Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.sys.raspberry-pi > #38307 > unrolled thread

Are Pi Zeros more fussy about SD cards than other Pis?

Back to article view | Back to comp.sys.raspberry-pi

Contents

  Are Pi Zeros more fussy about SD cards than other Pis? Chris Green <cl@isbd.net> - 2026-08-03 15:45 +0100
    Re: Are Pi Zeros more fussy about SD cards than other Pis? John R Walliker <jrwalliker@gmail.com> - 2026-08-03 18:05 +0100
      Re: Are Pi Zeros more fussy about SD cards than other Pis? Theo <theom+news@chiark.greenend.org.uk> - 2026-08-03 18:23 +0100
      Re: Are Pi Zeros more fussy about SD cards than other Pis? "s|b" <me@privacy.invalid> - 2026-08-03 22:48 +0200
    Re: Are Pi Zeros more fussy about SD cards than other Pis? The Natural Philosopher <tnp@invalid.invalid> - 2026-08-03 18:29 +0100
      Re: Are Pi Zeros more fussy about SD cards than other Pis? "s|b" <me@privacy.invalid> - 2026-08-03 22:48 +0200
        Re: Are Pi Zeros more fussy about SD cards than other Pis? Chris Green <cl@isbd.net> - 2026-08-04 10:15 +0100
          Re: Are Pi Zeros more fussy about SD cards than other Pis? The Natural Philosopher <tnp@invalid.invalid> - 2026-08-04 11:10 +0100
          Re: Are Pi Zeros more fussy about SD cards than other Pis? Ian <${send-direct-email-to-news1021-at-jusme-dot-com-if-you-must}@jusme.com> - 2026-08-04 11:23 +0000
            Re: Are Pi Zeros more fussy about SD cards than other Pis? Chris Green <cl@isbd.net> - 2026-08-04 17:53 +0100
              Re: Are Pi Zeros more fussy about SD cards than other Pis? Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-08-05 00:46 +0000
              Re: Are Pi Zeros more fussy about SD cards than other Pis? Ian <${send-direct-email-to-news1021-at-jusme-dot-com-if-you-must}@jusme.com> - 2026-08-05 08:30 +0000
                Re: Are Pi Zeros more fussy about SD cards than other Pis? The Natural Philosopher <tnp@invalid.invalid> - 2026-08-05 11:37 +0100
                Re: Are Pi Zeros more fussy about SD cards than other Pis? Chris Green <cl@isbd.net> - 2026-08-05 12:45 +0100
                  Re: Are Pi Zeros more fussy about SD cards than other Pis? The Natural Philosopher <tnp@invalid.invalid> - 2026-08-05 13:22 +0100
          Re: Are Pi Zeros more fussy about SD cards than other Pis? "s|b" <me@privacy.invalid> - 2026-08-04 14:56 +0200
            Re: Are Pi Zeros more fussy about SD cards than other Pis? [OT] "s|b" <me@privacy.invalid> - 2026-08-04 15:09 +0200
              Re: Are Pi Zeros more fussy about SD cards than other Pis? [OT] The Natural Philosopher <tnp@invalid.invalid> - 2026-08-04 15:11 +0100
                Re: Are Pi Zeros more fussy about SD cards than other Pis? [OT] "s|b" <me@privacy.invalid> - 2026-08-05 20:25 +0200
              Re: Are Pi Zeros more fussy about SD cards than other Pis? [OT] Chris Green <cl@isbd.net> - 2026-08-04 17:55 +0100
                Re: Are Pi Zeros more fussy about SD cards than other Pis? [OT] Andy Burns <usenet@andyburns.uk> - 2026-08-04 18:08 +0100
                Re: Are Pi Zeros more fussy about SD cards than other Pis? [OT] "s|b" <me@privacy.invalid> - 2026-08-05 19:50 +0200
            Re: Are Pi Zeros more fussy about SD cards than other Pis? The Natural Philosopher <tnp@invalid.invalid> - 2026-08-04 15:10 +0100
          Re: Are Pi Zeros more fussy about SD cards than other Pis? druck <news@druck.org.uk> - 2026-08-08 00:29 +0100
            Re: Are Pi Zeros more fussy about SD cards than other Pis? Chris Green <cl@isbd.net> - 2026-08-08 08:59 +0100
              Re: Are Pi Zeros more fussy about SD cards than other Pis? The Natural Philosopher <tnp@invalid.invalid> - 2026-08-08 12:27 +0100
              Re: Are Pi Zeros more fussy about SD cards than other Pis? druck <news@druck.org.uk> - 2026-08-10 19:06 +0100
                Re: Are Pi Zeros more fussy about SD cards than other Pis? The Natural Philosopher <tnp@invalid.invalid> - 2026-08-10 19:21 +0100
                Re: Are Pi Zeros more fussy about SD cards than other Pis? Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-08-11 00:03 +0000
                  Re: Are Pi Zeros more fussy about SD cards than other Pis? Jim Jackson <jj@franjam.org.uk> - 2026-08-11 07:33 +0000
                    Re: Are Pi Zeros more fussy about SD cards than other Pis? Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-08-12 01:25 +0000
                      Re: Are Pi Zeros more fussy about SD cards than other Pis? Jim Jackson <jj@franjam.org.uk> - 2026-08-13 16:55 +0000
                        Re: Are Pi Zeros more fussy about SD cards than other Pis? Lawrence D’Oliveiro <ldo@nz.invalid> - 2026-08-13 23:29 +0000
            Re: Are Pi Zeros more fussy about SD cards than other Pis? The Natural Philosopher <tnp@invalid.invalid> - 2026-08-08 12:25 +0100
    Re: Are Pi Zeros more fussy about SD cards than other Pis? Stephan Elinghaus <seli@spamgourmet.com> - 2026-08-11 12:49 +0000

Page 2 of 2 — ← Prev page 1 [2]

#38325 — Re: Are Pi Zeros more fussy about SD cards than other Pis? [OT]

Chris Green wrote:

> I use both ES and NIN, via leafnode running on my desktop machine. NIN
> is stopping in September though isn't it?

That's *next* september ...

[toc] | [prev] | [next] | [standalone]

#38333 — Re: Are Pi Zeros more fussy about SD cards than other Pis? [OT]

On Tue, 4 Aug 2026 17:55:25 +0100, Chris Green wrote:

> I use both ES and NIN, via leafnode running on my desktop machine. NIN
> is stopping in September though isn't it?

<https://individual.net/>

| Discontinuation of the NetNews / Usenet service as of 30 September 2027

-- 
s|b

[toc] | [prev] | [next] | [standalone]

#38320

On 04/08/2026 13:56, s|b wrote:
> On Tue, 4 Aug 2026 10:15:29 +0100, Chris Green wrote:
> 
>> s|b <me@privacy.invalid> wrote:
>>
>>> This is helpful.
> 
>> What would be more helpful would be some guidance on how to implement
>> it! :-)
> 
> Exactly my point.
>   
>> E.g.  how do you stop systemd's logging from writing to 'hard' disk?
>> It would need a **lot** of reconfiguration before it would use a
>> reasonable amount of space such as is available on a RAM drive.
>>
>> Maybe there are versions of the OS that are optimised for minimal disk
>> writing, if so I'd like to know about them.
> 
> Maybe he's just 'not interested' in explaining.
> 
I already did

-- 
The difference bweteen a psychopath and a saint is that the psychpoath 
takes what he can and gives only what he must, but the saint gives 
everything he can and takes only what he needs.

[toc] | [prev] | [next] | [standalone]

#38372

On 04/08/2026 10:15, Chris Green wrote:
> E.g.  how do you stop systemd's logging from writing to 'hard' disk?
> It would need a **lot** of reconfiguration before it would use a
> reasonable amount of space such as is available on a RAM drive.

Here are the important lines of /etc/systemd/journald.conf

[Journal]
Storage=volatile
RuntimeMaxUse=32M
RuntimeMaxFileSize=8M
RuntimeMaxFiles=4
ForwardToConsole=no
ForwardToWall=no


That will log to RAM only, a maximum of 32M in up to 4 files, so it 
should automatically prune 25% when it gets full, but I'm often having 
to do it manually with:-

journalctl --vacuum-files 3

> Maybe there are versions of the OS that are optimised for minimal disk
> writing, if so I'd like to know about them.

It's the config that needs to be optimised.

---druck

[toc] | [prev] | [next] | [standalone]

#38375

druck <news@druck.org.uk> wrote:
> On 04/08/2026 10:15, Chris Green wrote:
> > E.g.  how do you stop systemd's logging from writing to 'hard' disk?
> > It would need a **lot** of reconfiguration before it would use a
> > reasonable amount of space such as is available on a RAM drive.
> 
> Here are the important lines of /etc/systemd/journald.conf
> 
> [Journal]
> Storage=volatile
> RuntimeMaxUse=32M
> RuntimeMaxFileSize=8M
> RuntimeMaxFiles=4
> ForwardToConsole=no
> ForwardToWall=no
> 
> 
> That will log to RAM only, a maximum of 32M in up to 4 files, so it 
> should automatically prune 25% when it gets full, but I'm often having 
> to do it manually with:-
> 
> journalctl --vacuum-files 3
> 
That last rather spoils things, I'm expecting the systems in question
to run for months (at least) without any intervention from me.

... and systemd's journalling isn't the only thing that writes, it was
just the first thing that I thought of (hence the 'E.g.').

-- 
Chris Green
·

[toc] | [prev] | [next] | [standalone]

#38378

On 08/08/2026 08:59, Chris Green wrote:
> druck <news@druck.org.uk> wrote:
>> On 04/08/2026 10:15, Chris Green wrote:
>>> E.g.  how do you stop systemd's logging from writing to 'hard' disk?
>>> It would need a **lot** of reconfiguration before it would use a
>>> reasonable amount of space such as is available on a RAM drive.
>>
>> Here are the important lines of /etc/systemd/journald.conf
>>
>> [Journal]
>> Storage=volatile
>> RuntimeMaxUse=32M
>> RuntimeMaxFileSize=8M
>> RuntimeMaxFiles=4
>> ForwardToConsole=no
>> ForwardToWall=no
>>
>>
>> That will log to RAM only, a maximum of 32M in up to 4 files, so it
>> should automatically prune 25% when it gets full, but I'm often having
>> to do it manually with:-
>>
>> journalctl --vacuum-files 3
>>
> That last rather spoils things, I'm expecting the systems in question
> to run for months (at least) without any intervention from me.
> 
I haven't done mine manually. Ever.

> ... and systemd's journalling isn't the only thing that writes, it was
> just the first thing that I thought of (hence the 'E.g.').
> 
Well that was where a ramdisk is handy to replace /var/log

There is a command that looks at real I/O calls to e.g. disk drives.  It 
should return zero ponce you achieve Nirvana...

-- 
“But what a weak barrier is truth when it stands in the way of an 
hypothesis!”

Mary Wollstonecraft

[toc] | [prev] | [next] | [standalone]

#38393

On 08/08/2026 08:59, Chris Green wrote:
> druck <news@druck.org.uk> wrote:
>> That will log to RAM only, a maximum of 32M in up to 4 files, so it
>> should automatically prune 25% when it gets full, but I'm often having
>> to do it manually with:-
>>
> That last rather spoils things, I'm expecting the systems in question
> to run for months (at least) without any intervention from me.
> 
> ... and systemd's journalling isn't the only thing that writes, it was
> just the first thing that I thought of (hence the 'E.g.').

Exactly why I wish systemd would FOAD!

---druck

[toc] | [prev] | [next] | [standalone]

#38394

On 10/08/2026 19:06, druck wrote:
> On 08/08/2026 08:59, Chris Green wrote:
>> druck <news@druck.org.uk> wrote:
>>> That will log to RAM only, a maximum of 32M in up to 4 files, so it
>>> should automatically prune 25% when it gets full, but I'm often having
>>> to do it manually with:-
>>>
>> That last rather spoils things, I'm expecting the systems in question
>> to run for months (at least) without any intervention from me.
>>
>> ... and systemd's journalling isn't the only thing that writes, it was
>> just the first thing that I thought of (hence the 'E.g.').
> 
> Exactly why I wish systemd would FOAD!
> 
> ---druck
> 

Trust me, it can be done
systenmd does NOT carry on logging. My system is up months at a time and 
I never need to manually shrink it

Everything else is a matter of picking through anything that logss an 
either diverting it to a RAM disk at configuration level, or symlinking 
is entry in /var/log to a ramdisk
It takes time and the thing you may well lack, patience.


-- 
Labour - a bunch of rich people convincing poor people to vote for rich 
people by telling poor people that "other" rich people are the reason 
they are poor.

Peter Thompson

[toc] | [prev] | [next] | [standalone]

#38395

On Mon, 10 Aug 2026 19:06:57 +0100, druck wrote:

> On 08/08/2026 08:59, Chris Green wrote:
>>
>> ... and systemd's journalling isn't the only thing that writes, it
>> was just the first thing that I thought of (hence the 'E.g.').
>
> Exactly why I wish systemd would FOAD!

Ironically, it’s the one thing you can configure to work exactly the
way you want.

[toc] | [prev] | [next] | [standalone]

#38396

On 2026-08-11, Lawrence D?Oliveiro <ldo@nz.invalid> wrote:
> On Mon, 10 Aug 2026 19:06:57 +0100, druck wrote:
>
>> On 08/08/2026 08:59, Chris Green wrote:
>>>
>>> ... and systemd's journalling isn't the only thing that writes, it
>>> was just the first thing that I thought of (hence the 'E.g.').
>>
>> Exactly why I wish systemd would FOAD!
>
> Ironically, it???s the one thing you can configure to work exactly the
> way you want.

As long as its the systemd way :-)

Actually linux has always been very configurable the way you want, 
probably too configurable, because the plethora of options/alternatives 
etc can put people off.

[toc] | [prev] | [next] | [standalone]

#38398

On Tue, 11 Aug 2026 07:33:18 -0000 (UTC), Jim Jackson wrote:

> On Tue, 11 Aug 2026 00:03:39 -0000 (UTC), Lawrence D’Oliveiro wrote:
>
>> Ironically, it’s the one thing you can configure to work exactly
>> the way you want.
>
> As long as its the systemd way :-)

The “systemd way” lets you send the logging wherever you want. Want it
all to go to rsyslog? Send it there. Then you can blame rsyslog for
eating your SD-card, not systemd.

[toc] | [prev] | [next] | [standalone]

#38406

On 2026-08-12, Lawrence D?Oliveiro <ldo@nz.invalid> wrote:
> On Tue, 11 Aug 2026 07:33:18 -0000 (UTC), Jim Jackson wrote:
>
>> On Tue, 11 Aug 2026 00:03:39 -0000 (UTC), Lawrence D???Oliveiro wrote:
>>
>>> Ironically, it???s the one thing you can configure to work exactly
>>> the way you want.
>>
>> As long as its the systemd way :-)
>
> The ???systemd way??? lets you send the logging wherever you want.

Oh dear you took it waaayyy too seriously! I never said it couldn't! I 
merely meant that, in the bits you edited out, that you could even 
Before Systemd "configure [it] to work exactly the way you want". But 
obviously I touched one of your buttons, as you did mine :-)


And to be more on topic, I'd suggest to the original poster, that having 
a local syslog log server is useful.  Configure your RPi machine to send 
the log entries to that. No SD writes and the logs are there even if the 
SD card gets corrupt for other reasons.

[toc] | [prev] | [next] | [standalone]

#38408

On Thu, 13 Aug 2026 16:55:03 -0000 (UTC), Jim Jackson wrote:

> I merely meant that, in the bits you edited out, that you could even
> Before Systemd "configure [it] to work exactly the way you want".

As I recall, the original complaint was about logfiles shortening the
life of SD cards. How would this have helped?

[toc] | [prev] | [next] | [standalone]

#38377

On 08/08/2026 00:29, druck wrote:
> On 04/08/2026 10:15, Chris Green wrote:
>> E.g.  how do you stop systemd's logging from writing to 'hard' disk?
>> It would need a **lot** of reconfiguration before it would use a
>> reasonable amount of space such as is available on a RAM drive.
> 
> Here are the important lines of /etc/systemd/journald.conf
> 
> [Journal]
> Storage=volatile
> RuntimeMaxUse=32M
> RuntimeMaxFileSize=8M
> RuntimeMaxFiles=4
> ForwardToConsole=no
> ForwardToWall=no
> 
> 
> That will log to RAM only, a maximum of 32M in up to 4 files, so it 
> should automatically prune 25% when it gets full, but I'm often having 
> to do it manually with:-
> 
> journalctl --vacuum-files 3
> 
>> Maybe there are versions of the OS that are optimised for minimal disk
>> writing, if so I'd like to know about them.
> 
> It's the config that needs to be optimised.
> 
> ---druck
> 
> 
+1. And this actually works

-- 
“But what a weak barrier is truth when it stands in the way of an 
hypothesis!”

Mary Wollstonecraft

[toc] | [prev] | [next] | [standalone]

#38397

Chris Green <cl@isbd.net> schrieb:
> I'd welcome any recommendations for cards that last well.

Since I began using industrial grade cards (e.g. Kingston industrial) I
never had any issues with a defective card any more. Those cards may
not be the fastest, but they have proven to be very reliable.

-- 
[place signature here]

[toc] | [prev] | [standalone]

Page 2 of 2 — ← Prev page 1 [2]

Back to top | Article view | comp.sys.raspberry-pi

csiph-web