only
associated to XC.

But all the new options we would add will have this problem, at some or
> other point. Should we provide a solution which will take care of such
> problems for once and all? Changing user interfaces from release to release
> would alienate users from us.
>
Yes, and we also cannot forget that XC is directly a child of Postgres, so
in case of an option conflict, well postgres has to win and will win.
What we can do is choosing our new options enough wisely (launch options,
guc params, etc) to avoid any conflicts in future merges.
The past choice of -C and -X was indeed really bad as 9.2 merge is showing.


Hence, long options with really XC-specific keywords is a good deal. I am
open to other suggestions though.


> On Mon, Jul 30, 2012 at 7:01 AM, Michael Paquier <
> mic...@gm...> wrote:
>
>> Hi all,
>>
>> XC is now using -C to start up a node as a Coordinator, and -X to start
>> up a node as a Datanode.
>> In PostgreSQL 9.2, -C is now used for configuration parameters, so we
>> need to change the way we start nodes.
>>
>> In order to definitely take care of this problem that may happen in the
>> next release, I think we should change the start-up option
>> of XC nodes to long options, with --coordinator and --datanode. By using
>> that, we are sure that out options won't interact with postgres ones.
>>
>> Comments?
>> --
>> 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
>>
>>
>
>
> --
> Best Wishes,
> Ashutosh Bapat
> EntepriseDB Corporation
> The Enterprise Postgres Company
>
>


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

We need to look at this in broader perspective. Using long options would
fix the -C problem for now. But all the new options we would add will have
this problem, at some or other point. Should we provide a solution which
will take care of such problems for once and all? Changing user interfaces
from release to release would alienate users from us.

On Mon, Jul 30, 2012 at 7:01 AM, Michael Paquier
<mic...@gm...>wrote:

> Hi all,
>
> XC is now using -C to start up a node as a Coordinator, and -X to start up
> a node as a Datanode.
> In PostgreSQL 9.2, -C is now used for configuration parameters, so we need
> to change the way we start nodes.
>
> In order to definitely take care of this problem that may happen in the
> next release, I think we should change the start-up option
> of XC nodes to long options, with --coordinator and --datanode. By using
> that, we are sure that out options won't interact with postgres ones.
>
> Comments?
> --
> 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
>
>


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

On 07/29/2012 06:31 PM, Michael Paquier wrote:
> Hi all,
>
> XC is now using -C to start up a node as a Coordinator, and -X to start
> up a node as a Datanode.
> In PostgreSQL 9.2, -C is now used for configuration parameters, so we
> need to change the way we start nodes.
>
> In order to definitely take care of this problem that may happen in the
> next release, I think we should change the start-up option
> of XC nodes to long options, with --coordinator and --datanode. By using
> that, we are sure that out options won't interact with postgres ones.

Only long options works for me. I prefer verbose options anyway. It 
allows us to understand exactly what is going on.

Sincerely,

Joshua D. Drake



-- 
Command Prompt, Inc. - http://www.commandprompt.com/
PostgreSQL Support, Training, Professional Services and Development
High Availability, Oracle Conversion, Postgres-XC
@cmdpromptinc - 509-416-6579

On Mon, Jul 30, 2012 at 10:48 AM, Koichi Suzuki <ko...@in...>wrote:

> On Mon, 30 Jul 2012 10:31:38 +0900
> Michael Paquier <mic...@gm...> wrote:
>
> > Hi all,
> >
> > XC is now using -C to start up a node as a Coordinator, and -X to start
> up
> > a node as a Datanode.
> > In PostgreSQL 9.2, -C is now used for configuration parameters, so we
> need
> > to change the way we start nodes.
> >
> > In order to definitely take care of this problem that may happen in the
> > next release, I think we should change the start-up option
> > of XC nodes to long options, with --coordinator and --datanode. By using
> > that, we are sure that out options won't interact with postgres ones.
> >
> > Comments?
>
> So we're going to have only long options for this?   I've no objection to
> it anyway.
>
Yes, looks like a better long-term option than anything else.
However I am open to other ideas.

> Also, we're maintaining -Z option, right?
>
There is no problem on this side, so we can keep it as is.
-- 
Michael Paquier
http://michael.otacoo.com

On Mon, 30 Jul 2012 10:31:38 +0900
Michael Paquier <mic...@gm...> wrote:

> Hi all,
> 
> XC is now using -C to start up a node as a Coordinator, and -X to start up
> a node as a Datanode.
> In PostgreSQL 9.2, -C is now used for configuration parameters, so we need
> to change the way we start nodes.
> 
> In order to definitely take care of this problem that may happen in the
> next release, I think we should change the start-up option
> of XC nodes to long options, with --coordinator and --datanode. By using
> that, we are sure that out options won't interact with postgres ones.
> 
> Comments?

So we're going to have only long options for this?   I've no objection to it anyway.
Also, we're maintaining -Z option, right?

---
Koichi

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

Hi all,

XC is now using -C to start up a node as a Coordinator, and -X to start up
a node as a Datanode.
In PostgreSQL 9.2, -C is now used for configuration parameters, so we need
to change the way we start nodes.

In order to definitely take care of this problem that may happen in the
next release, I think we should change the start-up option
of XC nodes to long options, with --coordinator and --datanode. By using
that, we are sure that out options won't interact with postgres ones.

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