I've bought a new disk (HDD, WD purple) but instead of
replacing the old, I disconnected internal DVD recorder and
I moved the old one there (I had no spare "molex" nor esata
channels left).
Now I have found a strange thing : the timing of the facts
is too suspicious to be decorrelated : my USB alcatel model
fails to be charged at a pace fast enough to keep up with
battery consumption and after 1-2 hours of intense use,
switches off.

I am thinking the power (we call it "alimentatore" but dunno
the English word ... the power source, let's say) is
insufficient to cover up all devices properly and fails to
send enough current to external devices (the HDD behaves
normally)

now the true question (The drive currently is mounted via FSTAB)

would unmount the drive cause it to stop spinning ?

There is a faster way to sendo some low level command to
make it stop spinning ? I.g. some "IOCTL" sequence.

the /dev/sde1 interface for the block device can receive
IOCTLs codes (assuming I'd be able to find proper commands
to send to the controller).

The drive will be seldom used (I don't have a big enough
extern box to build an USB unit)

I'd largely prefer to send direct commands to stop spinning
and park the heads to unmounting (assumed unmounting will
cut power consumption, thing that I dunno, actually), since
later the drive would still be responding requests,
automatically restarting spinning on receiving requests, I
guess.

for those really knowledgeable in internal Linux working :
does a drive not actively used (no files opened, no file
manager operating on it) require some form of "polling" its
state that would cause frequent automatical restartig of
spinning ? In this case, stopping it would be harmful and
enhance preature wearing of the motors and all and it would
have no benefit to manually stop spinning.

I ask this latter because I have an INTENSO 4 TB external
drive (a monster of speed, it reaches 160 MB/s !) that on
another machine (always debian / MX) frequently stops and
later at some time for no apparent reason at all restarts
spinning (there the blue led makes it further evident beyond
the ears can probe). Now surely it could be the internal
"controller" aggressive power saving strategies, independent
on the OS, to switch the drive off even if no "warranty" of
no further requests was given by the OS (unmounting for
examople).
But that is an USB external device and such strategies seem
reasonable.

I dunno if a drive connected on esata and molex is allowed
to decide alike (or if the OS can / must somehow save power
if the disk is not in active use).

If sb knows about this topic, pls share opinions.

tnx

-- CCCPMario or MarioCCCP

Am 29.09.2023 schrieb Fonntuggnio <JoeFonntuggnio@libbbero.it>:

> I've bought a new disk (HDD, WD purple) but instead of
> replacing the old, I disconnected internal DVD recorder and
> I moved the old one there (I had no spare "molex" nor esata
> channels left).
> Now I have found a strange thing : the timing of the facts
> is too suspicious to be decorrelated : my USB alcatel model
> fails to be charged at a pace fast enough to keep up with
> battery consumption and after 1-2 hours of intense use,
> switches off.

Seems to be an USB problem. A bus has limited amperage.

> would unmount the drive cause it to stop spinning ?

Not directly. Even when it is mounted it can be spinned down and
spinned up again when needed.

> There is a faster way to sendo some low level command to
> make it stop spinning ? I.g. some "IOCTL" sequence.

hdparm
https://superuser.com/questions/852727/safe-to-manually-spin-down-hard-drive-with-hdparm

> for those really knowledgeable in internal Linux working :
> does a drive not actively used (no files opened, no file
> manager operating on it) require some form of "polling" its
> state that would cause frequent automatical restartig of
> spinning ? In this case, stopping it would be harmful and
> enhance preature wearing of the motors and all and it would
> have no benefit to manually stop spinning.

Every read request (not in cache) will cause it to spin up.
Write request can be cached and will be synced at unmount (can be
invoked by "sync").

On 29/09/23 10:56, Marco Moock wrote:
> Am 29.09.2023 schrieb Fonntuggnio <JoeFonntuggnio@libbbero.it>:
>
>> I've bought a new disk (HDD, WD purple) but instead of
>> replacing the old, I disconnected internal DVD recorder and
>> I moved the old one there (I had no spare "molex" nor esata
>> channels left).
>> Now I have found a strange thing : the timing of the facts
>> is too suspicious to be decorrelated : my USB alcatel model
>> fails to be charged at a pace fast enough to keep up with
>> battery consumption and after 1-2 hours of intense use,
>> switches off.
>
> Seems to be an USB problem. A bus has limited amperage.
>
>> would unmount the drive cause it to stop spinning ?
>
> Not directly. Even when it is mounted it can be spinned down and
> spinned up again when needed.
>
>> There is a faster way to sendo some low level command to
>> make it stop spinning ? I.g. some "IOCTL" sequence.
>
> hdparm
> https://superuser.com/questions/852727/safe-to-manually-spin-down-hard-drive-with-hdparm

tnx for this very prompt reply ! I'll study the case

>
>> for those really knowledgeable in internal Linux working :
>> does a drive not actively used (no files opened, no file
>> manager operating on it) require some form of "polling" its
>> state that would cause frequent automatical restartig of
>> spinning ? In this case, stopping it would be harmful and
>> enhance preature wearing of the motors and all and it would
>> have no benefit to manually stop spinning.
>
> Every read request (not in cache) will cause it to spin up.
> Write request can be cached and will be synced at unmount (can be
> invoked by "sync").
>

tnx !

--
1) Resistere, resistere, resistere.
2) Se tutti pagano le tasse, le tasse le pagano tutti
MarioCPPP

On 29/09/23 12:12, MarioCCCP wrote:
> On 29/09/23 10:56, Marco Moock wrote:
>> Am 29.09.2023 schrieb Fonntuggnio
>> <JoeFonntuggnio@libbbero.it>:
>>
>>> I've bought a new disk (HDD, WD purple) but instead of
>>> replacing the old, I disconnected internal DVD recorder and
>>> I moved the old one there (I had no spare "molex" nor esata
>>> channels left).
>>> Now I have found a strange thing : the timing of the facts
>>> is too suspicious to be decorrelated : my USB alcatel model
>>> fails to be charged at a pace fast enough to keep up with
>>> battery consumption and after 1-2 hours of intense use,
>>> switches off.
>>
>> Seems to be an USB problem. A bus has limited amperage.
>>
>>> would unmount the drive cause it to stop spinning ?
>>
>> Not directly. Even when it is mounted it can be spinned
>> down and
>> spinned up again when needed.
>>
>>> There is a faster way to sendo some low level command to
>>> make it stop spinning ? I.g. some "IOCTL" sequence.
>>
>> hdparm
>> https://superuser.com/questions/852727/safe-to-manually-spin-down-hard-drive-with-hdparm
>
> tnx for this very prompt reply ! I'll study the case
>
>>
>>> for those really knowledgeable in internal Linux working :
>>> does a drive not actively used (no files opened, no file
>>> manager operating on it) require some form of "polling" its
>>> state that would cause frequent automatical restartig of
>>> spinning ? In this case, stopping it would be harmful and
>>> enhance preature wearing of the motors and all and it would
>>> have no benefit to manually stop spinning.
>>
>> Every read request (not in cache) will cause it to spin up.
>> Write request can be cached and will be synced at unmount
>> (can be
>> invoked by "sync").
>>
>
>
> tnx !
>

in the replies of the SuperUser thread you signalled, I
found the HD-IDLE utility that does exactly what I required
!!! Wonderful Linux

--
1) Resistere, resistere, resistere.
2) Se tutti pagano le tasse, le tasse le pagano tutti
MarioCPPP

On 2023-09-29 10:43, Fonntuggnio wrote:
>
> I've bought a new disk (HDD, WD purple) but instead of replacing the
> old, I disconnected internal DVD recorder and I moved the old one there
> (I had no spare "molex" nor esata channels left).
> Now I have found a strange thing : the timing of the facts is too
> suspicious to be decorrelated : my USB alcatel model fails to be charged
> at a pace fast enough to keep up with battery consumption and after 1-2
> hours of intense use, switches off.
>
> I am thinking the power (we call it "alimentatore" but dunno the English
> word ... the power source, let's say) is insufficient to cover up all
> devices properly and fails to send enough current to external devices
> (the HDD behaves normally)
>
> now the true question (The drive currently is mounted via FSTAB)
>
> would unmount the drive cause it to stop spinning ?

Not inmediately, there is a timeout.

hdparm - get/set SATA/IDE device parameters

-C Check the current IDE power mode status, which will
always be one of unknown (drive does not support
this command), active/idle (normal operation),
standby (low power mode, drive has spun down), or
sleeping (lowest power mode, drive is completely
shut down). The -S, -y, -Y, and -Z options can be
used to manipulate the IDE power modes.

-s Enable/disable the power-on in standby feature, if
supported by the drive. VERY DANGEROUS. Do not
use unless you are absolutely certain that both the
system BIOS (or firmware) and the operating system
kernel (Linux >= 2.6.22) support probing for drives
that use this feature. When enabled, the drive is
powered-up in the standby mode to allow the con-
troller to sequence the spin-up of devices, reduc-
ing the instantaneous current draw burden when many
drives share a power supply. Primarily for use in
large RAID setups. This feature is usually dis-
abled and the drive is powered-up in the active
mode (see -C above). Note that a drive may also
allow enabling this feature by a jumper. Some SATA
drives support the control of this feature by pin
11 of the SATA power connector. In these cases,
this command may be unsupported or may have no ef-
fect.

-S Put the drive into idle (low-power) mode, and also
set the standby (spindown) timeout for the drive.
This timeout value is used by the drive to deter-
mine how long to wait (with no disk activity) be-
fore turning off the spindle motor to save power.
Under such circumstances, the drive may take as
long as 30 seconds to respond to a subsequent disk
access, though most drives are much quicker. The
encoding of the timeout value is somewhat peculiar.
A value of zero means "timeouts are disabled": the
device will not automatically enter standby mode.
Values from 1 to 240 specify multiples of 5 sec-
onds, yielding timeouts from 5 seconds to 20 min-
utes. Values from 241 to 251 specify from 1 to 11
units of 30 minutes, yielding timeouts from 30 min-
utes to 5.5 hours. A value of 252 signifies a
timeout of 21 minutes. A value of 253 sets a ven-
dor-defined timeout period between 8 and 12 hours,
and the value 254 is reserved. 255 is interpreted
as 21 minutes plus 15 seconds. Note that some
older drives may have very different interpreta-
tions of these values.

>
> There is a faster way to sendo some low level command to make it stop
> spinning ? I.g. some "IOCTL" sequence.

The "-S" option above.

> for those really knowledgeable in internal Linux working : does a drive
> not actively used (no files opened, no file manager operating on it)
> require some form of "polling" its state that would cause frequent
> automatical restartig of spinning ? In this case, stopping it would be
> harmful and enhance preature wearing of the motors and all and it would
> have no benefit to manually stop spinning.

There can be operations that trigger periodically that would cause the
disk to spin up again. Typical culprits were syslog. I seem to remember
some 7 minutes timeout somewhere, just a foggy memory.

If the disk is umounted, then hardly. Well, a command such as "lsblk"
could wake up all disks, because it probes them all. Smartctl... I think
there is an option telling it to wake or not wake a sleeping disk.

--
Cheers,
Carlos E.R.

