"Carlos E.R." <robin_listas@es.invalid> wrote:
>On 2024-05-01 22:42, Kenny McCormack wrote:
>> In article <v0tkdc$3780t$1@dont-email.me>,
>> Lew Pitcher <lew.pitcher@digitalfreehold.ca> wrote:
>>> On Wed, 01 May 2024 11:32:18 +0100, James Harris wrote:
>>>
>>>> Not a question, just an observation.
>> ...
>>>
>>> So, if you want help, how about providing some details.
>>
>> OP was pretty clear that he was *not* seeking help.
>
>Because he is blaming Linux at large, when he has a problem in *his*
>system. He made a wrong diagnosis.

And a frequent one for crying out loud. Heck, there is an FAQ about
this very issue.

Memory management is always "too little, too slow", impossible to do
right for everybody.

Greetings
Marc
--
----------------------------------------------------------------------------
Marc Haber | " Questions are the | Mailadresse im Header
Rhein-Neckar, DE | Beginning of Wisdom " |
Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 6224 1600402

Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
>On Thu, 02 May 2024 07:47:19 +0200, Marc Haber wrote:
>> No mechanic ever had to appear in front of a board to justify the use of
>> spanner and screw driver, right?
>
>I’m sure they do, to whomever controls the purse strings at the garage.

Show me a single garage without spanner or screw driver.

Greetings
Marc
--
----------------------------------------------------------------------------
Marc Haber | " Questions are the | Mailadresse im Header
Rhein-Neckar, DE | Beginning of Wisdom " |
Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 6224 1600402

"Carlos E.R." <robin_listas@es.invalid> wrote:
>On 2024-05-02 09:35, The Natural Philosopher wrote:
>> On 02/05/2024 07:45, Lawrence D'Oliveiro wrote:
>>> On Thu, 02 May 2024 07:47:19 +0200, Marc Haber wrote:
>>>
>>>> No mechanic ever had to appear in front of a board to justify the use of
>>>> spanner and screw driver, right?
>>>
>>> I’m sure they do, to whomever controls the purse strings at the garage.
>>
>> Well indeed. Although not a board, just the boss.
>
>My mechanic boss has to query the board. It is a chain.
>Not for a spanner, but for expensive diagnostic tools.

htop or atop are not expensive. They're available in all major
distribution's repositories ready for installation.

Greetings
Marc, who is responsible for atop turning up in Debian's repositories
--
----------------------------------------------------------------------------
Marc Haber | " Questions are the | Mailadresse im Header
Rhein-Neckar, DE | Beginning of Wisdom " |
Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 6224 1600402

On 01/05/2024 11:46, Marc Haber wrote:
> James Harris <james.harris.1@gmail.com> wrote:
>> Not a question, just an observation.
>
> A freqently wrong one as well.
>
>> I say that Linux doesn't seem to handle memory well because my laptop
>> had 8GB RAM (which, frankly, Windows seems to find perfectly adequate
>> for a similar workload). Under Linux the RAM would fill up and then swap
>> space would be used. Then the machine would become largely unresponsive
>> - e.g. taking minutes to switch between windows.
>>
>> So I upgraded the RAM. It now has three times as much (i.e. 24GB)! But
>> even so, RAM has still steadily filled up until reaching the full 24GB.
>> What's more, it's now showing 4.8GB of swap space in use.
>
> Linux swaps out lesser used pages anyway to make the backing memory
> available for buffers or cache.
>
> Those people writing memory management are really smart. Smarter than
> you and me together. Don't assume them being stupid.

It may be unwise to assume there's such expertise. A number of years ago
I came up with what I thought was a sensible design for a paging system.
I later found that it corresponded very closely to the one used in
Windows and differed markedly from the paging system used in Linux. I
remember thinking that the latter was not a good design and since then
I've experienced problems with Linux when memory fills up which I tend
to assume are down to the design of its paging system - hence the,
admittedly provocative, title of this thread.

I can't remember much about it now and Linux may well have changed the
way it handles page depletion but my experience as a Linux user seems to
bear out my negative view of its page management.

Whatever the details, in defence of Linux it should be noted that the
Linux design may have to cope with a much broader range of hardware -
e.g. CPUs which have no accessed or dirty bits - whereas Windows was x86
focussed.

--
James Harris

On 01/05/2024 11:53, The Natural Philosopher wrote:
> On 01/05/2024 11:32, James Harris wrote:
>> Not a question, just an observation.
>>
>> I say that Linux doesn't seem to handle memory well because my laptop
>> had 8GB RAM (which, frankly, Windows seems to find perfectly adequate
>> for a similar workload). Under Linux the RAM would fill up and then
>> swap space would be used. Then the machine would become largely
>> unresponsive - e.g. taking minutes to switch between windows.
>>
>> So I upgraded the RAM. It now has three times as much (i.e. 24GB)! But
>> even so, RAM has still steadily filled up until reaching the full
>> 24GB. What's more, it's now showing 4.8GB of swap space in use.
>>
>>
> That's not Linux. That's a rogue program with a memory leak...

It may not be a memory leak but it's certainly a rogue program -
chrome/chromium in this case, though I've had similar and worse problems
with Firefox which seems to me more likely to have a genuine memory leak.

However, the point of the original post was to vent a frustration about
the way that Linux handles such memory deletion when it happens.

--
James Harris

On 01/05/2024 15:06, David W. Hodgins wrote:
> On Wed, 01 May 2024 06:32:18 -0400, James Harris
> <james.harris.1@gmail.com> wrote:
>> Not a question, just an observation.
>> I say that Linux doesn't seem to handle memory well because my laptop
>> had 8GB RAM (which, frankly, Windows seems to find perfectly adequate
>> for a similar workload). Under Linux the RAM would fill up and then swap
>> space would be used. Then the machine would become largely unresponsive
>> - e.g. taking minutes to switch between windows.
>> So I upgraded the RAM. It now has three times as much (i.e. 24GB)! But
>> even so, RAM has still steadily filled up until reaching the full 24GB.
>> What's more, it's now showing 4.8GB of swap space in use.
>
> Likely you're using a system that's configured for maximum total
> throughput,
> not best response time.
>
> Create a file, /etc/sysctl.d/tales.conf with the contents ...
> # Reduce applications being swapped
> vm.swappiness=1
> # Don't shrink the inode cache
> vm.vfs_cache_pressure=50
>
> Then run the command "sysctl --system" (as root).
>
> See
> https://rudd-o.com/linux-and-free-software/tales-from-responsivenessland-why-linux-feels-slow-and-how-to-fix-that
> for an explanation.

I very much like the control that Linux gives to a user but ISTM that
what you say here is a great example of the issue I was pointing out in
the original post: IMO page depletion should be handled better *by
default* and not need human intervention for normal scenarios.

--
James Harris

On 01/05/2024 21:42, Kenny McCormack wrote:
> In article <v0tkdc$3780t$1@dont-email.me>,
> Lew Pitcher <lew.pitcher@digitalfreehold.ca> wrote:
>> On Wed, 01 May 2024 11:32:18 +0100, James Harris wrote:
>>
>>> Not a question, just an observation.
> ...
>>
>> So, if you want help, how about providing some details.
>
> OP was pretty clear that he was *not* seeking help.

Correct. I don't want to overtax the expert assistance that's clearly
available from posters to this newsgroup and would rather ask for help
relatively rarely, so I am not looking (at least, yet!) for a solution.
My point, in this thread, was, instead, to comment on what I perceive to
be an OS-design issue - a long-term interest.

--
James Harris

On 02/05/2024 07:01, Carlos E.R. wrote:
> On 2024-05-01 22:42, Kenny McCormack wrote:
>> In article <v0tkdc$3780t$1@dont-email.me>,
>> Lew Pitcher  <lew.pitcher@digitalfreehold.ca> wrote:
>>> On Wed, 01 May 2024 11:32:18 +0100, James Harris wrote:
>>>
>>>> Not a question, just an observation.
>> ...
>>>
>>> So, if you want help, how about providing some details.
>>
>> OP was pretty clear that he was *not* seeking help.
>
> Because he is blaming Linux at large, when he has a problem in *his*
> system. He made a wrong diagnosis.

Yes, I am (at least tentatively) blaming Linux but not for the problem
you think. I wouldn't expect Linux to prevent programs from gobbling up
memory but I would expect it to manage memory hogs more gracefully than
it does. IME Windows handles the same situation better - and that may be
down to the different designs of their paging systems.

Don't get me wrong. I much prefer Linux to Windows and have often seen
Windows get into a worse situation under different circumstances. But
for page management ISTM that the design of Linux's paging subsystem may
not be the best.

--
James Harris

Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
> On Thu, 02 May 2024 07:47:19 +0200, Marc Haber wrote:
>
>> No mechanic ever had to appear in front of a board to justify the use of
>> spanner and screw driver, right?
>
> I’m sure they do, to whomever controls the purse strings at the garage.

At nearly all auto-shops, the basic tools (spanners/screwdrivers/etc.)
are purchased and owned by the actual mechanics using them and not the
operator of the garage. What the shop owner provides are the
"infrastructure" items (lifts, air compressor, some of the more
specialized tools, such as the tyre changer or the 'computer diagnostic
rack', etc.).

On 01/05/2024 15:53, Grant Taylor wrote:
> On 5/1/24 05:32, James Harris wrote:
>> Not a question, just an observation.
>
> There more to it than just the amount of free RAM.
>
>> I say that Linux doesn't seem to handle memory well because my laptop
>> had 8GB RAM (which, frankly, Windows seems to find perfectly adequate
>> for a similar workload). Under Linux the RAM would fill up and then
>> swap space would be used. Then the machine would become largely
>> unresponsive - e.g. taking minutes to switch between windows.
>
> Linux uses otherwise unused RAM for disk cache:
>
> Link - Help! Linux ate my RAM!
>  - https://www.linuxatemyram.com/

Thanks. Some very useful info there. Bookmarked.

--
James Harris

Carlos E.R. <robin_listas@es.invalid> wrote:
> On 2024-05-01 12:32, James Harris wrote:
>> Not a question, just an observation.
>>
>> I say that Linux doesn't seem to handle memory well because my laptop
>> had 8GB RAM (which, frankly, Windows seems to find perfectly adequate
>> for a similar workload). Under Linux the RAM would fill up and then swap
>> space would be used. Then the machine would become largely unresponsive
>> - e.g. taking minutes to switch between windows.
>>
>> So I upgraded the RAM. It now has three times as much (i.e. 24GB)! But
>> even so, RAM has still steadily filled up until reaching the full 24GB.
>> What's more, it's now showing 4.8GB of swap space in use.
>
> No, your analysis is not correct.
>
> You have some buggy application or service that is eating all the
> memory. You have to search and hunt for it.

Another reply indicates that his issue is the fact that browsers
(Chrome, Firefox) are bloated, memory hungry, "leaks" that eventually
eat all of ram, driving any system's memory management to insanity.

This, unfortuantely, is just a fact of browsers. They eat memory to no
end.

On 02/05/2024 06:59, Carlos E.R. wrote:
> On 2024-05-01 12:32, James Harris wrote:
>> Not a question, just an observation.

....

> Do you want our help?

Not on this. At least not at this stage. Thanks for asking. However, I
may well come back on separate threads about other things which are more
genuinely problematic for me.

--
James Harris

James Harris <james.harris.1@gmail.com> wrote:
>It may be unwise to assume there's such expertise. A number of years ago
>I came up with what I thought was a sensible design for a paging system.
>I later found that it corresponded very closely to the one used in
>Windows and differed markedly from the paging system used in Linux. I
>remember thinking that the latter was not a good design and since then
>I've experienced problems with Linux when memory fills up which I tend
>to assume are down to the design of its paging system - hence the,
>admittedly provocative, title of this thread.

