postgres-xc-developers Mailing List for Postgres-XC (Page 3)

sorry if this message is a dup, previous one sent to postgres-xc-developers
was bounded back.


hi, Koichi,

thanks for your response. I am trying to connect to a hdfs system, and
reading hdfs files through FDW. Assuming I have 10 files mapping to one
foreign table, and have 3 datanode on PG-XC, I'd like coordinator node to
do some simple scheduling so that each dataNode of PG-XC will read roughtly
same amount of data. That is why I'd like to hook a value(a String will be
fine) that generated by coordinator node, and visible to every data node.
The string will be the same value for all datanodes.

Demai

On Tue, Nov 4, 2014 at 1:34 AM, Koichi Suzuki <koi...@gm...> wrote:

> Demai;
>
> Anyway, could you describe your point more specifically?   What would
> you do this in vanilla postgresql?  This will help to find how to do
> it in XC.
>
> Regards;
> ---
> Koichi Suzuki
>
>
> 2014-11-04 10:48 GMT+09:00 Amit Langote <ami...@gm...>:
> > Hi Demai,
> >
> > On Tue, Nov 4, 2014 at 10:40 AM, Demai Ni <ni...@gm...> wrote:
> >> hi, hackers,
> >>
> >> I am looking for a simple way to pass a value(ideally a hashtable, and
> use a
> >> string for easy implementation for now), that is generated by
> coordinator
> >> node, and pass it to all data nodes. Plans may be one approach, I am
> just
> >> wondering whether there is a simpler hook?
> >>
> >
> > Correct list for this question would be (added to 'cc'):
> >
> > pos...@li...
> >
> > Thanks,
> > Amit
> >
> >
> ------------------------------------------------------------------------------
> > _______________________________________________
> > Postgres-xc-developers mailing list
> > Pos...@li...
> > https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers
>

hi, Koichi,

thanks for your response. I am trying to connect to a hdfs system, and
reading hdfs files through FDW. Assuming I have 10 files mapping to one
foreign table, and have 3 datanode on PG-XC, I'd like coordinator node to
do some simple scheduling so that each dataNode of PG-XC will read roughtly
same amount of data. That is why I'd like to hook a value(a String will be
fine) that generated by coordinator node, and visible to every data node.

Demai

>
> Demai;
>
> Anyway, could you describe your point more specifically?   What would
> you do this in vanilla postgresql?  This will help to find how to do
> it in XC.
>
> Regards;
> ---
> Koichi Suzuki


>> hi, hackers,
>>
>> I am looking for a simple way to pass a value(ideally a hashtable, and
use a
>> string for easy implementation for now), that is generated by coordinator
>> node, and pass it to all data nodes. Plans may be one approach, I am just
>> wondering whether there is a simpler hook?
>>

Demai;

Anyway, could you describe your point more specifically?   What would
you do this in vanilla postgresql?  This will help to find how to do
it in XC.

Regards;
---
Koichi Suzuki


2014-11-04 10:48 GMT+09:00 Amit Langote <ami...@gm...>:
> Hi Demai,
>
> On Tue, Nov 4, 2014 at 10:40 AM, Demai Ni <ni...@gm...> wrote:
>> hi, hackers,
>>
>> I am looking for a simple way to pass a value(ideally a hashtable, and use a
>> string for easy implementation for now), that is generated by coordinator
>> node, and pass it to all data nodes. Plans may be one approach, I am just
>> wondering whether there is a simpler hook?
>>
>
> Correct list for this question would be (added to 'cc'):
>
> pos...@li...
>
> Thanks,
> Amit
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Postgres-xc-developers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers

Hi Demai,

On Tue, Nov 4, 2014 at 10:40 AM, Demai Ni <ni...@gm...> wrote:
> hi, hackers,
>
> I am looking for a simple way to pass a value(ideally a hashtable, and use a
> string for easy implementation for now), that is generated by coordinator
> node, and pass it to all data nodes. Plans may be one approach, I am just
> wondering whether there is a simpler hook?
>

Correct list for this question would be (added to 'cc'):

pos...@li...