On 29/09/23 14:38, Carlos E. R. wrote:
> On 2023-09-29 10:43, Fonntuggnio wrote:
>>
>> I've bought a new disk (HDD, WD purple) but instead of
>> replacing the old, I disconnected internal DVD recorder
>> and I moved the old one there (I had no spare "molex" nor
>> esata channels left).
>> Now I have found a strange thing : the timing of the facts
>> is too suspicious to be decorrelated : my USB alcatel
>> model fails to be charged at a pace fast enough to keep up
>> with battery consumption and after 1-2 hours of intense
>> use, switches off.
>>
>> I am thinking the power (we call it "alimentatore" but
>> dunno the English word ... the power source, let's say) is
>> insufficient to cover up all devices properly and fails to
>> send enough current to external devices (the HDD behaves
>> normally)
>>
>> now the true question (The drive currently is mounted via
>> FSTAB)
>>
>> would unmount the drive cause it to stop spinning ?
>
> Not inmediately, there is a timeout.
>
>        hdparm - get/set SATA/IDE device parameters
>
>        -C     Check the current IDE power mode status,
> which will
>               always  be  one  of unknown (drive does not
> support
>               this  command),  active/idle  (normal
> operation),
>               standby  (low  power mode, drive has spun
> down), or
>               sleeping (lowest power mode,  drive  is
> completely
>               shut  down).  The -S, -y, -Y, and -Z options
> can be
>               used to manipulate the IDE power modes.
>
>        -s     Enable/disable the power-on in standby
> feature,  if
>               supported  by  the  drive.  VERY DANGEROUS.
> Do not
>               use unless you are absolutely certain that
> both the
>               system  BIOS (or firmware) and the operating
> system
>               kernel (Linux >= 2.6.22) support probing for
> drives
>               that  use this feature.  When enabled, the
> drive is
>               powered-up in the standby mode to  allow
> the  con-
>               troller  to sequence the spin-up of devices,
> reduc-
>               ing the instantaneous current draw burden
> when many
>               drives  share a power supply.  Primarily for
> use in
>               large RAID setups.  This feature  is
> usually  dis-
>               abled  and  the  drive  is powered-up in the
> active
>               mode (see -C above).  Note that a  drive
> may  also
>               allow enabling this feature by a jumper.
> Some SATA
>               drives support the control of this feature
> by  pin
>               11  of  the  SATA  power connector. In these
> cases,
>               this command may be unsupported or may have
> no  ef-
>               fect.
>
>        -S     Put  the drive into idle (low-power) mode,
> and also
>               set the standby (spindown) timeout for  the
> drive.
>               This  timeout  value is used by the drive to
> deter-
>               mine how long to wait (with no disk
> activity)  be-
>               fore  turning  off the spindle motor to save
> power.
>               Under such circumstances, the  drive  may
> take  as
>               long  as 30 seconds to respond to a
> subsequent disk
>               access, though most drives are much
> quicker.   The
>               encoding of the timeout value is somewhat
> peculiar.
>               A value of zero means "timeouts are
> disabled":  the
>               device  will  not automatically enter standby
> mode.
>               Values from 1 to 240 specify multiples  of
> 5  sec-
>               onds,  yielding  timeouts from 5 seconds to
> 20 min-
>               utes.  Values from 241 to 251 specify from 1
> to  11
>               units of 30 minutes, yielding timeouts from
> 30 min-
>               utes to 5.5 hours.  A  value  of  252
> signifies  a
>               timeout  of  21 minutes. A value of 253 sets
> a ven-
>               dor-defined timeout period between 8 and 12
> hours,
>               and  the value 254 is reserved.  255 is
> interpreted
>               as 21 minutes plus  15  seconds.   Note
> that  some
>               older  drives  may  have very different
> interpreta-
>               tions of these values.
>
>
>>
>> There is a faster way to sendo some low level command to
>> make it stop spinning ? I.g. some "IOCTL" sequence.
>
> The "-S" option above.
>
>
>> for those really knowledgeable in internal Linux working :
>> does a drive not actively used (no files opened, no file
>> manager operating on it) require some form of "polling"
>> its state that would cause frequent automatical restartig
>> of spinning ? In this case, stopping it would be harmful
>> and enhance preature wearing of the motors and all and it
>> would have no benefit to manually stop spinning.
>
> There can be operations that trigger periodically that would
> cause the disk to spin up again. Typical culprits were
> syslog. I seem to remember some 7 minutes timeout somewhere,
> just a foggy memory.
>
> If the disk is umounted, then hardly. Well, a command such
> as "lsblk" could wake up all disks, because it probes them
> all. Smartctl... I think there is an option telling it to
> wake or not wake a sleeping disk.
>
tnx !
--
1) Resistere, resistere, resistere.
2) Se tutti pagano le tasse, le tasse le pagano tutti
MarioCPPP