There is more than way to do it. Windows and Linux are both operating
systems, and still fundamentally different. For example, firing up a
new process is vastly more expensive in Windows than Linux's fork()
mechanism. There is no "better" or "worse" in that comparision.

I don't know enough about VMS, but didn't the guy who led the design
the process and memory management layer for Windows NT come from
Digital Equipment?

Greetings
Marc
--
----------------------------------------------------------------------------
Marc Haber | " Questions are the | Mailadresse im Header
Rhein-Neckar, DE | Beginning of Wisdom " |
Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 6224 1600402

Marc Haber <mh+usenetspam1118@zugschl.us> wrote:
> James Harris <james.harris.1@gmail.com> wrote:
>>It may be unwise to assume there's such expertise. A number of years ago
>>I came up with what I thought was a sensible design for a paging system.
>>I later found that it corresponded very closely to the one used in
>>Windows and differed markedly from the paging system used in Linux. I
>>remember thinking that the latter was not a good design and since then
>>I've experienced problems with Linux when memory fills up which I tend
>>to assume are down to the design of its paging system - hence the,
>>admittedly provocative, title of this thread.
>
> There is more than way to do it. Windows and Linux are both operating
> systems, and still fundamentally different. For example, firing up a
> new process is vastly more expensive in Windows than Linux's fork()
> mechanism. There is no "better" or "worse" in that comparision.
>
> I don't know enough about VMS, but didn't the guy who led the design
> the process and memory management layer for Windows NT come from
> Digital Equipment?

Yes, Dave Cutler, chief architect for VMS, plus much of the DEC VMS
team that remained at the time.

https://en.wikipedia.org/wiki/Dave_Cutler

The core innards of NT are supposedly *very* VMS like.

Rich <rich@example.invalid> wrote:
>The core innards of NT are supposedly *very* VMS like.

And VMS was VERY different to Unix, so it is hardly surprising that
Windows today does many things different than Linux. And I think
that's a good thing.

Greetings
Marc
--
----------------------------------------------------------------------------
Marc Haber | " Questions are the | Mailadresse im Header
Rhein-Neckar, DE | Beginning of Wisdom " |
Nordisch by Nature | Lt. Worf, TNG "Rightful Heir" | Fon: *49 6224 1600402

On 2024-05-02 16:11, Rich wrote:
> Lawrence D'Oliveiro <ldo@nz.invalid> wrote:
>> On Thu, 02 May 2024 07:47:19 +0200, Marc Haber wrote:
>>
>>> No mechanic ever had to appear in front of a board to justify the use of
>>> spanner and screw driver, right?
>>
>> I’m sure they do, to whomever controls the purse strings at the garage.
>
> At nearly all auto-shops, the basic tools (spanners/screwdrivers/etc.)
> are purchased and owned by the actual mechanics using them and not the
> operator of the garage. What the shop owner provides are the
> "infrastructure" items (lifts, air compressor, some of the more
> specialized tools, such as the tyre changer or the 'computer diagnostic
> rack', etc.).

Not here.

The entrepeneur, the boss, the owner, hires the mechanics and must
provide everything. Protective clothing and gear, tools. Yes, every tool
is owned by the garage. The employee is not required to provide anything
except his person and knowledge and time and effort.

--
Cheers, Carlos.

On 2024-05-02 15:54, James Harris wrote:
> On 01/05/2024 15:06, David W. Hodgins wrote:
>> On Wed, 01 May 2024 06:32:18 -0400, James Harris
>> <james.harris.1@gmail.com> wrote:
>>> Not a question, just an observation.
>>> I say that Linux doesn't seem to handle memory well because my laptop
>>> had 8GB RAM (which, frankly, Windows seems to find perfectly adequate
>>> for a similar workload). Under Linux the RAM would fill up and then swap
>>> space would be used. Then the machine would become largely unresponsive
>>> - e.g. taking minutes to switch between windows.
>>> So I upgraded the RAM. It now has three times as much (i.e. 24GB)! But
>>> even so, RAM has still steadily filled up until reaching the full 24GB.
>>> What's more, it's now showing 4.8GB of swap space in use.
>>
>> Likely you're using a system that's configured for maximum total
>> throughput,
>> not best response time.
>>
>> Create a file, /etc/sysctl.d/tales.conf with the contents ...
>> # Reduce applications being swapped
>> vm.swappiness=1
>> # Don't shrink the inode cache
>> vm.vfs_cache_pressure=50
>>
>> Then run the command "sysctl --system" (as root).
>>
>> See
>> https://rudd-o.com/linux-and-free-software/tales-from-responsivenessland-why-linux-feels-slow-and-how-to-fix-that
>> for an explanation.
>
> I very much like the control that Linux gives to a user but ISTM that
> what you say here is a great example of the issue I was pointing out in
> the original post: IMO page depletion should be handled better *by
> default* and not need human intervention for normal scenarios.

It is up to you to set the defaults.

The defaults you get were set by the distribution, and each distribution
does this differently.

hint: ulimit.

--
Cheers, Carlos.

On Thu, 02 May 2024 09:54:30 -0400, James Harris <james.harris.1@gmail.com> wrote:
<snip>
> I very much like the control that Linux gives to a user but ISTM that
> what you say here is a great example of the issue I was pointing out in
> the original post: IMO page depletion should be handled better *by
> default* and not need human intervention for normal scenarios.

A system designed for a single user will likely try to make fast response time for
that one user the priority.

Unix and Linux were not designed for that. They were designed to be multi-user
systems that make the best use of available resources the priority, even if
that is at the expense of the response time for one of the poeple sitting at one
of the terminals.

The settings can easily be tweaked to suit the system administrator's priorities.

A distribution is free to alter the defaults it provides, but most choose to keep
the upstream defaults, to ensure it matches the documentation from the author
of the software.

Regards, Dave Hodgins

On 2024-05-02 16:06, James Harris wrote:
> On 02/05/2024 07:01, Carlos E.R. wrote:
>> On 2024-05-01 22:42, Kenny McCormack wrote:
>>> In article <v0tkdc$3780t$1@dont-email.me>,
>>> Lew Pitcher  <lew.pitcher@digitalfreehold.ca> wrote:
>>>> On Wed, 01 May 2024 11:32:18 +0100, James Harris wrote:
>>>>
>>>>> Not a question, just an observation.
>>> ...
>>>>
>>>> So, if you want help, how about providing some details.
>>>
>>> OP was pretty clear that he was *not* seeking help.
>>
>> Because he is blaming Linux at large, when he has a problem in *his*
>> system. He made a wrong diagnosis.
>
> Yes, I am (at least tentatively) blaming Linux but not for the problem
> you think. I wouldn't expect Linux to prevent programs from gobbling up
> memory but I would expect it to manage memory hogs more gracefully than
> it does. IME Windows handles the same situation better - and that may be
> down to the different designs of their paging systems.
>
> Don't get me wrong. I much prefer Linux to Windows and have often seen
> Windows get into a worse situation under different circumstances. But
> for page management ISTM that the design of Linux's paging subsystem may
> not be the best.

You can configure Linux to crash the application that is behaving badly
by grabbing all the memory. It is up to you, the boss.

The philosophy is not to nanny care for you. It does what you asked.
More memory? Yes sir. Till death does us part. Your orders will be obeyed.

:-)

--
Cheers, Carlos.

Carlos E.R. <robin_listas@es.invalid> wrote:
> On 2024-05-02 16:06, James Harris wrote:
>>
>> Yes, I am (at least tentatively) blaming Linux but not for the problem
>> you think. I wouldn't expect Linux to prevent programs from gobbling up
>> memory but I would expect it to manage memory hogs more gracefully than
>> it does. IME Windows handles the same situation better - and that may be
>> down to the different designs of their paging systems.
>>
>> Don't get me wrong. I much prefer Linux to Windows and have often seen
>> Windows get into a worse situation under different circumstances. But
>> for page management ISTM that the design of Linux's paging subsystem may
>> not be the best.
>
> You can configure Linux to crash the application that is behaving badly
> by grabbing all the memory. It is up to you, the boss.
>
> The philosophy is not to nanny care for you. It does what you asked.
> More memory? Yes sir. Till death does us part. Your orders will be obeyed.
>
> :-)

I didn't know there was an option not to have it do that (via the
Out Of Memory Reaper). It seems the alternatives are to reboot or
risk the kernel crashing, so your description seems about right.

https://www.oracle.com/technical-resources/articles/it-infrastructure/dev-oom-killer.html

The OP has noted now that the process that consumes their RAM is
Chrome or Firefox. I've not seen a detailed description of why it
happens, but I've long noted that Firefox seems to expand its RAM
usage to the available space, but different from a memory leak in
that it usually leaves a certain amount free. I assume that this
in intended behaviour. I run current Firefox on a PC with 2GB RAM
and I don't have it getting killed by the kernel, nor do I have
problems with kernel crashes/reboots. I've also tried running
recent Firefox on a PC with 512MB RAM and noticed that it performs
much worse than with 2GB RAM, slowing down to a crawl while loading
some websites, suggesting that it really does need more RAM in that
case.

I don't use web browsers to play video. If you're streaming super
high resolution video through your browser with the latest and
greatest compression algorithms, then it probably has the right
to chew up a lot of RAM.

--
__ __
#_ < |\| |< _#

On Thu, 2 May 2024 14:47:24 +0100, James Harris wrote:

> A number of years ago
> I came up with what I thought was a sensible design for a paging system.
> I later found that it corresponded very closely to the one used in
> Windows and differed markedly from the paging system used in Linux. I
> remember thinking that the latter was not a good design ...

Look at all the effort those smart Microsoft engineers had to go to,
just to get Windows file copying to work properly:
<https://techcommunity.microsoft.com/t5/windows-blog-archive/inside-vista-sp1-file-copy-improvements/ba-p/723622>

On 02/05/2024 14:50, James Harris wrote:
> On 01/05/2024 11:53, The Natural Philosopher wrote:
>> On 01/05/2024 11:32, James Harris wrote:
>>> Not a question, just an observation.
>>>
>>> I say that Linux doesn't seem to handle memory well because my laptop
>>> had 8GB RAM (which, frankly, Windows seems to find perfectly adequate
>>> for a similar workload). Under Linux the RAM would fill up and then
>>> swap space would be used. Then the machine would become largely
>>> unresponsive - e.g. taking minutes to switch between windows.
>>>
>>> So I upgraded the RAM. It now has three times as much (i.e. 24GB)!
>>> But even so, RAM has still steadily filled up until reaching the full
>>> 24GB. What's more, it's now showing 4.8GB of swap space in use.
>>>
>>>
>> That's not Linux. That's a rogue program with a memory leak...
>
> It may not be a memory leak but it's certainly a rogue program -
> chrome/chromium in this case, though I've had similar and worse problems
> with Firefox which seems to me more likely to have a genuine memory leak.
>
> However, the point of the original post was to vent a frustration about
> the way that Linux handles such memory deletion when it happens.
>
>
It is not hard to write a javascript program that eats all memory and
place it on a website.

I even encountered a website that silently ran some sort of proxy server
in my browser and used up all my upload bandwidth.

Linux at least protects you from such sites having root access by default...

--
Future generations will wonder in bemused amazement that the early
twenty-first century’s developed world went into hysterical panic over a
globally average temperature increase of a few tenths of a degree, and,
on the basis of gross exaggerations of highly uncertain computer
projections combined into implausible chains of inference, proceeded to
contemplate a rollback of the industrial age.

Richard Lindzen

On 02/05/2024 14:54, James Harris wrote:
ISTM that
> what you say here is a great example of the issue I was pointing out in
> the original post: IMO page depletion should be handled better *by
> default* and not need human intervention for normal scenarios.
>
IME that is exactly what Linux *does* do.
Handles everything perfectly by default.
Swapping is what happens when *all else fails*.

In Windows, programs simply crash.

--
Microsoft : the best reason to go to Linux that ever existed.