Thanks,
Amit

On Oct 2, 2014 12:04 PM, <tic...@ir...> wrote:
>
> ________________________________
>
> This is a system generated mail. Please do not reply to this email ID.
Ifwas  you have a query or need any clarification you may:
> (1) Call our 24-hour Customer Care or
> (2) Email Us ca...@ir...
> ________________________________
>
> Ticket Confirmation
>
> Dear Customer,
>
> Congratulations! Thank you for using IRCTC's online rail reservation
facility.w22¢£™|`` Your e-ticket h`as been booked and the details are
indicated below.
>
> Please take a printout of the Electronic Reservation Slip (ERS)/a
screenshot of ERS i.e. Virtual Reservation Message(VRM) OF YOUR TICKET FROM
YOUR Booked Tickets History page .You have to carry this printout/VRM along
with any one from the listed identity cards (Voter Identity Card / Passport
/ PAN Card / Driving License / Photo ID card issued by Central / State Govt
/ Student Identity Card with photograph issued by recognized School or
College for their students / Nationalised Bank Passbook with photograph /
Credit Cards issued by Banks with laminated photograph) during train
journey in original. Both these will be examined by ticket checking staff
on stations/trains for verification purpose.
>
>
> Ticket Details
> ________________________________
> Transaction ID :
> 100000090432490
> PNR No. :
> 8728342198
> Train No. & Name :
> 01651 / PUNE HBG SPL
> Date of Booking :
> 02-Oct-2014
> Class :
> SLEEPER CLASS
> Quota :
> GENERAL
> Date of Journey :
> 07-Oct-2014
> From :
> PUNE
> To :
> HBJ
> Boarding At :
> PUNE
> Date Of Boarding :
> 07-Oct-2014
> Reservation Upto :
> HBJ
>
> Passenger Details
> ________________________________
> Sl. No.
> Name
> Age
> Gender
> Status
> Coach
> Seat / Berth / WL No
> 1
> VIVEK RAGHUWANSH
> 34
> Male
> RAC
>
> 99
>
> Fare Details
> ________________________________
> Distance :
> 889 KM
> Ticket Fare :
> Rs. 460
> Service Charge :
> Rs. 11.24
> Total Fare :
> Rs. 471.24 *
> * Payment Gateway charges as applicable.
>
> For cancellation of your ticket, please log in to your account, select
the PNR you wish to cancel and cancel your ticket online. Cancellation of
your e-Ticket is possible only on the website www.irctc.co.in, and is NOT
possible on PRS Counters.
>
> You can cancel any of the passenger names from the list of passengers on
this ticket without affecting the ticket as a whole.
>
> CANCELLATION OF E-TICKETS
>
> (1) E-Ticket can be cancelled online till preparation of charts (which is
normally 4 to 6 hours before the scheduled departure of the train from the
originating stations. (For trains starting up to 12 noon the chart
preparation is usually done on the previous night)). If by any reason you
are unable to cancel your ticket on the website till the time of charting
please mail to eti...@ir.... User can check their refund status
online by clicking "Refund Status" link after logging in. All refund will
be processed as per extent Railway Refund Rules. Please refer refund rules
here.
>
> (2) After preparation of charts, the user shall have to file online TDR
for claiming refund. IRCTC forward the case to concerned Zonal Railway on
the basis of Customer request through EDR and Enhance coaching refund
system. The refund case will be decided by Railways. The TDR can be filed
by the same user who books the ticket. After receipt of the refund amount
from Railway, it shall be credited only in the same account through which
booking was made through opted payment gateway. If by any reason you are
unable to file TDR through website please mail to eti...@ir....
User can check TDR/Refund status through TDR History. All refund will be
process as per extent Railway Refund Rules. Please refer refund rules here.
>
> (3) Any refund amount will be credited only in the same account through
booking was made by user. Customer is advised not to close that account so
that refund amount will be credited back.
>
> For CHANGE OF BOARDING STATION & CHANGE IN NAME customer can now approach
the nearest Railway Reservation Office with Electronic Reservation Slip
print out and photo identity proof mentioned in the Electronic Reservation
Slip at least 24 hours before the scheduled departure of train as per
existing Railway rules.
>
> Once you have cancelled your ticket, the fresh details and changes made
by you will be available on the Cancelled e-Tickets History (as a link to
site). The refund amount due to you after cancellation will be credited
back to the same account which you had used for booking the ticket.
>
> Passengers are advised not to carry inflammable/dangerous/explosive
articles as part of their luggage and also to desist from smoking in the
trains.
>
> For any further assistance, please contact us at 24*7 Hrs.
> Customer Support at 011-39340000 or mail us at ca...@ir....
>
> Just dial 139 from your landline, mobile & CDMA phones for railway
enquiries.
>
> ******************************** Information
*************************************
>
> For any enquiries or information regarding your transaction with IRCTC,
do not provide your credit/debit card details by any means to IRCTC. All
your queries can be replied on the basis of 15 digit IRCTC Transaction id/
10 digit PNR no./ User id. IRCTC does not store the credit/ debit card
information in any form during the transaction.
>
>
**********************************************************************************
>
>
>
> Warm Regards,
> Customer Care
> Internet Ticketing
> IRCTC
>
>

Thanks a lot for the update.   This will be committed in the earliest occasion.

Best;
---
Koichi Suzuki

2014/09/16 11:35、Wang Diancheng <dia...@gm...> のメール：

> Hi,
> 
> it lead to following statement crash:
> 
> clean connection to all force for regression;
> 
> patch attached.
> 
> diff --git a/src/backend/pgxc/pool/poolcomm.c b/src/backend/pgxc/pool/poolcomm.c
> index 02d67f3..b775649 100644
> --- a/src/backend/pgxc/pool/poolcomm.c
> +++ b/src/backend/pgxc/pool/poolcomm.c
> @@ -778,7 +778,7 @@ pool_recvpids(PoolPort *port, int **pids)
> 	{
> 		int n;
> 		memcpy(&n, buf + 5 + i * sizeof(int), sizeof(int));
> -		*pids[i] = ntohl(n);
> +		(*pids)[i] = ntohl(n);
> 	}
> 	return n32;
> 
> ------------------------------------------------------------------------------
> Want excitement?
> Manually upgrade your production database.
> When you want reliability, choose Perforce.
> Perforce version control. Predictably reliable.
> http://pubads.g.doubleclick.net/gampad/clk?id=157508191&iu=/4140/ostg.clktrk_______________________________________________
> Postgres-xc-developers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers

Hi,

it lead to following statement crash:

clean connection to all force for regression;

patch attached.

On Tue, Sep 9, 2014 at 4:30 PM, Luyuanfei <ka...@hu...> wrote:
> /*
> * cost_remotequery
> * As of now the function just sets the costs to 0 to make this path the
> * cheapest.
> * PGXC_TODO: Ideally, we should estimate the costs of network transfer from
> * datanodes and any datanode costs involved.
> */
> It's very useful to some statement such as join or subquery.
In short, no. And I am not aware of anybody working in this area for
this project. Any patch would need a paper or research made in this
area to start on good basis IMO. Even without researches in this area,
I imagine that we'd finish for sure with a set of cost parameters that
can be set per-node (you don't want to have the same network cost for
a node on local network and another one thousands of miles away) and
that the planner could use those parameters when evaluating a plan
using an equal qual on an index used for hash. Another problem is that
we need a good idea of how they compare to other cost parameters to
avoid unbalanced plans. The issue remaining is then what are those
parameters, and what they symbolize.

> When you want reliability, choose Perforce.
Not much sure about that. Git is more appealing with its decentralized system.
-- 
Michael

There is a comment about function cost_remotequery as following:

/*
* cost_remotequery
* As of now the function just sets the costs to 0 to make this path the
* cheapest.
* PGXC_TODO: Ideally, we should estimate the costs of network transfer from
* datanodes and any datanode costs involved.
*/

It's very useful to some statement such as join or subquery.

Yours sincerely

Kaka Lu

Hi,

Unfortunatelly, I haven't heard any progress on SAVEPOINT.
The feature is on the wish list, but AFAIC it seems not to have high priority.
 * https://sourceforge.net/p/postgres-xc/xc-wiki/Roadmap/

This page mentions about isolation levels.
 * http://postgres-xc.sourceforge.net/docs/1_2_1/sql-start-transaction.html
It says XC supports READ COMMITTED and READ UNCOMMITTED.

Regards.

On 21 August 2014 23:48, Mikko Rantalainen <mik...@pe...> wrote:
> Hi,
>
> Our software is using PostgreSQL as backend and we're considering
> different choices for high availability. I found Postgres-XC and noticed
> that SAVEPOINTs are not supported yet. Other than that, it seems to be a
> good match for our needs.
>
> Previous discussion about this issue seems to have happened around a
> year ago (http://sourceforge.net/p/postgres-xc/mailman/message/31349199/).
>
> What's the current status? Will SAVEPOINTs be supported in near future
> (e.g. within a year)?
>
> Also related question is where's the documentation about actually
> supported transaction isolation levels?
>
> --
> B.R.
> Mikko Rantalainen
>
> ------------------------------------------------------------------------------
> Slashdot TV.
> Video for Nerds.  Stuff that matters.
> http://tv.slashdot.org/
> _______________________________________________
> Postgres-xc-developers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers

Hi,

Our software is using PostgreSQL as backend and we're considering 
different choices for high availability. I found Postgres-XC and noticed 
that SAVEPOINTs are not supported yet. Other than that, it seems to be a 
good match for our needs.

Previous discussion about this issue seems to have happened around a 
year ago (http://sourceforge.net/p/postgres-xc/mailman/message/31349199/).

What's the current status? Will SAVEPOINTs be supported in near future 
(e.g. within a year)?

Also related question is where's the documentation about actually 
supported transaction isolation levels?

-- 
B.R.
Mikko Rantalainen

Hello;

I removed PG packers ML from CC: list as advised.

BTW, replication by Trove must be for redundancy, which is different from XC’s replication.   I think you can still use Trove’s replication even for XC.   We may have to study Trove’s replication a bit to see if it supports multi-master configuration of XC coordinators.

Regards;
---
Koichi Suzuki

2014/08/20 15:11、Mark Kirkwood <mar...@ca...> のメール：

> On 19/08/14 23:14, Vivek Singh Raghuwanshi wrote:
>> 
>> Hi All,
>> 
>> Please let me know is that possible to use Openstack Trove with Postgres-XC.
>> With instances and Baremetal (after Juno Release).
>> I Know it is possible to use other medium like MySQL or PostgreSQL, but
>> i am not sure about XC.
> 
> AFAIK [1], vanilla Postgres support for Trove is still at the review 
> stage, and there is quite a bit to finish, and some stuff to fix.
> 
> I'd think that getting Trove to provision a complete XC VM (or more 
> sensibly a set of VMs that run XC) is going to require fair bit of extra 
> work.
> 
> It looks like replication will be deployable via Trove for the supported 
> flavours (Mysql, Mongo, Cassandra, Counch, Redis) see 
> https://wiki.openstack.org/wiki/Trove/Replication-And-Clustering . I'm 
> not seeing any specific discussion of sharding notice.
> 
> Cheers
> 
> Mark
> 
> [1] I've been helping debug the Postgres Trove patch
> 
> 
> ------------------------------------------------------------------------------
> Slashdot TV.  
> Video for Nerds.  Stuff that matters.
> http://tv.slashdot.org/
> _______________________________________________
> Postgres-xc-developers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers
> 

On 19/08/14 23:14, Vivek Singh Raghuwanshi wrote:
>
> Hi All,
>
> Please let me know is that possible to use Openstack Trove with Postgres-XC.
> With instances and Baremetal (after Juno Release).
> I Know it is possible to use other medium like MySQL or PostgreSQL, but
> i am not sure about XC.

AFAIK [1], vanilla Postgres support for Trove is still at the review 
stage, and there is quite a bit to finish, and some stuff to fix.

I'd think that getting Trove to provision a complete XC VM (or more 
sensibly a set of VMs that run XC) is going to require fair bit of extra 
work.

It looks like replication will be deployable via Trove for the supported 
flavours (Mysql, Mongo, Cassandra, Counch, Redis) see 
https://wiki.openstack.org/wiki/Trove/Replication-And-Clustering . I'm 
not seeing any specific discussion of sharding notice.

Cheers

Mark

[1] I've been helping debug the Postgres Trove patch

On 08/19/2014 04:14 AM, Vivek Singh Raghuwanshi wrote:
> 
> Hi All,
> 
> Please let me know is that possible to use Openstack Trove with Postgres-XC.
> With instances and Baremetal (after Juno Release).
> I Know it is possible to use other medium like MySQL or PostgreSQL, but
> i am not sure about XC.

Please do not cross-post this to pgsql-hackers, thanks.

-- 
Josh Berkus
PostgreSQL Experts Inc.
http://pgexperts.com

On 08/19/2014 07:46 PM, Vivek Singh Raghuwanshi wrote:
> Thanks,
> One more question, is this library support multitenancy or we need to
> launch separate VPC (virtual public cloud) every time for each customer.
> its good if we have both option.

Please don't cross-post this to pgsql-hackers; remove that CC from
future replies.

-- 
 Craig Ringer                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services

Thanks,
One more question, is this library support multitenancy or we need to
launch separate VPC (virtual public cloud) every time for each customer.
its good if we have both option.



On Tue, Aug 19, 2014 at 5:04 PM, 鈴木 幸市 <ko...@in...> wrote:

>  XC’s libpq is binary compatible with PG.   So as long as Openstack uses
> libpq to connect to PG, XC should work too.
>
>  Appreciate if you have a chance to try.
>
>  Thanks.
> ---
> Koichi Suzuki
>
>  2014/08/19 20:14、Vivek Singh Raghuwanshi <viv...@gm...>
> のメール：
>
>
>  Hi All,
>
>  Please let me know is that possible to use Openstack Trove with
> Postgres-XC.
>  With instances and Baremetal (after Juno Release).
>  I Know it is possible to use other medium like MySQL or PostgreSQL, but
> i am not sure about XC.
>
>
>  --
> ViVek Raghuwanshi
> Mobile -+91-09595950504
> Skype - vivek_raghuwanshi
> IRC vivekraghuwanshi
> http://vivekraghuwanshi.wordpress.com/
> http://in.linkedin.com/in/vivekraghuwanshi
>
> ------------------------------------------------------------------------------
> _______________________________________________
> Postgres-xc-developers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers
>
>
>


-- 
ViVek Raghuwanshi
Mobile -+91-09595950504
Skype - vivek_raghuwanshi
IRC vivekraghuwanshi
http://vivekraghuwanshi.wordpress.com/
http://in.linkedin.com/in/vivekraghuwanshi

XC’s libpq is binary compatible with PG.   So as long as Openstack uses libpq to connect to PG, XC should work too.

Appreciate if you have a chance to try.

Thanks.
---
Koichi Suzuki

2014/08/19 20:14、Vivek Singh Raghuwanshi <viv...@gm...<mailto:viv...@gm...>> のメール：


Hi All,

Please let me know is that possible to use Openstack Trove with Postgres-XC.
With instances and Baremetal (after Juno Release).
I Know it is possible to use other medium like MySQL or PostgreSQL, but i am not sure about XC.


--
ViVek Raghuwanshi
Mobile -+91-09595950504
Skype - vivek_raghuwanshi
IRC vivekraghuwanshi
http://vivekraghuwanshi.wordpress.com/
http://in.linkedin.com/in/vivekraghuwanshi
------------------------------------------------------------------------------
_______________________________________________
Postgres-xc-developers mailing list
Pos...@li...<mailto:Pos...@li...>
https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers

Hi All,

Please let me know is that possible to use Openstack Trove with Postgres-XC.
With instances and Baremetal (after Juno Release).
I Know it is possible to use other medium like MySQL or PostgreSQL, but i
am not sure about XC.


-- 
ViVek Raghuwanshi
Mobile -+91-09595950504
Skype - vivek_raghuwanshi
IRC vivekraghuwanshi
http://vivekraghuwanshi.wordpress.com/
http://in.linkedin.com/in/vivekraghuwanshi

I'm sorry. I was wrong.

list_append_unique is also replaced to list_append_unique_ptr in your patch.
It looks fine.

Regards.

On 5 August 2014 16:15, Masataka Saito <pg...@gm...> wrote:
> Hi,
>
> I think this patch is good, but there is still some bad API calls,
> e.g. list_append_unique.
>
> Regards.
>
> On 31 July 2014 10:29, Wang Diancheng <dia...@gm...> wrote:
>> Hi,
>>
>> because struct pgxc_node_handle is not a kind of Node, we could not use
>> function list_member, list_delete.
>>
>> Patch attached.
>>
>>
>> ------------------------------------------------------------------------------
>> Infragistics Professional
>> Build stunning WinForms apps today!
>> Reboot your WinForms applications with our WinForms controls.
>> Build a bridge from your legacy apps to the future.
>> http://pubads.g.doubleclick.net/gampad/clk?id=153845071&iu=/4140/ostg.clktrk
>> _______________________________________________
>> Postgres-xc-developers mailing list
>> Pos...@li...
>> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers
>>

Hi,

I think this patch is good, but there is still some bad API calls,
e.g. list_append_unique.

Regards.

On 31 July 2014 10:29, Wang Diancheng <dia...@gm...> wrote:
> Hi,
>
> because struct pgxc_node_handle is not a kind of Node, we could not use
> function list_member, list_delete.
>
> Patch attached.
>
>
> ------------------------------------------------------------------------------
> Infragistics Professional
> Build stunning WinForms apps today!
> Reboot your WinForms applications with our WinForms controls.
> Build a bridge from your legacy apps to the future.
> http://pubads.g.doubleclick.net/gampad/clk?id=153845071&iu=/4140/ostg.clktrk
> _______________________________________________
> Postgres-xc-developers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers
>

Hi,

because struct pgxc_node_handle is not a kind of Node, we could not use
function list_member, list_delete.

Patch attached.

On Thu, Jul 24, 2014 at 9:38 PM, 鈴木 幸市 <ko...@in...> wrote:

>  Background is to make configuration as simple as possible and to make
> promotion simpler.    XC primitive nodes support different port for master
> and slave, as in vanilla PG.
>
>  When promoting slave with the same port number, VIP will help to connect
> to the node without change in the target port/host.   VIP cannot do this if
> port number is different.   I understand this helps only for coordinator.
> I think it’s also better to make coordinator/datanode configuration as same
> as possible.
>

I had assumed this to be the reason. It makes sense to do as the default. I
think it would be a nice option to be able to configure differently for
development and debugging.

I added this to Postgres-XL as a ticket with a low priority, since I would
like to see it at least get in there eventually, to optionally override.

