My computer becomes unusable while a file transfer is in progress

This is how it looks:


just tested it too !

If I copy&paste the command I got the same error as you.
It’s cause the last sign ( ’ ) after the word “warn” got changed somehow during copy&paste from this forum to my terminal ( NOT the other way around !)

The htop screenshot shows a bunch of swap was used. If you ran out of RAM during the transfer, that would certainly cause your system to become unresponsive, even possibly crash with the kernel killing processes due to being out of memory (OOM killer).

1 Like

just out of interest: where do you see this in the htop screenshot ?
I see:
Mem 1.26/3.71G [1]
Swp 572M/3.84G

or does the yellow line indicate that all RAM was used short time ago ?

Ok, here you go!:


this doesn’t indicate to me any bad influences regarding responsiveness during transfers.

There is no kernel bug that I was thinking off.

So we need to search the other corners.
What are the answers to my questions in my first post in this thread ???

but try with a new created user first and then try to answer the remaining question afterwards, cause I think off an right problem as next ( => cit.: “This problem seems to happen only when copying or moving files from external USB storages (HDD or Flash) to home…”)

If your issue does NOT occur with a new user created (see above) I would do the following:
in an terminal sudo touch /.autorelabel [1] , reboot and test again transfers to you familiar user home dir (the newly created user is obsolet now)

Cause then it has something to do with the rights in your current home directory.
one part of that should get fixed with the relabel command above (man selinux)…

a file ist created (and vanishes after 2. reboot)

ls -al /

-rw-r–r--. 1 root root 0 28. Mai 02:48 .autorelabel

What? Why do suggest trying so many different things… especially without explaining what it will do and why you suspect SELinux to be involved in this issue.

Also, for clarity use proper code blocks to avoid typo problems as above. i.e. code

Mmmmh, I’m now running ~ 20 years linux. so it happens that I unnoticed don’t explain things in that deep it sometimes need to, cause it’s - to me - sooo obvious.

you answered to my contribution while I was still editing/changing it.
so see my previous contribution regarding rights. selinux is one part of it. (I added - while you was answering - “man selinux”)

typo problem ?
=> read why the user - and I - were unable to get the command running.

You are better up with helping to fix the users problem then criticizing guy’s they do !!!

I ask myself the same question… by the way, AHCI is turned on in BIOS…

Ok, I’ll try this…

maybe cause I want to fix your problem ?
should I better stop - I still have other things to do too - ?

Can you please type again step by step what I should, I’m confused with the above steps?

Know that it will likely take a long time to complete autorelabling your entire system.

I/we need to know what’s up with your home directory cause you noticed that the issue occur during transfers to your home dir. so there are mainly 2 parts involved: your home dir (deeper the disk with home on) and the extern disk and the minic between !

it could be user rights in your home dir. selinux is a part of rights. relabeling resets this part to the defaults (a clean(er) test environment).

again, if relabeling doesn’t fix your maybe somehow changed home rights (if issue still occur during transfer to your current home directory).
I suggest:

  1. to create a new user
  2. login to that user
  3. transfer files to that new home user (please test always with the same files !)
  4. does the issue thereby occur too ?

no => it’s your home directory (the old user one; don’t mix it up with the new user home dir.)

  • we need to look at the rights in deep

yes => something else

=> Questions to answer:

  1. why do issues occur when files transfered to (all) home directories and NOT from them ?
  2. what is up/ involved on the transfer side
  • disk,
  • Bios setting: AHCI (checked: okay ! ),
  • kernel bugs (checked: okay !)…)
  1. when did the issue start to appear
    3a. ever => box maybe to slow, but why are issues during transfers from extern to intern only
    3b. sometime ago => what was done sometimes ago (software update, …)
  2. what’s up with internal disk ?
    => gnome disks => smart
    => fsck

That’s a good list of questions. Before doing procedures involving change, such as autorelabeling or creating new users, why not try diagnostics first?

The iostat command may be useful to try. The sysstat package may first need to be installed.

iostat -mxh 5

This will print extended stats in human readable form every 5 seconds. Try this when transfering to and then from the internal disk and see if anything jumps out.

I think I wasn’t clear about the HOME directory, I said home because that’s the user directory but I’m actually transfering the files to the downloads folder, or music folder, etc… is there any reason to do all that if I wasn’t working with that directory directly?

Those are all in the home directory.

1 Like

Ok, I’m gonna leave this for tomorrow, I’m pretty tired, today was a busy day, I will follow all the instructions from @sixpack13 and @fasulia thank you for your attention today, have a good night everyone…

1 Like

change, such as autorelabeling
is an reset to defaults or a clean test env to begin with !

“…or creating new users
cause the user clearly mentioned that the issues occur when he/she transfers files to his/her directory (we are aware now it’s NOT it’s home dir)

Usually if those issues occur it’s a kernel bug (a trace but a still somehow running box) => checked with my dmesg command (a builtin command as relabeling is, too [need some time on an celeron, but …)

I guess it’s an right problem and I guess you won’t fetch them with iostat. (maybe you get “it’s slow” (we already know), but not why)
With a per some clicks created new user and login/out I get an idea what’s up with that “special” directory where the files transfered to …

I learned “Do the simplest thing first” (RHCE) !

man iostat
iostat - Report Central Processing Unit (CPU) statistics and input/output statistics for devices, partitions and network filesystems (NFS).

nice, but we already know it’s slow.
why to prove it again ?

I need to know WHY to get an idea how to fix it !

It doesn’t if the code formatting tool is used. That makes it monospace which can be copy-pasted to a terminal. If you don’t use it, it is treated as unicode text and formatted accordingly. However, unicode symbols, of course, do not work as expected terminals (or is that a shell issue and not a terminal emulator issue? I forget :thinking:) .

Anyway, I’ve edited that bit for you now. Please use the formatting tools. :+1: