From:
Simon Riggs <simon(at)2ndQuadrant(dot)com>
To:
Aidan Van Dyk <aidan(at)highrise(dot)ca>
Cc:
Markus Wanner <markus(at)bluegap(dot)ch>,
Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>,
Kevin Grittner <Kevin(dot)Grittner(at)wicourts(dot)gov>,
MARK CALLAGHAN <mdcallag(at)gmail(dot)com>,
Alvaro Herrera <alvherre(at)commandprompt(dot)com>,
Andrew Dunstan <andrew(at)dunslane(dot)net>,
pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject:
Re: Re: [COMMITTERS] pgsql: Efficient
transaction-controlled synchronous replication.
Date:
2011-03-18 21:26:00
Message-ID:
1300483560.18619.19276.camel@ebony (view raw or flat )
Thread:
2011-03-06 22:51:56 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-06 23:09:53 from Andrew Dunstan <andrew(at)dunslane(dot)net>
2011-03-06 23:28:45 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-07 07:29:48 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2011-03-07 07:48:11 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-07 07:54:19 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2011-03-07 13:30:34 from Andrew Dunstan <andrew(at)dunslane(dot)net>
2011-03-07 14:02:44 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2011-03-07 14:21:46 from Andrew Dunstan <andrew(at)dunslane(dot)net>
2011-03-07 14:29:04 from Aidan Van Dyk <aidan(at)highrise(dot)ca>
2011-03-07 15:03:10 from Andrew Dunstan <andrew(at)dunslane(dot)net>
2011-03-07 15:13:46 from "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
2011-03-07 15:46:52 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2011-03-07 15:51:49 from Andrew Dunstan <andrew(at)dunslane(dot)net>
2011-03-07 16:09:49 from Alvaro Herrera <alvherre(at)commandprompt(dot)com>
2011-03-18 09:27:13 from Markus Wanner <markus(at)bluegap(dot)ch>
2011-03-18 13:16:24 from MARK CALLAGHAN <mdcallag(at)gmail(dot)com>
2011-03-18 13:30:56 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-18 13:40:22 from "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
2011-03-18 14:37:26 from Markus Wanner <markus(at)bluegap(dot)ch>
2011-03-18 16:03:03 from MARK CALLAGHAN <mdcallag(at)gmail(dot)com>
2011-03-18 14:19:18 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-18 14:52:13 from "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
2011-03-18 15:43:38 from Markus Wanner <markus(at)bluegap(dot)ch>
2011-03-18 15:47:46 from Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com>
2011-03-18 16:19:31 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-18 16:33:26 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-18 16:48:57 from "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
2011-03-18 17:35:46 from Greg Stark <gsstark(at)mit(dot)edu>
2011-03-18 19:17:56 from Markus Wanner <markus(at)bluegap(dot)ch>
2011-03-18 19:19:06 from Markus Wanner <markus(at)bluegap(dot)ch>
2011-03-18 19:29:50 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-18 19:34:18 from "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
2011-03-18 19:41:00 from Markus Wanner <markus(at)bluegap(dot)ch>
2011-03-18 21:08:11 from Aidan Van Dyk <aidan(at)highrise(dot)ca>
2011-03-18 21:26:00 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-22 20:33:11 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-23 07:27:13 from Markus Wanner <markus(at)bluegap(dot)ch>
2011-03-23 11:52:22 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-23 12:16:39 from Markus Wanner <markus(at)bluegap(dot)ch>
2011-03-23 15:24:30 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-18 21:18:28 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-18 21:24:03 from "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
2011-03-18 21:30:04 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-18 21:43:32 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-18 21:48:33 from "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
2011-03-18 22:47:54 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-19 19:27:04 from Markus Wanner <markus(at)bluegap(dot)ch>
2011-03-25 12:12:05 from Fujii Masao <masao(dot)fujii(at)gmail(dot)com>
2011-03-18 16:27:57 from "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
2011-03-18 19:22:20 from Markus Wanner <markus(at)bluegap(dot)ch>
2011-03-18 14:19:56 from Markus Wanner <markus(at)bluegap(dot)ch>
2011-03-18 15:52:18 from MARK CALLAGHAN <mdcallag(at)gmail(dot)com>
2011-03-07 15:04:22 from Aidan Van Dyk <aidan(at)highrise(dot)ca>
2011-03-07 14:55:11 from "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>
2011-03-06 23:28:13 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2011-03-06 23:36:30 from Jaime Casanova <jaime(at)2ndquadrant(dot)com>
2011-03-06 23:38:30 from Jaime Casanova <jaime(at)2ndquadrant(dot)com>
2011-03-06 23:52:49 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-07 08:27:16 from Fujii Masao <masao(dot)fujii(at)gmail(dot)com>
2011-03-07 08:44:52 from Fujii Masao <masao(dot)fujii(at)gmail(dot)com>
2011-03-07 09:30:35 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-07 09:47:20 from Fujii Masao <masao(dot)fujii(at)gmail(dot)com>
2011-03-10 21:28:12 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-18 14:10:03 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-19 01:12:52 from Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
2011-03-23 15:53:30 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-07 09:20:07 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-07 09:28:25 from Fujii Masao <masao(dot)fujii(at)gmail(dot)com>
2011-03-07 11:21:39 from Fujii Masao <masao(dot)fujii(at)gmail(dot)com>
2011-03-10 20:04:57 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-11 10:46:03 from Fujii Masao <masao(dot)fujii(at)gmail(dot)com>
2011-03-17 17:46:37 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-17 17:56:36 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-18 06:25:06 from Fujii Masao <masao(dot)fujii(at)gmail(dot)com>
2011-03-18 12:25:49 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-18 14:55:56 from Greg Stark <gsstark(at)mit(dot)edu>
2011-03-18 15:07:23 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-18 15:40:57 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-25 11:53:15 from Fujii Masao <masao(dot)fujii(at)gmail(dot)com>
2011-03-17 17:52:56 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-18 05:45:44 from Fujii Masao <masao(dot)fujii(at)gmail(dot)com>
2011-03-18 07:52:07 from Simon Riggs <simon(at)2ndQuadrant(dot)com>
2011-03-18 12:16:08 from Robert Haas <robertmhaas(at)gmail(dot)com>
2011-03-10 20:44:33 from Robert Haas <robertmhaas(at)gmail(dot)com>
Lists:
pgsql-committers pgsql-hackers
On Fri, 2011-03-18 at 17:08 -0400, Aidan Van Dyk wrote:
> On Fri, Mar 18, 2011 at 3:41 PM, Markus Wanner <markus(at)bluegap(dot)ch> wrote:
> > On 03/18/2011 08:29 PM, Simon Riggs wrote:
> >> We could do that easily enough, actually, if we wished.
> >>
> >> Do we wish?
> >
> > I personally don't see any problem letting a standby show a snapshot
> > before the master. I'd consider it unneeded network traffic. But then
> > again, I'm completely biased.
>
> In fact, we *need* to have standbys show a snapshot before the master.
>
> By the time the master acks the commit to the client, the snapshot
> must be visible to all client connected to both the master and the
> syncronous slave.
>
> Even with just a single server postgresql cluster, other
> clients(backends) can see the commit before the commiting client
> receives the ACK. Just that on a single server, the time period for
> that is small.
>
> Sync rep increases that time period by the length of time from when
> the slave reaches the commit point in the WAL stream to when it's ack
> of that point get's back to the wal sender. Ideally, that ACK time is
> small.
>
> Adding another round trip in there just for a "go almost to $COMIT,
> ok, now go to $COMMIT" type of WAL/ack is going to be pessimal for
> performance, and still not improve the *guarentees* it can make.
>
> It can only slightly reduce, but not eliminated that window where them
> master has WAL that the slave doesn't, and without a complete
> elimination (where you just switch the problem to be the slave has the
> data that the master doesn't), you haven't changed any of the
> guarantees sync rep can make (or not).
Well explained observation. Agreed.
--
Simon Riggs http://www.2ndQuadrant.com/books/
PostgreSQL Development, 24x7 Support, Training and Services
In response to
pgsql-hackers by date
Next :From: Simon RiggsDate: 2011-03-18 21:30:04Subject : Re: Re: [COMMITTERS] pgsql: Efficient
transaction-controlled synchronous replication.Previous :From : Kevin GrittnerDate : 2011-03-18 21:24:03Subject : Re: Re: [COMMITTERS] pgsql: Efficient transaction-controlled synchronous replication.
pgsql-committers by date
Next :From: Simon RiggsDate: 2011-03-18 21:30:04Subject : Re: Re: [COMMITTERS] pgsql: Efficient
transaction-controlled synchronous replication.Previous :From : Kevin GrittnerDate : 2011-03-18 21:24:03Subject : Re: Re: [COMMITTERS] pgsql: Efficient transaction-controlled synchronous replication.
