Thread: Documentaion fix.
Hello. I found that an example output in the following page is not correct for the version. https://www.postgresql.org/docs/10/static/warm-standby.html postgres=# SELECT * FROM pg_replication_slots; > slot_name | slot_type | datoid | database | active | xmin | restart_lsn | confirmed_flush_lsn > -------------+-----------+--------+----------+--------+------+-------------+--------------------- > node_a_slot | physical | | | f | | | The correct one is > slot_name | plugin | slot_type | datoid | database | temporary | active | active_pid | xmin | catalog_xmin | restart_lsn| confirmed_flush_lsn > -------------+--------+-----------+--------+----------+-----------+--------+------------+------+--------------+-------------+--------------------- > node_a_slot | | physical | | | f | f | | | | | From 9.4, 9.5, 9.6 and 10 and later have an example output wrong in individual way of the version. I'm not sure how the documentation was updated but each of the attached patches fixes the version designated in the file name. But, just fixing it makes the line seemingly a bit too long.. regards. -- Kyotaro Horiguchi NTT Open Source Software Center From 93c69b9cccaf527fdfe765cfe9b9e88ee86f226c Mon Sep 17 00:00:00 2001 From: Kyotaro Horiguchi <horiguchi.kyotaro@lab.ntt.co.jp> Date: Tue, 31 Jul 2018 18:38:03 +0900 Subject: [PATCH] Documentation fix of "Log-Shipping Standy Servers" for 9.4. The example output of pg_replication_slot is stale. Replaces with the up-to-date example. --- doc/src/sgml/high-availability.sgml | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/doc/src/sgml/high-availability.sgml b/doc/src/sgml/high-availability.sgml index cd5305620d..fce09f1799 100644 --- a/doc/src/sgml/high-availability.sgml +++ b/doc/src/sgml/high-availability.sgml @@ -932,9 +932,9 @@ postgres=# SELECT * FROM pg_create_physical_replication_slot('node_a_slot'); node_a_slot | postgres=# SELECT * FROM pg_replication_slots; - slot_name | slot_type | datoid | database | active | xmin | restart_lsn --------------+-----------+--------+----------+--------+------+------------- - node_a_slot | physical | | | f | | + slot_name | plugin | slot_type | datoid | database | active | xmin | catalog_xmin | restart_lsn +-------------+--------+-----------+--------+----------+--------+------+--------------+------------- + node_a_slot | | physical | | | f | | | (1 row) </programlisting> To configure the standby to use this slot, <varname>primary_slot_name</> -- 2.16.3 From eba78232e51bed3a4dc3fca6e311ece191d163cd Mon Sep 17 00:00:00 2001 From: Kyotaro Horiguchi <horiguchi.kyotaro@lab.ntt.co.jp> Date: Tue, 31 Jul 2018 18:33:20 +0900 Subject: [PATCH] Documentation fix of "Log-Shipping Standy Servers" for 9.5. The example output of pg_replication_slot is stale. Replaces with the up-to-date example. --- doc/src/sgml/high-availability.sgml | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/doc/src/sgml/high-availability.sgml b/doc/src/sgml/high-availability.sgml index 35daf0103a..780851c976 100644 --- a/doc/src/sgml/high-availability.sgml +++ b/doc/src/sgml/high-availability.sgml @@ -932,9 +932,9 @@ postgres=# SELECT * FROM pg_create_physical_replication_slot('node_a_slot'); node_a_slot | postgres=# SELECT * FROM pg_replication_slots; - slot_name | slot_type | datoid | database | active | xmin | restart_lsn --------------+-----------+--------+----------+--------+------+------------- - node_a_slot | physical | | | f | | + slot_name | plugin | slot_type | datoid | database | active | active_pid | xmin | catalog_xmin | restart_lsn +-------------+--------+-----------+--------+----------+--------+------------+------+--------------+------------- + node_a_slot | | physical | | | f | | | | (1 row) </programlisting> To configure the standby to use this slot, <varname>primary_slot_name</> -- 2.16.3 From 8a97a8defc674a03dc63c93439430d88278335cf Mon Sep 17 00:00:00 2001 From: Kyotaro Horiguchi <horiguchi.kyotaro@lab.ntt.co.jp> Date: Tue, 31 Jul 2018 18:29:57 +0900 Subject: [PATCH] Documentation fix of "Log-Shipping Standy Servers" for 9.6. The example output of pg_replication_slot is stale. Replaces with the up-to-date example. --- doc/src/sgml/high-availability.sgml | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/doc/src/sgml/high-availability.sgml b/doc/src/sgml/high-availability.sgml index a52580e3dd..ab6d146b68 100644 --- a/doc/src/sgml/high-availability.sgml +++ b/doc/src/sgml/high-availability.sgml @@ -932,9 +932,9 @@ postgres=# SELECT * FROM pg_create_physical_replication_slot('node_a_slot'); node_a_slot | postgres=# SELECT * FROM pg_replication_slots; - slot_name | slot_type | datoid | database | active | xmin | restart_lsn | confirmed_flush_lsn --------------+-----------+--------+----------+--------+------+-------------+--------------------- - node_a_slot | physical | | | f | | | + slot_name | plugin | slot_type | datoid | database | active | active_pid | xmin | catalog_xmin | restart_lsn | confirmed_flush_lsn +-------------+--------+-----------+--------+----------+--------+------------+------+--------------+-------------+--------------------- + node_a_slot | | physical | | | f | | | | | (1 row) </programlisting> To configure the standby to use this slot, <varname>primary_slot_name</> -- 2.16.3 From 79e433a81fa84feb597cec5da3a049ab355327b4 Mon Sep 17 00:00:00 2001 From: Kyotaro Horiguchi <horiguchi.kyotaro@lab.ntt.co.jp> Date: Tue, 31 Jul 2018 18:44:18 +0900 Subject: [PATCH] Documentation fix of "Log-Shipping Standy Servers" for devel. The example output of pg_replication_slot is stale. Replaces with the up-to-date example. --- doc/src/sgml/high-availability.sgml | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/doc/src/sgml/high-availability.sgml b/doc/src/sgml/high-availability.sgml index 934eb9052d..bd2747de25 100644 --- a/doc/src/sgml/high-availability.sgml +++ b/doc/src/sgml/high-availability.sgml @@ -967,9 +967,9 @@ postgres=# SELECT * FROM pg_create_physical_replication_slot('node_a_slot'); node_a_slot | postgres=# SELECT * FROM pg_replication_slots; - slot_name | slot_type | datoid | database | active | xmin | restart_lsn | confirmed_flush_lsn --------------+-----------+--------+----------+--------+------+-------------+--------------------- - node_a_slot | physical | | | f | | | + slot_name | plugin | slot_type | datoid | database | temporary | active | active_pid | xmin | catalog_xmin | restart_lsn| confirmed_flush_lsn +-------------+--------+-----------+--------+----------+-----------+--------+------------+------+--------------+-------------+--------------------- + node_a_slot | | physical | | | f | f | | | | | (1 row) </programlisting> To configure the standby to use this slot, <varname>primary_slot_name</varname> -- 2.16.3
On 2018-Jul-31, Kyotaro HORIGUCHI wrote: > But, just fixing it makes the line seemingly a bit too long.. How about pasting it like this? alvherre=# select * from pg_replication_slots \gx ─[ RECORD 1 ]───────┬──────────── slot_name │ node_a_slot plugin │ slot_type │ physical datoid │ database │ temporary │ f active │ f active_pid │ xmin │ catalog_xmin │ restart_lsn │ confirmed_flush_lsn │ -- Álvaro Herrera https://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
On Tue, Jul 31, 2018 at 10:59:14AM -0400, Alvaro Herrera wrote: > How about pasting it like this? > > alvherre=# select * from pg_replication_slots \gx > ─[ RECORD 1 ]───────┬──────────── > slot_name │ node_a_slot > plugin │ > slot_type │ physical > datoid │ > database │ > temporary │ f > active │ f > active_pid │ > xmin │ > catalog_xmin │ > restart_lsn │ > confirmed_flush_lsn │ Each time this catalog is changed, this would become incorrect. My suggestion would be to change the query to that and call it a day: SELECT slot_name, slot_type, active FROM pg_replication_slots; -- Michael
Attachment
At Wed, 1 Aug 2018 02:47:32 +0900, Michael Paquier <michael@paquier.xyz> wrote in <20180731174732.GD2878@paquier.xyz>
> On Tue, Jul 31, 2018 at 10:59:14AM -0400, Alvaro Herrera wrote:
> > How about pasting it like this?
> >
> > alvherre=# select * from pg_replication_slots \gx
> > ─[ RECORD 1 ]───────┬────────────
> > slot_name │ node_a_slot
> > plugin │
> > slot_type │ physical
> > datoid │
> > database │
> > temporary │ f
> > active │ f
> > active_pid │
> > xmin │
> > catalog_xmin │
> > restart_lsn │
> > confirmed_flush_lsn │
>
> Each time this catalog is changed, this would become incorrect. My
> suggestion would be to change the query to that and call it a day:
> SELECT slot_name, slot_type, active FROM pg_replication_slots;
Yeah, I'm also concerned about the unstability. I didn't came up
with a list of columns that makes sense at that time but I
rethought on that.
> slot_name │ required
> plugin │ <not requried for physical slots>
> slot_type │ required
> datoid │ <not requried>
> database │ <not requried>
> temporary │ <not requried>
> active │ required?
> active_pid │ <not requried>
> xmin │ <not requried>
> catalog_xmin │ <not requried>
> restart_lsn │ better to be shown?
> confirmed_flush_lsn │ <not requried for physical slots>
The query and the result with four columns fit the current width.
regards.
--
Kyotaro Horiguchi
NTT Open Source Software Center
From 2b5788b3f1ecf4ebd31cf36eecdd4619cf3f394e Mon Sep 17 00:00:00 2001
From: Kyotaro Horiguchi <horiguchi.kyotaro@lab.ntt.co.jp>
Date: Wed, 1 Aug 2018 12:41:03 +0900
Subject: [PATCH] Documentation fix of "Log-Shipping Standy Servers" for PG95.
The example output of pg_replication_slot is wrong. Correct and make
the output stable by explicitly specifying a part of the all columns
in pg_replication_slots.
---
doc/src/sgml/high-availability.sgml | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/doc/src/sgml/high-availability.sgml b/doc/src/sgml/high-availability.sgml
index 35daf0103a..ae831192d0 100644
--- a/doc/src/sgml/high-availability.sgml
+++ b/doc/src/sgml/high-availability.sgml
@@ -931,10 +931,10 @@ postgres=# SELECT * FROM pg_create_physical_replication_slot('node_a_slot');
-------------+---------------
node_a_slot |
-postgres=# SELECT * FROM pg_replication_slots;
- slot_name | slot_type | datoid | database | active | xmin | restart_lsn
--------------+-----------+--------+----------+--------+------+-------------
- node_a_slot | physical | | | f | |
+postgres=# SELECT slot_name, slot_type, active, restart_lsn FROM pg_replication_slots;
+ slot_name | slot_type | active | restart_lsn
+-------------+-----------+--------+-------------
+ node_a_slot | physical | f |
(1 row)
</programlisting>
To configure the standby to use this slot, <varname>primary_slot_name</>
--
2.16.3
From 4367cdea3d6ac19ef9c6ec1545fbc2329dda47bb Mon Sep 17 00:00:00 2001
From: Kyotaro Horiguchi <horiguchi.kyotaro@lab.ntt.co.jp>
Date: Wed, 1 Aug 2018 11:50:03 +0900
Subject: [PATCH] Documentation fix of "Log-Shipping Standy Servers" for
master.
The example output of pg_replication_slot is wrong. Correct and make
the output stable by explicitly specifying a part of the all columns
in pg_replication_slots.