On 02/05/2024 15:01, James Harris wrote:
> On 01/05/2024 21:42, Kenny McCormack wrote:
>> In article <v0tkdc$3780t$1@dont-email.me>,
>> Lew Pitcher  <lew.pitcher@digitalfreehold.ca> wrote:
>>> On Wed, 01 May 2024 11:32:18 +0100, James Harris wrote:
>>>
>>>> Not a question, just an observation.
>> ...
>>>
>>> So, if you want help, how about providing some details.
>>
>> OP was pretty clear that he was *not* seeking help.
>
> Correct. I don't want to overtax the expert assistance that's clearly
> available from posters to this newsgroup and would rather ask for help
> relatively rarely, so I am not looking (at least, yet!) for a solution.
> My point, in this thread, was, instead, to comment on what I perceive to
> be an OS-design issue - a long-term interest.
>
>
In short anti-linux trolling.
--
“when things get difficult you just have to lie”

― Jean Claud Jüncker

On 03/05/2024 00:23, Computer Nerd Kev wrote:
> Carlos E.R. <robin_listas@es.invalid> wrote:
>> On 2024-05-02 16:06, James Harris wrote:
>>>
>>> Yes, I am (at least tentatively) blaming Linux but not for the problem
>>> you think. I wouldn't expect Linux to prevent programs from gobbling up
>>> memory but I would expect it to manage memory hogs more gracefully than
>>> it does. IME Windows handles the same situation better - and that may be
>>> down to the different designs of their paging systems.
>>>
>>> Don't get me wrong. I much prefer Linux to Windows and have often seen
>>> Windows get into a worse situation under different circumstances. But
>>> for page management ISTM that the design of Linux's paging subsystem may
>>> not be the best.
>>
>> You can configure Linux to crash the application that is behaving badly
>> by grabbing all the memory. It is up to you, the boss.
>>
>> The philosophy is not to nanny care for you. It does what you asked.
>> More memory? Yes sir. Till death does us part. Your orders will be obeyed.
>>
>> :-)
>
> I didn't know there was an option not to have it do that (via the
> Out Of Memory Reaper). It seems the alternatives are to reboot or
> risk the kernel crashing, so your description seems about right.
>
> https://www.oracle.com/technical-resources/articles/it-infrastructure/dev-oom-killer.html
>
> The OP has noted now that the process that consumes their RAM is
> Chrome or Firefox. I've not seen a detailed description of why it
> happens, but I've long noted that Firefox seems to expand its RAM
> usage to the available space, but different from a memory leak in
> that it usually leaves a certain amount free. I assume that this
> in intended behaviour. I run current Firefox on a PC with 2GB RAM
> and I don't have it getting killed by the kernel, nor do I have
> problems with kernel crashes/reboots. I've also tried running
> recent Firefox on a PC with 512MB RAM and noticed that it performs
> much worse than with 2GB RAM, slowing down to a crawl while loading
> some websites, suggesting that it really does need more RAM in that
> case.
>
> I don't use web browsers to play video. If you're streaming super
> high resolution video through your browser with the latest and
> greatest compression algorithms, then it probably has the right
> to chew up a lot of RAM.
>
I use browsers to play videos. RAM usage is not massive. CPU usage is.

What chews memory are *commercial* websites loaded with (deliberately)
buggy javaScript that cause javaScript engines to go into meltdown. How
that is handled is browser dependent.

Ublock Origin helps massively, but is not a complete answer.
What is a massive help at leats in Mint Mate is the System Monitor
widget that I keep in the task bar permanently displaying CPU, RAM and
Network usage as teeny graphs.
It is very easy to see when memory is all grabbed by a process rather
than simply cache
And monitor how much gets released when you close a website page.

It is configurable to do what htop, top and atop do in a much simpler
way for GUI users
Right now, it looks like Thunderbird is my biggest surprise at almost a
gigabyte?

I will close it and reopen it. It shouldnt be THAT heavy on memory.

--
“when things get difficult you just have to lie”

― Jean Claud Jüncker