Date: Fri, 16 Oct 2020 16:21:47 +0800
From: "movead.li@highgo.ca" <movead.li@highgo.ca>
To: Horiguchi <horikyota.ntt@gmail.com>
Cc: andres <andres@anarazel.de>,
	michael <michael@paquier.xyz>,
	pgsql-hackers <pgsql-hackers@postgresql.org>,
	ahsan.hadi <ahsan.hadi@highgo.ca>
Subject: Re: Wrong statistics for size of XLOG_SWITCH during pg_waldump.
References: <20201014204613.lxts3w6lc4lsv7zh@alap3.anarazel.de>,
	<20201015.114443.2231235644531652368.horikyota.ntt@gmail.com>,
	<2020101512560020200319@highgo.ca>,
	<20201015.173210.1595235318844221199.horikyota.ntt@gmail.com>
Mime-Version: 1.0
Message-ID: <2020101616214467451338@highgo.ca>
Content-Type: multipart/mixed;
	boundary="----=_001_NextPart485402434310_=----"
Precedence: bulk

This is a multi-part message in MIME format.

------=_001_NextPart485402434310_=----
Content-Type: multipart/alternative;
	boundary="----=_002_NextPart611358342521_=----"


------=_002_NextPart611358342521_=----
Content-Type: text/plain;
	charset="ISO-8859-1"
Content-Transfer-Encoding: base64
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------=_002_NextPart611358342521_=----
Content-Type: text/html;
	charset="ISO-8859-1"
Content-Transfer-Encoding: quoted-printable

<html><head><meta http-equiv=3D"content-type" content=3D"text/html; charse=
t=3DISO-8859-1"><style>body { line-height: 1.5; }blockquote { margin-top: =
0px; margin-bottom: 0px; margin-left: 0.5em; }body { font-size: 10.5pt; fo=
nt-family: 'Microsoft YaHei UI'; color: rgb(0, 0, 0); line-height: 1.5; }<=
/style></head><body>=0A<div><span></span>Thanks for all the suggestion, an=
d new patch attached.</div><div><br></div>=0A<div><div>&gt;Andres suggeste=
d that we don't need that description with per-record</div><div><span styl=
e=3D"font-size: 10.5pt; line-height: 1.5; background-color: transparent;">=
&gt;</span>basis. Do you have a reason to do that?&nbsp; (For clarity, I'm=
 not</div><div><span style=3D"font-size: 10.5pt; line-height: 1.5; backgro=
und-color: transparent;">&gt;</span>suggesting that you should reving it.)=
</div></div><div>I think Andres is saying if we just log it in&nbsp;<span =
style=3D"font-family: 'Microsoft YaHei UI', Tahoma; line-height: 22px; whi=
te-space: pre; font-size: 10.5pt; background-color: transparent;">xlog_des=
c() then we can not get</span></div><div><span style=3D"font-family: 'Micr=
osoft YaHei UI', Tahoma; line-height: 22px; white-space: pre; font-size: 1=
0.5pt; background-color: transparent;">the result for '</span><span style=
=3D"font-family: 'Microsoft YaHei UI', Tahoma; line-height: normal; font-s=
ize: 10.5pt; background-color: transparent;">--stats=3Drecord</span><span =
style=3D"font-family: 'Microsoft YaHei UI', Tahoma; font-size: 10.5pt; lin=
e-height: 22px; white-space: pre; background-color: transparent;">' case. =
And the patch solve the problem.</span></div><div><span style=3D"font-fami=
ly: 'Microsoft YaHei UI', Tahoma; font-size: 10.5pt; line-height: 22px; wh=
ite-space: pre; background-color: transparent;"><br></span></div><div><div=
>&gt;+	XLByteToSeg(record-&gt;EndRecPtr - 1, endSegNo, record-&gt;segcxt.w=
s_segsize);</div><div><span style=3D"font-size: 10.5pt; line-height: 1.5; =
background-color: transparent;">&gt;</span>We use XLByteToPrevSeg instead =
for this purpose.</div></div><div><span style=3D"font-family: 'Microsoft Y=
aHei UI', Tahoma; font-size: 10.5pt; line-height: 22px; white-space: pre; =
background-color: transparent;">Thanks and follow the suggestion.</span></=
div><div><span style=3D"font-family: 'Microsoft YaHei UI', Tahoma; font-si=
ze: 10.5pt; line-height: 22px; white-space: pre; background-color: transpa=
rent;"><br></span></div><div><span style=3D"font-family: 'Microsoft YaHei =
UI', Tahoma; font-size: 10.5pt; line-height: 22px; white-space: pre; backg=
round-color: transparent;">&gt;</span><span style=3D"font-size: 10.5pt; li=
ne-height: 1.5; background-color: transparent;">You forgot to add a correc=
tion for short headers.</span></div><div><span style=3D"font-size: 10.5pt;=
 line-height: 1.5; background-color: transparent;">Infact, we need to cons=
ider this matter when&nbsp;</span><span style=3D"font-family: 'Microsoft Y=
aHei UI', Tahoma; line-height: 22px; white-space: pre; font-size: 10.5pt; =
background-color: transparent;">the </span><span style=3D"font-size: 10.5p=
t; line-height: 22px; white-space: pre; background-color: transparent;"><f=
ont face=3D"Microsoft YaHei UI, Tahoma">remain&nbsp;size&nbsp;of&nbsp;a&nb=
sp;wal</font></span></div><div><span style=3D"font-size: 10.5pt; line-heig=
ht: 22px; white-space: pre; background-color: transparent;"><font face=3D"=
Microsoft YaHei UI, Tahoma">segment&nbsp;can&nbsp;not </font></span><span =
style=3D"font-size: 10.5pt; font-family: 'Microsoft YaHei UI', Tahoma; lin=
e-height: 22px; white-space: pre; background-color: transparent;">afford&n=
bsp;a&nbsp;XLogRecord&nbsp;struct&nbsp;for </span><span style=3D"font-fami=
ly: 'Microsoft YaHei UI', Tahoma; font-size: 10.5pt; line-height: 22px; wh=
ite-space: pre; background-color: transparent;">XLOG_SWITCH.&nbsp;</span><=
/div><div><span style=3D"font-family: 'Microsoft YaHei UI', Tahoma; font-s=
ize: 10.5pt; line-height: 22px; white-space: pre; background-color: transp=
arent;">It's </span><span style=3D"font-family: 'Microsoft YaHei UI', Taho=
ma; font-size: 10.5pt; line-height: 22px; white-space: pre; background-col=
or: transparent;">mean that if </span><span style=3D"font-family: 'Microso=
ft YaHei UI', Tahoma; font-size: 10.5pt; line-height: 22px; white-space: p=
re; background-color: transparent;">record-&gt;ReadRecPtr is on A wal segm=
ent, then</span></div><div><span style=3D"font-family: 'Microsoft YaHei UI=
', Tahoma; line-height: 22px; white-space: pre; font-size: 10.5pt; backgro=
und-color: transparent;">record-&gt;EndRecPtr is on (A+2) wal segment. </s=
pan><span style=3D"font-family: 'Microsoft YaHei UI', Tahoma; font-size: 1=
0.5pt; line-height: 22px; white-space: pre; background-color: transparent;=
">So we need to minus</span></div><div><span style=3D"font-family: 'Micros=
oft YaHei UI', Tahoma; font-size: 10.5pt; line-height: 22px; white-space: =
pre; background-color: transparent;">the longpagehead size on (A+1) wal se=
gment.</span></div><div><font face=3D"Microsoft YaHei UI, Tahoma"><span st=
yle=3D"line-height: 22px; white-space: pre;">In my thought we need not to =
care the </span></font><span style=3D"font-size: 10.5pt; line-height: 1.5;=
 background-color: transparent;">short header, if my understand</span></di=
v><div><span style=3D"font-size: 10.5pt; line-height: 1.5; background-colo=
r: transparent;">is wrong?</span></div><div><span style=3D"font-size: 10.5=
pt; line-height: 1.5; background-color: transparent;"><br></span></div><di=
v><div>&gt;+	if(RM_XLOG_ID =3D=3D rmid &amp;&amp; XLOG_SWITCH =3D=3D info)=
</div><div><span style=3D"font-size: 10.5pt; line-height: 1.5; background-=
color: transparent;">&gt;</span>&nbsp;</div><div><span style=3D"font-size:=
 10.5pt; line-height: 1.5; background-color: transparent;">&gt;</span>We n=
eed a comment for the special code path.</div><div><span style=3D"font-siz=
e: 10.5pt; line-height: 1.5; background-color: transparent;">&gt;</span>We=
 don't follow this kind of convension. Rather use "variable =3D</div><div>=
<span style=3D"font-size: 10.5pt; line-height: 1.5; background-color: tran=
sparent;">&gt;</span>constant".</div><div><div><span style=3D"font-size: 1=
0.5pt; line-height: 1.5; background-color: transparent;">&gt;</span>+	{</d=
iv><div><span style=3D"font-size: 10.5pt; line-height: 1.5; background-col=
or: transparent;">&gt;</span>+	junk_len =3D xlog_switch_junk_len(record);<=
/div><div><span style=3D"font-size: 10.5pt; line-height: 1.5; background-c=
olor: transparent;">&gt;</span>+	stats-&gt;count_xlog_switch++;</div><div>=
<span style=3D"font-size: 10.5pt; line-height: 1.5; background-color: tran=
sparent;">&gt;</span>+	stats-&gt;junk_size +=3D junk_len;</div><div><span =
style=3D"font-size: 10.5pt; line-height: 1.5; background-color: transparen=
t;">&gt;</span>&nbsp;</div><div><span style=3D"font-size: 10.5pt; line-hei=
ght: 1.5; background-color: transparent;">&gt;</span>junk_len is used only=
 the last line above.&nbsp; We don't need that</div><div><span style=3D"fo=
nt-size: 10.5pt; line-height: 1.5; background-color: transparent;">&gt;</s=
pan>temporary variable.</div><div><span style=3D"font-size: 10.5pt; line-h=
eight: 1.5; background-color: transparent;">&gt;</span>&nbsp;</div><div><s=
pan style=3D"font-size: 10.5pt; line-height: 1.5; background-color: transp=
arent;">&gt;</span>+	* If the wal switch record spread on two segments, we=
 should extra minus the</div><div>&gt;This comment is sticking out of 80-c=
olumn border.&nbsp; However, I'm not</div><div>&gt;sure if we have reached=
 a conclustion about the target console-width.</div></div><div><div><span =
style=3D"font-size: 10.5pt; line-height: 1.5; background-color: transparen=
t;">&gt;</span>+	info =3D (rj &lt;&lt; 4) &amp; ~XLR_INFO_MASK;</div><div>=
<span style=3D"font-size: 10.5pt; line-height: 1.5; background-color: tran=
sparent;">&gt;</span>+	switch_junk_id =3D "XLOG/SWITCH_JUNK";</div><div><s=
pan style=3D"font-size: 10.5pt; line-height: 1.5; background-color: transp=
arent;">&gt;</span>+	if( XLOG_SWITCH =3D=3D info &amp;&amp; stats-&gt;coun=
t_xlog_switch &gt; 0)</div><div><span style=3D"font-size: 10.5pt; line-hei=
ght: 1.5; background-color: transparent;">&gt;</span>&nbsp;</div><div><spa=
n style=3D"font-size: 10.5pt; line-height: 1.5; background-color: transpar=
ent;">&gt;</span>This line order is strange. At least switch_junk_id is us=
ed only in</div><div><span style=3D"font-size: 10.5pt; line-height: 1.5; b=
ackground-color: transparent;">&gt;</span>the if-then block.</div></div><d=
iv><span style=3D"font-family: 'Microsoft YaHei UI', Tahoma; line-height: =
22px; white-space: pre; font-size: 10.5pt; background-color: transparent;"=
>Thanks and follow the suggestions.</span></div><div><br></div><div>&nbsp;=
</div><div>&gt;I'm not confindent on which is better, but I feel that this=
 is not a</div><div><span style=3D"font-size: 10.5pt; line-height: 1.5; ba=
ckground-color: transparent;">&gt;</span>work for display side, but for th=
e recorder side like attached.</div></div><div><span style=3D"font-family:=
 'Microsoft YaHei UI', Tahoma; line-height: 22px; white-space: pre; font-s=
ize: 10.5pt; background-color: transparent;">The patch really seems more c=
learly, but the new 'OTHERS' may confuse</span></div><div><span style=3D"f=
ont-family: 'Microsoft YaHei UI', Tahoma; line-height: 22px; white-space: =
pre; font-size: 10.5pt; background-color: transparent;">the users and we h=
ard to handle it with '</span><span style=3D"background-color: transparent=
; line-height: 22px; white-space: pre;"><font face=3D"Microsoft YaHei UI, =
Tahoma">--rmgr=3DRMGR</font></span><span style=3D"font-family: 'Microsoft =
YaHei UI', Tahoma; font-size: 10.5pt; line-height: 22px; white-space: pre;=
 background-color: transparent;">' option. So I have</span></div><div><spa=
n style=3D"font-family: 'Microsoft YaHei UI', Tahoma; font-size: 10.5pt; l=
ine-height: 22px; white-space: pre; background-color: transparent;">not us=
e this design in this patch, let's see other's opinion.</span></div><div><=
br></div><div><div>&gt;Sorry for the confusion, but it would be a separate=
 topic even if we</div><div><span style=3D"font-size: 10.5pt; line-height:=
 1.5; background-color: transparent;">&gt;</span>are going to fix that..</=
div></div><div>Sorry, I remove the code, make sense we should discuss it i=
n a</div><div><span style=3D"font-size: 10.5pt; line-height: 1.5; backgrou=
nd-color: transparent;">separate topic.</span></div><div><br></div><hr sty=
le=3D"width: 210px; height: 1px;" color=3D"#b5c4df" size=3D"1" align=3D"le=
ft">=0A<div><span><div style=3D"MARGIN: 10px; FONT-FAMILY: verdana; FONT-S=
IZE: 10pt"><span style=3D"color: rgb(13, 10, 11); font-family: monospace, =
monospace; font-size: 10.7333px; font-variant-ligatures: normal; orphans: =
2; white-space: pre-wrap; widows: 2;">Regards,</span></div><div style=3D"M=
ARGIN: 10px; FONT-FAMILY: verdana; FONT-SIZE: 10pt"><span style=3D"color: =
rgb(13, 10, 11); font-family: monospace, monospace; font-size: 10.7333px; =
font-variant-ligatures: normal; orphans: 2; white-space: pre-wrap; widows:=
 2;">Highgo Software (Canada/China/Pakistan) </span><br style=3D"box-sizin=
g: border-box; color: rgb(13, 10, 11); font-family: monospace, monospace; =
font-size: 10.7333px; font-variant-ligatures: normal; orphans: 2; white-sp=
ace: pre-wrap; widows: 2;"><span style=3D"color: rgb(13, 10, 11); font-fam=
ily: monospace, monospace; font-size: 10.7333px; font-variant-ligatures: n=
ormal; orphans: 2; white-space: pre-wrap; widows: 2;">URL : </span><a href=
=3D"http://www.highgo.ca/" rel=3D"nofollow" style=3D"box-sizing: border-bo=
x; color: rgb(132, 0, 50); text-decoration: none; font-weight: 600; transi=
tion: color 0.2s ease-in-out 0s; font-family: monospace, monospace; font-s=
ize: 10.7333px; font-variant-ligatures: normal; orphans: 2; white-space: p=
re-wrap; widows: 2;">www.highgo.ca</a><span style=3D"color: rgb(13, 10, 11=
); font-family: monospace, monospace; font-size: 10.7333px; font-variant-l=
igatures: normal; orphans: 2; white-space: pre-wrap; widows: 2;"> </span><=
br style=3D"box-sizing: border-box; color: rgb(13, 10, 11); font-family: m=
onospace, monospace; font-size: 10.7333px; font-variant-ligatures: normal;=
 orphans: 2; white-space: pre-wrap; widows: 2;"><span style=3D"color: rgb(=
13, 10, 11); font-family: monospace, monospace; font-size: 10.7333px; font=
-variant-ligatures: normal; orphans: 2; white-space: pre-wrap; widows: 2;"=
>EMAIL: mailto:movead(dot)li(at)highgo(dot)ca</span></div></span></div><bl=
ockquote style=3D"margin-Top: 0px; margin-Bottom: 0px; margin-Left: 0.5em;=
 margin-Right: inherit"><div>=0A</div></blockquote>=0A</body></html>
------=_002_NextPart611358342521_=------

------=_001_NextPart485402434310_=----
Content-Type: application/octet-stream;
	name="fix_waldump_size_for_wal_switch_v4.patch"
Content-Transfer-Encoding: base64
Content-Disposition: attachment;
	filename="fix_waldump_size_for_wal_switch_v4.patch"
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------=_001_NextPart485402434310_=------