---
doc/src/sgml/high-availability.sgml | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/doc/src/sgml/high-availability.sgml b/doc/src/sgml/high-availability.sgml
index 934eb9052d..39cada1278 100644
--- a/doc/src/sgml/high-availability.sgml
+++ b/doc/src/sgml/high-availability.sgml
@@ -966,10 +966,10 @@ postgres=# SELECT * FROM pg_create_physical_replication_slot('node_a_slot');
-------------+-----
node_a_slot |
-postgres=# SELECT * FROM pg_replication_slots;
- slot_name | slot_type | datoid | database | active | xmin | restart_lsn | confirmed_flush_lsn
--------------+-----------+--------+----------+--------+------+-------------+---------------------
- node_a_slot | physical | | | f | | |
+postgres=# SELECT slot_name, slot_type, active, restart_lsn FROM pg_replication_slots;
+ slot_name | slot_type | active | restart_lsn
+-------------+-----------+--------+-------------
+ node_a_slot | physical | f |
(1 row)
</programlisting>
To configure the standby to use this slot, <varname>primary_slot_name</varname>
--
2.16.3
On Wed, Aug 01, 2018 at 01:04:37PM +0900, Kyotaro HORIGUCHI wrote: > The query and the result with four columns fit the current width. Just wondering, what is your reason behind the addition of restart_lsn? This part of the documentation focuses on slot creation, so slot_name, slot_type and active would be representative enough, no? -- Michael
Attachment
At Fri, 3 Aug 2018 04:13:56 +0900, Michael Paquier <michael@paquier.xyz> wrote in <20180802191356.GB2247@paquier.xyz>
> On Wed, Aug 01, 2018 at 01:04:37PM +0900, Kyotaro HORIGUCHI wrote:
> > The query and the result with four columns fit the current width.
>
> Just wondering, what is your reason behind the addition of restart_lsn?
> This part of the documentation focuses on slot creation, so slot_name,
> slot_type and active would be representative enough, no?
The first reason was the example looks having a bit too many
space around only with three other columns.
A more serious reason is that I wanted to show a slot with what
properties is created in the example and I thought that
restart_lsn and temporary are significant. Howerver restart_lsn
is always a part of the pg_replication_slots, we don't have
"temprary" before 10. Addition to that, adding it makes the
SELECT line stick out of the width.
That said, I don't object to reduce the columns. Please find the
attached.
regards.
--
Kyotaro Horiguchi
NTT Open Source Software Center
From 7e750c902f594b0a00c1f579518a74947a755d05 Mon Sep 17 00:00:00 2001
From: Kyotaro Horiguchi <horiguchi.kyotaro@lab.ntt.co.jp>
Date: Wed, 1 Aug 2018 11:50:03 +0900
Subject: [PATCH] Documentation fix of "Log-Shipping Standy Servers" for
master.
The example output of pg_replication_slot is wrong. Correct and make
the output stable by explicitly specifying a part of the all columns
in pg_replication_slots.
---
doc/src/sgml/high-availability.sgml | 10 ++++++----
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git a/doc/src/sgml/high-availability.sgml b/doc/src/sgml/high-availability.sgml
index 934eb9052d..3c5f3fc6a9 100644
--- a/doc/src/sgml/high-availability.sgml
+++ b/doc/src/sgml/high-availability.sgml
@@ -966,10 +966,12 @@ postgres=# SELECT * FROM pg_create_physical_replication_slot('node_a_slot');
-------------+-----
node_a_slot |
-postgres=# SELECT * FROM pg_replication_slots;
- slot_name | slot_type | datoid | database | active | xmin | restart_lsn | confirmed_flush_lsn
--------------+-----------+--------+----------+--------+------+-------------+---------------------
- node_a_slot | physical | | | f | | |
+postgres=# SELECT slot_name, slot_type, active FROM pg_replication_slots;
+ slot_name | slot_type | active
+-------------+-----------+--------
+ node_a_slot | physical | f
+(1 row)
+
(1 row)
</programlisting>
To configure the standby to use this slot, <varname>primary_slot_name</varname>
--
2.16.3
From 0192ad0d7c674eff11285f68ea2128ad50ecef10 Mon Sep 17 00:00:00 2001
From: Kyotaro Horiguchi <horiguchi.kyotaro@lab.ntt.co.jp>
Date: Wed, 1 Aug 2018 12:41:03 +0900
Subject: [PATCH] Documentation fix of "Log-Shipping Standy Servers" for PG95.
The example output of pg_replication_slot is wrong. Correct and make
the output stable by explicitly specifying a part of the all columns
in pg_replication_slots.
---
doc/src/sgml/high-availability.sgml | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/doc/src/sgml/high-availability.sgml b/doc/src/sgml/high-availability.sgml
index 35daf0103a..3f33c39bd2 100644
--- a/doc/src/sgml/high-availability.sgml
+++ b/doc/src/sgml/high-availability.sgml
@@ -931,10 +931,10 @@ postgres=# SELECT * FROM pg_create_physical_replication_slot('node_a_slot');
-------------+---------------
node_a_slot |
-postgres=# SELECT * FROM pg_replication_slots;
- slot_name | slot_type | datoid | database | active | xmin | restart_lsn
--------------+-----------+--------+----------+--------+------+-------------
- node_a_slot | physical | | | f | |
+postgres=# SELECT slot_name, slot_type, active FROM pg_replication_slots;
+ slot_name | slot_type | active
+-------------+-----------+--------
+ node_a_slot | physical | f
(1 row)
</programlisting>
To configure the standby to use this slot, <varname>primary_slot_name</>
--
2.16.3
On 2018-Aug-03, Kyotaro HORIGUCHI wrote: > That said, I don't object to reduce the columns. Please find the > attached. Thanks, pushed. -- Álvaro Herrera https://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
On Fri, Aug 03, 2018 at 04:37:05PM -0400, Alvaro Herrera wrote: > On 2018-Aug-03, Kyotaro HORIGUCHI wrote: >> That said, I don't object to reduce the columns. Please find the >> attached. > > Thanks, pushed. Thanks Alvaro for keeping the three-column version, I was going to look at the proposed patch and push as you did, until I noticed that you showed up :) -- Michael
Attachment
At Sat, 4 Aug 2018 05:58:52 +0900, Michael Paquier <michael@paquier.xyz> wrote in <20180803205852.GB20967@paquier.xyz> > On Fri, Aug 03, 2018 at 04:37:05PM -0400, Alvaro Herrera wrote: > > On 2018-Aug-03, Kyotaro HORIGUCHI wrote: > >> That said, I don't object to reduce the columns. Please find the > >> attached. > > > > Thanks, pushed. > > Thanks Alvaro for keeping the three-column version, I was going to look > at the proposed patch and push as you did, until I noticed that you > showed up :) Thanks. # I noticed that the subject has typo.. regards. -- Kyotaro Horiguchi NTT Open Source Software Center
On 2018-Aug-09, Kyotaro HORIGUCHI wrote: > # I noticed that the subject has typo.. Yeah. I suggest never changing subject lines, because Gmail has the nasty (mis-)feature of making such a response into a completely new thread. I don't know if Google paid mail service behaves in the same way. So if you change Subject, please do include a URL to the previous thread if necessary, for the benefit of people reading on Gmail. I wouldn't worry about this for any other individual email provider, but Gmail is by far the largest fraction of subscribers :-( I guess this shows just how much better Google made webmail systems than what existed at the time. -- Álvaro Herrera https://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
On Thu, Aug 09, 2018 at 03:58:09PM -0400, Alvaro Herrera wrote: > Yeah. I suggest never changing subject lines, because Gmail has the > nasty (mis-)feature of making such a response into a completely new > thread. I don't know if Google paid mail service behaves in the same > way. If one uses Gmail with another client like mutt, then those don't get broken. I don't know about Thunderbird which I believe people use a lot. > So if you change Subject, please do include a URL to the previous thread > if necessary, for the benefit of people reading on Gmail. +1. > I wouldn't worry about this for any other individual email provider, but > Gmail is by far the largest fraction of subscribers :-( I guess this > shows just how much better Google made webmail systems than what existed > at the time. If something is free, you are the product, still I have to admit that Gmail is not especially bad. -- Michael
Attachment
On 2018-Aug-03, Alvaro Herrera wrote: > On 2018-Aug-03, Kyotaro HORIGUCHI wrote: > > > That said, I don't object to reduce the columns. Please find the > > attached. > > Thanks, pushed. I had failed to push in the 9.4 branch. Done now. -- Álvaro Herrera https://www.2ndQuadrant.com/ PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services
On Wed, Aug 15, 2018 at 02:35:56PM -0300, Alvaro Herrera wrote: > I had failed to push in the 9.4 branch. Done now. Thanks Alvaro for working on the details. I would not have bothered much myself as the set of columns was correct in 9.4, but for consistency's sake that makes sense. -- Michael