Web lists-archives.com

Re: Debian Stretch freezes often

On 4/23/19, Cindy Sue Causey <butterflybytes@xxxxxxxxx> wrote:
> Knowing what type of program or programs is open each time wouldn't
> hurt. Knowing specifically which ones... ok, yeah, I know... the list
> could become seemingly endless. My thought process is that maybe
> certain ones or combinations of certain genres/families are calling
> something up that causes this.
> Knowing any personal tweaks done at installation wouldn't hurt, either..

> < blah-blah SNIPPED for brevity >

tl;dr... Dear Browsers, the ability to get back control of task
scheduling the autoupdate feature.. I just had a positive ending
experience that showed me networking transfer rates might not be the
only thing that is negatively affected by uncontrollable autoupdates.

The long version.. Just had a breakthrough in my own similar deal of
things freezing up A LOT and very frequently lately. Tripped over
autoupdate potentially running in my browser despite conscious efforts
to avoid that k/t dialup. There was no toggle switch to be found
during multiple searches lately.

A determined search today found SuperUser [0] with a slightly older
related thread. It seems a no-brainer to stretch that out to apply to
any package that has an autoupdate feature. I can't name any right
now, but, on rare occasion, I've seen that feature floating around
outside of the browser package family.

The RUMOR is that the autoupdate executable will have been
*relatively* safely disarmed if moving it or renaming it results in a
message similar to.... the one I'm... so not seeing just this second

"An error occurred while checking for updates"

That error message appears in places like "opera://about".

Seconds after that adjustment, my system is smoking right through
things.... flipping through tabs without hesitation and the cursor is
responding IMMEDIATELY to the touchpad.

This is right off the bat when it was previously already bogged down
even this early into a new browsing session. Prior to this change,
I've been having ALL KINDS of slow down issues to that point of having
to hit the hardware button to shut down WAY TOO OFTEN.

** Afterthought: This still might not help with the bogged down cursor
and other things that occur when some of us simply walk away from our
computers for a few minutes.

This find sounds like it may also be applicable to that recent thread
where I caught a blip about desktop PCs, massive amount of images, and
development gearing evermore toward mobile computing. That's a "boy,
howdy, is it ever!" Mobile phone and tablet-type computing puts money
in the hands of those who own the associated wireless services that
frequently are not free. *hm*

Last second observation: I just ran "free -m". HUGE chunk of my
limited memory is already swamped, yet here it is still smokin' right
on along.

Until this second, I leaned heavy toward memory usage as the front
leader for the freezing. I currently have 1GB memory and 1.66Ghz
processing oomph. I took those freezes all in stride, simply dealt
with it all this time because of those low numbers. Crocheting fills
the time in between webpage turns. :)

I figured autoupdate trashes networking speeds, yeah, but never
thought about it having a notable negative effect against the rest of
the operations, e.g. SLOW cursor response times. The change today was
immediate and is *notable* in my usage case.

SECURITY RELATED DISCLAIMER: Disorienting autoupdates in any way means
that one has to consciously schedule those updates to occur during
one's computing downtime.

There's also the possibility that disorienting updates could lead to
smoke inducing, unpredictable behaviors that could break a whole lot
more than just a cursor's slow response time or the even slower
CTRL+Tab/CTRL+SHIFT+Tab webpage navigation travels.

Cindy :)

[0] Yeah, I know, that other operating system mostly. It still worked
in my usage case.

PS Also just found something called "AskVG". Never heard of it before,
but information from their "How to Disable or Block Auto-Update" page
"felt" good.

Task schedulers... that's beyond my comprehension right now. Users
having control via system-wide task scheduling always sounds a whole
lot more *stable* and *appropriate* versus having to wing it via
functions such as "rename", "relocate", and "delete" from within a
package's Developer-implemented file hierarchy.

Cindy-Sue Causey
Talking Rock, Pickens County, Georgia, USA

* runs with birdseed *