https://sourceforge.net/p/postgres-xl/tickets/25/



>
>  Regards;
> ---
> Koichi Suzuki
>
>  2014/07/24 21:08、Pavan Deolasee <pav...@gm...> のメール：
>
>  I noticed that we can't specify port number for datanode slave nodes
> that are different from the master nodes. Is it that way for a good reason?
> I'm asking before if I need to test this feature on a single node, I would
> like to run master and slave on the same machine. But since they use the
> same port number, its not possible to do that. Should we consider
> supporting different ports for datanode (and coordinator slaves) just like
> we do for GTM slave?
>
>  Thanks,
> Pavan
>
>  --
> Pavan Deolasee
> http://www.linkedin.com/in/pavandeolasee
>
> ------------------------------------------------------------------------------
> Want fast and easy access to all the code in your enterprise? Index and
> search up to 200,000 lines of code with a free copy of Black Duck
> Code Sight - the same software that powers the world's largest code
> search on Ohloh, the Black Duck Open Hub! Try it now.
> http://p.sf.net/sfu/bds_______________________________________________
> Postgres-xc-developers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers
>
>
>
>
> ------------------------------------------------------------------------------
> Want fast and easy access to all the code in your enterprise? Index and
> search up to 200,000 lines of code with a free copy of Black Duck
> Code Sight - the same software that powers the world's largest code
> search on Ohloh, the Black Duck Open Hub! Try it now.
> http://p.sf.net/sfu/bds
> _______________________________________________
> Postgres-xc-developers mailing list
> Pos...@li...
> https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers
>
>


-- 
Mason Sharp

TransLattice - http://www.translattice.com
Distributed and Clustered Database Solutions

Background is to make configuration as simple as possible and to make promotion simpler.    XC primitive nodes support different port for master and slave, as in vanilla PG.

When promoting slave with the same port number, VIP will help to connect to the node without change in the target port/host.   VIP cannot do this if port number is different.   I understand this helps only for coordinator.   I think it’s also better to make coordinator/datanode configuration as same as possible.

Regards;
---
Koichi Suzuki

2014/07/24 21:08、Pavan Deolasee <pav...@gm...<mailto:pav...@gm...>> のメール：

I noticed that we can't specify port number for datanode slave nodes that are different from the master nodes. Is it that way for a good reason? I'm asking before if I need to test this feature on a single node, I would like to run master and slave on the same machine. But since they use the same port number, its not possible to do that. Should we consider supporting different ports for datanode (and coordinator slaves) just like we do for GTM slave?

Thanks,
Pavan

--
Pavan Deolasee
http://www.linkedin.com/in/pavandeolasee
------------------------------------------------------------------------------
Want fast and easy access to all the code in your enterprise? Index and
search up to 200,000 lines of code with a free copy of Black Duck
Code Sight - the same software that powers the world's largest code
search on Ohloh, the Black Duck Open Hub! Try it now.
http://p.sf.net/sfu/bds_______________________________________________
Postgres-xc-developers mailing list
Pos...@li...
https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers

Thanks Pavan;

The patch has been committed to all the relevant branches.
---
Koichi Suzuki

2014/07/24 20:15、Pavan Deolasee <pav...@gm...<mailto:pav...@gm...>> のメール：

On Thu, Jul 24, 2014 at 11:25 AM, Koichi Suzuki <koi...@gm...<mailto:koi...@gm...>> wrote:

Thanks a lot for pointing out.   Could you submit a patch for this?

Sure. Attached.

Thanks,
Pavan

--
Pavan Deolasee
http://www.linkedin.com/in/pavandeolasee
<xc_use_pclose.patch>------------------------------------------------------------------------------
Want fast and easy access to all the code in your enterprise? Index and
search up to 200,000 lines of code with a free copy of Black Duck
Code Sight - the same software that powers the world's largest code
search on Ohloh, the Black Duck Open Hub! Try it now.
http://p.sf.net/sfu/bds_______________________________________________
Postgres-xc-developers mailing list
Pos...@li...
https://lists.sourceforge.net/lists/listinfo/postgres-xc-developers

I noticed that we can't specify port number for datanode slave nodes that
are different from the master nodes. Is it that way for a good reason? I'm
asking before if I need to test this feature on a single node, I would like
to run master and slave on the same machine. But since they use the same
port number, its not possible to do that. Should we consider supporting
different ports for datanode (and coordinator slaves) just like we do for
GTM slave?

Thanks,
Pavan

-- 
Pavan Deolasee
http://www.linkedin.com/in/pavandeolasee