[pgpool-hackers: 3629] Re: DOC:wd_priority description is too promising

[Kenichiro Tanaka]

Hello,

I've attached English and Japanese documentation patch .
(I've add more description to my last patch)
Could you check it?

diff --git a/doc.ja/src/sgml/watchdog.sgml b/doc.ja/src/sgml/watchdog.sgml
index 51f09bc..0b6c659 100644
--- a/doc.ja/src/sgml/watchdog.sgml
+++ b/doc.ja/src/sgml/watchdog.sgml
@@ -751,10 +751,17 @@ watchdogプロセスがネットワークリンクの状態を監視するネッ
       to select master watchdog node.
       The node with the higher <varname>wd_priority</varname> value
will get selected
       as master watchdog node when cluster will be electing its new master node
-      at the time of cluster startup or in the event of old master
watchdog node failure
+      in the event of old master watchdog node failure.
+      <varname>wd_priority</varname> is also valid at the time of
cluster startup.
+      When some watchdog nodes start up at same time,a node with the
higher <varname>wd_priority</varname>
+      value is selected as a master node.
+      So we should start watchdog nodes order by
<varname>wd_priority</varname> priority to prevent unintended nodes
from being selected as masters.
 -->
 このパラメータによってローカルのwatchdogノードがマスターに選ばれる優先度を上げることができます。
-クラスタの初期起動時や古いマスターノードが故障した状況でクラスタがマスターノードの選択を行う際に、<varname>wd_priority</varname>が高いノードがマスターwatchdogノードに選ばれます。
+古いマスターノードが故障した状況でクラスタがマスターノードの選択を行う際に、<varname>wd_priority</varname>が高いノードがマスターwatchdogノードに選ばれます。
+<varname>wd_priority</varname>はクラスタの初期起動時にも有効なので、
+複数のwatchdogノードを同時に起動した場合、<varname>wd_priority</varname>が高いノードがマスターwatchdogノードに選ばれます。
+このため、watchdogノードは意図しないノードがマスターに選ばれることが無いように<varname>wd_priority</varname>の優先順位が高い順に起動するべきです。
    </para>
   <para>
 <!--
diff --git a/doc/src/sgml/watchdog.sgml b/doc/src/sgml/watchdog.sgml
index 09549af..1bff1f8 100644
--- a/doc/src/sgml/watchdog.sgml
+++ b/doc/src/sgml/watchdog.sgml
@@ -486,7 +486,11 @@
       to select master watchdog node.
       The node with the higher <varname>wd_priority</varname> value
will get selected
       as master watchdog node when cluster will be electing its new master node
-      at the time of cluster startup or in the event of old master
watchdog node failure
+      in the event of old master watchdog node failure.
+      <varname>wd_priority</varname> is also valid at the time of
cluster startup.
+      When some watchdog nodes start up at same time,a node with the
higher <varname>wd_priority</varname>
+      value is selected as a master node.
+      So we should start watchdog nodes order by
<varname>wd_priority</varname> priority to prevent unintended nodes
from being selected as masters.
    </para>
   <para>
      <varname>wd_priority</varname> is not available in versions prior to

Regards
Kenichiro Tanaka

2020年5月19日(火) 22:37 Kenichiro Tanaka <kenichirotanakapg at gmail.com>:

>
> Hello
> Thank you for your response.
>
> >Because we maintain English and Japanese documentation,
> >is it possible for you to create a patch for Japanese documentaion?
> Off course.
> I'll send it later.
>
> Regards Kenichiro Tanaka.
>
>
> 2020年5月19日(火) 16:52 Bo Peng <pengbo at sraoss.co.jp>:
> >
> > Hello,
> >
> > On Tue, 19 May 2020 09:33:35 +0900
> > Kenichiro Tanaka <kenichirotanakapg at gmail.com> wrote:
> >
> > > Hello pgpool hackers.
> > >
> > > I fell we should change and add some description about  "wd_priority"
> > > explanation of pgpool-II document .
> >
> > Thank you for your patch. Looks good.
> >
> > Because we maintain English and Japanese documentation,
> > is it possible for you to create a patch for Japanese documentaion?
> >
> > > https://www.pgpool.net/docs/latest/en/html/runtime-watchdog-config.html
> > >
> > > -----------------------------------------------------------------------
> > > wd_priority (integer)
> > > This parameter can be used to elevate the local watchdog node priority
> > > in the elections to select master watchdog node. The node with the higher
> > > wd_priority value will get selected as master watchdog node when cluster
> > > will be electing its new master node at the time of cluster startup or in
> > > the event of old master watchdog node failure
> > > -----------------------------------------------------------------------
> > > Documentation says "cluster will be electing its new master node at the
> > > time of cluster *startup* or in the event of old master watchdog node failure".
> > > This description allows the user to expect higher wd_priority node always
> > > to become new master node.But actually it is not elected if it starts late
> > > (This is natural behavior for cluster).
> > >
> > > So I think we should add or change some description about it.
> > >
> > > For example,
> > >
> > > diff --git a/doc/src/sgml/watchdog.sgml b/doc/src/sgml/watchdog.sgml
> > > index 09549af..30db22d 100644
> > > --- a/doc/src/sgml/watchdog.sgml
> > > +++ b/doc/src/sgml/watchdog.sgml
> > > @@ -486,7 +486,11 @@
> > >        to select master watchdog node.
> > >        The node with the higher <varname>wd_priority</varname> value
> > > will get selected
> > >        as master watchdog node when cluster will be electing its new master node
> > > -      at the time of cluster startup or in the event of old master
> > > watchdog node failure
> > > +      in the event of old master watchdog node failure.
> > > +      <varname>wd_priority</varname> is also valid at the time of
> > > cluster startup.
> > > +      When some watchdog nodes start up at same time,a node with the
> > > higher <varname>wd_priority</varname>
> > > +      value is selected as a master node.
> > > +      So we should start watchdog nodes order by
> > > <varname>wd_priority</varname> priority.
> > >     </para>
> > >    <para>
> > >       <varname>wd_priority</varname> is not available in versions prior to
> > >
> > > Regards Kenichiro Tanaka
> > > _______________________________________________
> > > pgpool-hackers mailing list
> > > pgpool-hackers at pgpool.net
> > > http://www.pgpool.net/mailman/listinfo/pgpool-hackers
> >
> >
> > --
> > Bo Peng <pengbo at sraoss.co.jp>
> > SRA OSS, Inc. Japan