Koichi Suzuki - 2015-11-25

Thanks for the report. Main Postgres-XC development site is now moved to
https://github.com/postgres-x2/postgres-x2. It's a great help if you
submit the issue to its issue list.

Here's an answer to your question:

  1. Master recovery is not provided automatically as a part of XC. You
    need to configure slave of each datanode.
  2. While a master node is down, any transaction involves such failed node
    will fail. Other transaction will not be affected.

So in the case of coordinator, DDL will fail. In the case of datanode,
updating replicated table will fail. Updating distributed table with the
failed node involved will fail too.

Regards;

Koichi Suzuki
https://www.linkedin.com/in/koichidbms

2015-11-25 19:12 GMT+09:00 Web Gourou webgourou2000@users.sf.net:

[bugs:#494] what happen if one master fail

Status: open
Group: 1.0 Dev Q
Created: Wed Nov 25, 2015 10:12 AM UTC by Web Gourou
Last Updated: Wed Nov 25, 2015 10:12 AM UTC
Owner: nobody

Hi,

it is not a ticket but a question, I was not able to know what happen if
one master fail. Does all data will be resynchronized when it will become
available again. Or should we reconfigure it manually ?

Thanks

Aurélien

Sent from sourceforge.net because you indicated interest in <
https://sourceforge.net/p/postgres-xc/bugs/494/>

To unsubscribe from further messages, please visit <
https://sourceforge.net/auth/subscriptions/>

 

Related

Bugs: #494