pg.ddx.io  pgsql-hackers@postgresql.org mailing list archive  

help / color / mirror / Atom feed
From: Markus Wanner <markus@bluegap.ch>
To: Kevin Grittner <Kevin.Grittner@wicourts.gov>
Cc: Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>
Cc: Simon Riggs <simon@2ndQuadrant.com>
Cc: Alvaro Herrera <alvherre@commandprompt.com>
Cc: Andrew Dunstan <andrew@dunslane.net>
Cc: MARK CALLAGHAN <mdcallag@gmail.com>
Cc: Aidan Van Dyk <aidan@highrise.ca>
Cc: pgsql-hackers <pgsql-hackers@postgresql.org>
Subject: Re: Re: [COMMITTERS] pgsql: Efficient	 transaction-controlled synchronous replication.
Date: Fri, 18 Mar 2011 20:22:20 +0100
Message-ID: <4D83B0EC.1010105@bluegap.ch> (raw)
In-Reply-To: <4D8341BD020000250003BAB2@gw.wicourts.gov>
References: <E1PwMns-00047O-UU@gemulon.postgresql.org>
	<4D741441.2090704@dunslane.net>
	<1299454125.1696.6138.camel@ebony>
	<4D74896C.5030402@enterprisedb.com>
	<4D74DDFA.70800@dunslane.net>
	<4D74E584.7080409@enterprisedb.com>
	<4D74E9FA.7070202@dunslane.net>
	<AANLkTinkzJ=UrLbThZy2KeQg+SjwYqj2r0mDomMwirB7@mail.gmail.com>
	<4D74F3AE.9000802@dunslane.net>
	<4D74FDEC.5040406@enterprisedb.com>
	<4D74FF15.1070804@dunslane.net>
	<1299513979-sup-1987@alvh.no-ip.org>
	<4D832571.5030808@bluegap.ch>
	<AANLkTi=B+cWE-pQJQM=zoA9=Zr25W7zPPMFvvBe11O4E@mail.gmail.com>
	<1300457958.18619.13780.camel@ebony>
	<4D832B4D020000250003BA98@gw.wicourts.gov>
	<4D837EA2.2070205@enterprisedb.com>
	<4D8341BD020000250003BAB2@gw.wicourts.gov>

On 03/18/2011 05:27 PM, Kevin Grittner wrote:
> Basically, what Heikki addresses.  It has to be committed after
> crash and recovery, and deal with replicas which may or may not have
> been notified and may or may not have applied the transaction.

Huh?  I'm not quite following here.  Committing additional transactions
isn't a problem, reverting committed transactions is.

And yes, given that we only wait for ACK from a single standby, you'd
have to failover to exactly *that* standby to guarantee consistency.

> In fact, on further reflection, allowing other transactions to see
> work before the committing transaction returns could lead to broken
> behavior if that viewing transaction took some action based on the
> that, the master crashed, recovery was done using a standby, and
> that standby hadn't persisted the transaction.  So this behavior is
> necessary for good behavior.

I fully agree to that.

Regards

Markus



view thread (85+ messages)  latest in thread

Message-ID: <4D83B0EC.1010105@bluegap.ch>
Permalink:  ../4D83B0EC.1010105@bluegap.ch/
Also on:    postgresql.org/message-id/4D83B0EC.1010105@bluegap.ch

 · 

reply
Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: pgsql-hackers@postgresql.org
  Cc: markus@bluegap.ch, Kevin.Grittner@wicourts.gov, heikki.linnakangas@enterprisedb.com, simon@2ndQuadrant.com, alvherre@commandprompt.com, andrew@dunslane.net, mdcallag@gmail.com, aidan@highrise.ca
  Subject: Re: Re: [COMMITTERS] pgsql: Efficient	 transaction-controlled synchronous replication.
  In-Reply-To: <4D83B0EC.1010105@bluegap.ch>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox