MIME-Version: 1.0
References: <CAD1Ag0wW0nw-BR1aqKkD66LLO_yndJAy__i6yg9TWvU_Z6QEDA.ref@mail.gmail.com>
 <CAD1Ag0wW0nw-BR1aqKkD66LLO_yndJAy__i6yg9TWvU_Z6QEDA@mail.gmail.com>
 <20240930202728._Cgmke2m@steffen%sdaoden.eu> <CAD1Ag0xwXkX7yuanRrcpnSzNCNLXHp9Bx76=HYoecbhCWPpJDw@mail.gmail.com>
In-Reply-To: <CAD1Ag0xwXkX7yuanRrcpnSzNCNLXHp9Bx76=HYoecbhCWPpJDw@mail.gmail.com>
From: Jerome Marc <marcjero@yahoo.com>
Date: Tue, 1 Oct 2024 14:51:17 +0200
Message-ID: <CAD1Ag0zxPH8RuDrscixr+_VYu27VC=2S1B=XTHQ=H1=WNDJAjQ@mail.gmail.com>
Subject: Re: freeze / connection delay after idling for a while ?
To: ~alpine/users@lists.alpinelinux.org
Content-Type: multipart/alternative; boundary="0000000000002b71d1062369c865"

--0000000000002b71d1062369c865
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Ok I found the culprit. It was a network issue. The problem was happening
when initiating the connection from a wlan which is bridged with the wired
lan.
For some reason, my Internet router forgets the presence of the RPI.
Therefore IP address resolution can be slow. I suppose it's because there
is no traffic coming from the RPI.
I installed the avahi daemon, I think it will make the RPI visible
continuously.


Le lun. 30 sept. 2024 =C3=A0 23:20, Jerome Marc <marcjero@yahoo.com> a =C3=
=A9crit :

> Hi Steffen,
>
> Thanks !
> Unfortunately there is cpupower rc service running and I don't have wifi.
> I installed cpupower and got this output :
>
> alpine:~# cpupower frequency-info
> analyzing CPU 0:
>   driver: cpufreq-dt
>   CPUs which run at the same hardware frequency: 0 1 2 3
>   CPUs which need to have their frequency coordinated by software: 0 1 2 =
3
>   maximum transition latency:  Cannot determine or is not supported.
>   hardware limits: 600 MHz - 900 MHz
>   available frequency steps:  600 MHz, 700 MHz, 800 MHz, 900 MHz
>   available cpufreq governors: conservative ondemand userspace performanc=
e
> schedutil
>   current policy: frequency should be within 600 MHz and 900 MHz.
>                   The governor "schedutil" may decide which speed to use
>                   within this range.
>   current CPU frequency: 600 MHz (asserted by call to hardware)
>
> Looks good...
>
> Le lun. 30 sept. 2024 =C3=A0 22:27, Steffen Nurpmeso <steffen@sdaoden.eu>=
 a
> =C3=A9crit :
>
>> Jerome Marc wrote in
>>  <CAD1Ag0wW0nw-BR1aqKkD66LLO_yndJAy__i6yg9TWvU_Z6QEDA@mail.gmail.com>:
>>  |I just installed Alpine 3.20.3 on a RPI 2 in diskless mode.(headless,
>>  |minimal installation)
>>  |
>>  |I noticed that if I let the device idling, I have a few seconds latenc=
y
>>  |when I try to connect the machine with SSH for example. Even when
>> connected
>>  |I can have my SSH session frozen for a few seconds if I do nothing in
>> the
>>  |terminal.
>>  |I notice the same when using a nodejs app (HTTP) therefore this is a
>>  |general thing (not just about SSH)
>>  |
>>  |I don't see anything weird in logs. Any idea ? Is that a power saving
>>  |feature ?
>>
>> It *could* be that cpupower-openrc is indeed the
>> tools/power/cpupower of the linux kernel sources, it seems to be
>> widely available (architecture-wise, on Alpine).
>> If that is so, then running it should output stuff like
>>
>>   #?0|kent:~# bin/cpupower.sh
>>   analyzing CPU 0:
>>     driver: intel_pstate
>>     CPUs which run at the same hardware frequency: 0
>>     CPUs which need to have their frequency coordinated by software: 0
>>     maximum transition latency:  Cannot determine or is not supported.
>>     hardware limits: 400 MHz - 3.40 GHz
>>     available cpufreq governors: performance powersave
>>     current policy: frequency should be within 400 MHz and 3.40 GHz.
>>                     The governor "powersave" may decide which speed to u=
se
>>                     within this range.
>>     current CPU frequency: Unable to call hardware
>>     current CPU frequency: 800 MHz (asserted by call to kernel)
>>     boost state support:
>>       Supported: yes
>>       Active: yes
>>
>> On the reserve laptop however the EFI(BIOS) "always" sets back
>> cpupower, and we start with "powersave" governor and low
>> frequency, and so the initial decrypting of the hard disk takes
>> seeeeconds.  I btw drive it via
>>
>>   #!/bin/sh -
>>   #@ /root/bin/cpupower.sh
>>   # cpupower is in Linux src, tools/power/cpupower
>>
>>   : ${HOSTNAME:=3D$(uname -n)}
>>
>>   if [ -f /root/hosts/${HOSTNAME}/cpupower ]; then
>>           . /root/hosts/${HOSTNAME}/cpupower
>>   else
>>           logger -s -t /root/bin/cpupower.sh "MISS
>> /root/hosts/${HOSTNAME}/cpupower"
>>           exit 1
>>   fi
>>
>>   if command -v cpupower >/dev/null 2>&1; then :; else
>>           logger -s -t /root/bin/cpupower.sh 'no cpupower tool'
>>           exit 1
>>   fi
>>
>>   if [ $# -gt 0 ]; then
>>           x=3D
>>           case $1 in
>>           lo) x=3D$lo;;
>>           med) x=3D$med;;
>>           hi) x=3D$hi;;
>>           default) x=3D$default;;
>>           75) x=3D$x75;;
>>           *) echo >&2 'Synopsis: cpupower [lo|med|hi|default[|75]]';;
>>           esac
>>           [ -n "$x" ] && cpupower frequency-set -u $x
>>   fi
>>
>>   cpupower frequency-info
>>
>> and that /root/hosts/$HOSTNAME/cpupower is (here)
>>
>>   #@ /root/hosts/self/cpupower
>>   lo=3D'400M -g powersave'
>>   med=3D'1600M -g powersave'
>>   hi=3D'3400M -g performance'
>>   default=3D'3400M -g powersave'
>>   x75=3D'2500M -g performance'
>>
>> (i use "hi" only for kernel and port builds).
>> It makes "a hell of a difference".
>>
>> Other than that i have seen similar things because (a) the
>> kernel's crng came up slowly (up to 30 minutes at times), but this
>> *should* be a thing of the past (dunno Alpine 3.20, i always use
>> [edge]), or because of WiFi chip powersafe stuff, i usually do
>> that though, i use "iw dev $dev set power_save on":
>>
>>   # iw dev wlp1s0 get power_save
>>   Power save: on
>>
>> Other than that i am out of ideas.
>>
>> --steffen
>> |
>> |Der Kragenbaer,                The moon bear,
>> |der holt sich munter           he cheerfully and one by one
>> |einen nach dem anderen runter  wa.ks himself off
>> |(By Robert Gernhardt)
>>
>

--0000000000002b71d1062369c865
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr">Ok I found the culprit. It was a network issue. The proble=
m was happening when initiating the connection from a wlan which is bridged=
 with the wired lan.<div>For some reason,=C2=A0my Internet router forgets t=
he presence of the RPI. Therefore IP address resolution can be slow. I supp=
ose it&#39;s because there is no traffic coming from the RPI.</div><div>I i=
nstalled the avahi daemon, I think it will make the RPI visible continuousl=
y.</div><div><br></div></div><br><div class=3D"gmail_quote"><div dir=3D"ltr=
" class=3D"gmail_attr">Le=C2=A0lun. 30 sept. 2024 =C3=A0=C2=A023:20, Jerome=
 Marc &lt;<a href=3D"mailto:marcjero@yahoo.com">marcjero@yahoo.com</a>&gt; =
a =C3=A9crit=C2=A0:<br></div><blockquote class=3D"gmail_quote" style=3D"mar=
gin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1=
ex"><div dir=3D"ltr">Hi Steffen,<div><br><div>Thanks !</div><div>Unfortunat=
ely there is cpupower rc service running and I don&#39;t have wifi.</div></=
div><div>I installed cpupower and got this output=C2=A0:</div><div><br></di=
v><div>alpine:~# cpupower frequency-info<br>analyzing CPU 0:<br>=C2=A0 driv=
er: cpufreq-dt<br>=C2=A0 CPUs which run at the same hardware frequency: 0 1=
 2 3<br>=C2=A0 CPUs which need to have their frequency coordinated by softw=
are: 0 1 2 3<br>=C2=A0 maximum transition latency: =C2=A0Cannot determine o=
r is not supported.<br>=C2=A0 hardware limits: 600 MHz - 900 MHz<br>=C2=A0 =
available frequency steps: =C2=A0600 MHz, 700 MHz, 800 MHz, 900 MHz<br>=C2=
=A0 available cpufreq governors: conservative ondemand userspace performanc=
e schedutil<br>=C2=A0 current policy: frequency should be within 600 MHz an=
d 900 MHz.<br>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=
=A0 The governor &quot;schedutil&quot; may decide which speed to use<br>=C2=
=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 within this ran=
ge.<br>=C2=A0 current CPU frequency: 600 MHz (asserted by call to hardware)=
<br></div><div><br></div><div>Looks good...</div></div><br><div class=3D"gm=
ail_quote"><div dir=3D"ltr" class=3D"gmail_attr">Le=C2=A0lun. 30 sept. 2024=
 =C3=A0=C2=A022:27, Steffen Nurpmeso &lt;<a href=3D"mailto:steffen@sdaoden.=
eu" target=3D"_blank">steffen@sdaoden.eu</a>&gt; a =C3=A9crit=C2=A0:<br></d=
iv><blockquote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;bord=
er-left:1px solid rgb(204,204,204);padding-left:1ex">Jerome Marc wrote in<b=
r>
=C2=A0&lt;<a href=3D"mailto:CAD1Ag0wW0nw-BR1aqKkD66LLO_yndJAy__i6yg9TWvU_Z6=
QEDA@mail.gmail.com" target=3D"_blank">CAD1Ag0wW0nw-BR1aqKkD66LLO_yndJAy__i=
6yg9TWvU_Z6QEDA@mail.gmail.com</a>&gt;:<br>
=C2=A0|I just installed Alpine 3.20.3 on a RPI 2 in diskless mode.(headless=
,<br>
=C2=A0|minimal installation)<br>
=C2=A0|<br>
=C2=A0|I noticed that if I let the device idling, I have a few seconds late=
ncy<br>
=C2=A0|when I try to connect the machine with SSH for example. Even when co=
nnected<br>
=C2=A0|I can have my SSH session frozen for a few seconds if I do nothing i=
n the<br>
=C2=A0|terminal.<br>
=C2=A0|I notice the same when using a nodejs app (HTTP) therefore this is a=
<br>
=C2=A0|general thing (not just about SSH)<br>
=C2=A0|<br>
=C2=A0|I don&#39;t see anything weird in logs. Any idea ? Is that a power s=
aving<br>
=C2=A0|feature ?<br>
<br>
It *could* be that cpupower-openrc is indeed the<br>
tools/power/cpupower of the linux kernel sources, it seems to be<br>
widely available (architecture-wise, on Alpine).<br>
If that is so, then running it should output stuff like<br>
<br>
=C2=A0 #?0|kent:~# bin/cpupower.sh<br>
=C2=A0 analyzing CPU 0:<br>
=C2=A0 =C2=A0 driver: intel_pstate<br>
=C2=A0 =C2=A0 CPUs which run at the same hardware frequency: 0<br>
=C2=A0 =C2=A0 CPUs which need to have their frequency coordinated by softwa=
re: 0<br>
=C2=A0 =C2=A0 maximum transition latency:=C2=A0 Cannot determine or is not =
supported.<br>
=C2=A0 =C2=A0 hardware limits: 400 MHz - 3.40 GHz<br>
=C2=A0 =C2=A0 available cpufreq governors: performance powersave<br>
=C2=A0 =C2=A0 current policy: frequency should be within 400 MHz and 3.40 G=
Hz.<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 The g=
overnor &quot;powersave&quot; may decide which speed to use<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 withi=
n this range.<br>
=C2=A0 =C2=A0 current CPU frequency: Unable to call hardware<br>
=C2=A0 =C2=A0 current CPU frequency: 800 MHz (asserted by call to kernel)<b=
r>
=C2=A0 =C2=A0 boost state support:<br>
=C2=A0 =C2=A0 =C2=A0 Supported: yes<br>
=C2=A0 =C2=A0 =C2=A0 Active: yes<br>
<br>
On the reserve laptop however the EFI(BIOS) &quot;always&quot; sets back<br=
>
cpupower, and we start with &quot;powersave&quot; governor and low<br>
frequency, and so the initial decrypting of the hard disk takes<br>
seeeeconds.=C2=A0 I btw drive it via<br>
<br>
=C2=A0 #!/bin/sh -<br>
=C2=A0 #@ /root/bin/cpupower.sh<br>
=C2=A0 # cpupower is in Linux src, tools/power/cpupower<br>
<br>
=C2=A0 : ${HOSTNAME:=3D$(uname -n)}<br>
<br>
=C2=A0 if [ -f /root/hosts/${HOSTNAME}/cpupower ]; then<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 . /root/hosts/${HOSTNAME}/cpupower<br>
=C2=A0 else<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 logger -s -t /root/bin/cpupower.sh &quot=
;MISS /root/hosts/${HOSTNAME}/cpupower&quot;<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 exit 1<br>
=C2=A0 fi<br>
<br>
=C2=A0 if command -v cpupower &gt;/dev/null 2&gt;&amp;1; then :; else<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 logger -s -t /root/bin/cpupower.sh &#39;=
no cpupower tool&#39;<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 exit 1<br>
=C2=A0 fi<br>
<br>
=C2=A0 if [ $# -gt 0 ]; then<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 x=3D<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 case $1 in<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 lo) x=3D$lo;;<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 med) x=3D$med;;<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 hi) x=3D$hi;;<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 default) x=3D$default;;<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 75) x=3D$x75;;<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 *) echo &gt;&amp;2 &#39;Synopsis: cpupow=
er [lo|med|hi|default[|75]]&#39;;;<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 esac<br>
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 [ -n &quot;$x&quot; ] &amp;&amp; cpupowe=
r frequency-set -u $x<br>
=C2=A0 fi<br>
<br>
=C2=A0 cpupower frequency-info<br>
<br>
and that /root/hosts/$HOSTNAME/cpupower is (here)<br>
<br>
=C2=A0 #@ /root/hosts/self/cpupower<br>
=C2=A0 lo=3D&#39;400M -g powersave&#39;<br>
=C2=A0 med=3D&#39;1600M -g powersave&#39;<br>
=C2=A0 hi=3D&#39;3400M -g performance&#39;<br>
=C2=A0 default=3D&#39;3400M -g powersave&#39;<br>
=C2=A0 x75=3D&#39;2500M -g performance&#39;<br>
<br>
(i use &quot;hi&quot; only for kernel and port builds).<br>
It makes &quot;a hell of a difference&quot;.<br>
<br>
Other than that i have seen similar things because (a) the<br>
kernel&#39;s crng came up slowly (up to 30 minutes at times), but this<br>
*should* be a thing of the past (dunno Alpine 3.20, i always use<br>
[edge]), or because of WiFi chip powersafe stuff, i usually do<br>
that though, i use &quot;iw dev $dev set power_save on&quot;:<br>
<br>
=C2=A0 # iw dev wlp1s0 get power_save<br>
=C2=A0 Power save: on<br>
<br>
Other than that i am out of ideas.<br>
<br>
--steffen<br>
|<br>
|Der Kragenbaer,=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 The=
 moon bear,<br>
|der holt sich munter=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0he cheerfully=
 and one by one<br>
|einen nach dem anderen runter=C2=A0 wa.ks himself off<br>
|(By Robert Gernhardt)<br>
</blockquote></div>
</blockquote></div>

--0000000000002b71d1062369c865--