Date: Wed, 4 Sep 2019 08:21:27 +0700
From: Alexey Zagarin <zagarin@gmail.com>
To: Euler Taveira <euler@timbira.com.br>, Erik Rijkers <er@xs4all.nl>
Cc: Andres Freund <andres@anarazel.de>, Petr Jelinek
 <petr.jelinek@2ndquadrant.com>, pgsql-hackers
 <pgsql-hackers@postgresql.org>, Suzuki Hironobu <hironobu@interdb.jp>,
 a.kondratov@postgrespro.ru
Message-ID: <c6a2bd52-c5cb-4ca3-aace-ebfe5a330b91@Spark>
In-Reply-To: <9f25bd90d9f959f364efe79d92468682@xs4all.nl>
References: 
 <CAHE3wggb715X+mK_DitLXF25B=jE6xyNCH4YOwM860JR7HarGQ@mail.gmail.com>
 <CAHE3wgiHOdFHcEpm_w5f8vAXiugDmTciTiegNGVktezMbFkkrg@mail.gmail.com>
 <7106a0fc-8017-c0fe-a407-9466c9407ff8@2ndquadrant.com>
 <CAHE3wggWAtbT+Yt_MDy0uQwZcc6x820OwHp3_TBtAqtBeW4+Ug@mail.gmail.com>
 <20190203101439.26xgw5sv7vhgdcz2@alap3.anarazel.de>
 <CAHE3wgga+Z4PkzN4FgR5pO_y4Mn30mRcjfno5aXbnQE9YniEkw@mail.gmail.com>
 <d815a4a0f59e9c1b51caceae485b1193@xs4all.nl>
 <CAHE3wghn9hu10-YSr7p4TyRUurVAif3O+NhX7MsD=XEvPROSxQ@mail.gmail.com>
 <94b31725538135038f2fccaf08bf3811@xs4all.nl>
 <5720d79c-3d09-4ac2-b233-355220471838@Spark>
 <CAHE3wgjEQ9O5N8SvWJsVJU-zGTwd0VFqnijTabWpB9Un9-s0_w@mail.gmail.com>
 <9f25bd90d9f959f364efe79d92468682@xs4all.nl>
Subject: Re: row filtering for logical replication
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="5d6f119c_cc1016f_180"
Precedence: bulk

--5d6f119c_cc1016f_180
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

> OK, I'll consider Alexey Kondratov's set of patches as the current
> state-of-the-art then. (They still apply.)

Alexey's patch is the rebased version of=C2=A0previous=C2=A0Euler's patch=
 set, with slot type mistake fixed, and adapted to current changes in the=
 master branch. It also has testing improvements. On the other hand, the =
new patches from Euler include more fixes and the implementation of filte=
ring in COPY (as far as I can tell from code) which addresses my particul=
ar pain point with BDR. Hope they'll be joined soon. :)

> It turns out this did not work until I added a wait state after the
> CREATE SUBSCRIPTION. It always fails without the wait state, and always=

> works with the wait state.
>
> Do you agree this is a bug=3F

I'm not sure this is a bug as after the subscription is added (or a new t=
able added to the publication and then the subscription is refreshed), th=
e whole table is synchronized using COPY statement. Depending on size of =
the table it can take some time. You may want to check srsubstate in pg=5F=
subscription=5Frel instead of just sleep for more reliable implementation=
.

Alexey

--5d6f119c_cc1016f_180
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

<html xmlns=3D=22http://www.w3.org/1999/xhtml=22>
<head>
<title></title>
</head>
<body>
<div name=3D=22messageBodySection=22>
<div dir=3D=22auto=22>
<blockquote type=3D=22cite=22 class=3D=22spark=5Fquote=22 style=3D=22care=
t-color: rgb(39, 39, 40); margin: 5px; padding-left: 10px; border-left-wi=
dth: thin; border-left-style: solid; border-left-color: rgb(26, 188, 156)=
;=22>OK, I'll consider Alexey Kondratov's set of patches as the current<b=
r />
state-of-the-art then. (They still apply.)<br /></blockquote>
</div>
<div dir=3D=22auto=22><br /></div>
<div dir=3D=22auto=22>Alexey's patch is the rebased version of<span style=
=3D=22caret-color: rgb(39, 39, 40);=22>&=23160;</span><span style=3D=22ca=
ret-color: rgb(39, 39, 40);=22>previous</span>&=23160;Euler's patch set, =
with slot type mistake fixed, and adapted to current changes in the maste=
r branch. It also has testing improvements. On the other hand, the new pa=
tches from Euler include more fixes and the implementation of filtering i=
n COPY (as far as I can tell from code) which addresses my particular pai=
n point with BDR. Hope they'll be joined soon. :)</div>
<div dir=3D=22auto=22><br /></div>
<div dir=3D=22auto=22>
<blockquote type=3D=22cite=22 class=3D=22spark=5Fquote=22 style=3D=22care=
t-color: rgb(39, 39, 40); margin: 5px; padding-left: 10px; border-left-wi=
dth: thin; border-left-style: solid; border-left-color: rgb(26, 188, 156)=
;=22>It turns out this did not work until I added a wait state after the<=
br />
CREATE SUBSCRIPTION. It always fails without the wait state, and always<b=
r />
works with the wait state.<br />
<br />
Do you agree this is a bug=3F<br /></blockquote>
</div>
<div dir=3D=22auto=22><br /></div>
<div dir=3D=22auto=22>I'm not sure this is a bug as after the subscriptio=
n is added (or a new table added to the publication and then the subscrip=
tion is refreshed), the whole table is synchronized using COPY statement.=
 Depending on size of the table it can take some time. You may want to ch=
eck srsubstate in pg=5Fsubscription=5Frel instead of just sleep for more =
reliable implementation.</div>
</div>
<div name=3D=22messageSignatureSection=22><br />
<div dir=3D=22auto=22>Alexey</div>
</div>
</body>
</html>

--5d6f119c_cc1016f_180--