5

Two example cases:

  • Example 1: Mount an external unencrypted drive (GPT, ext4) and copy a 8GB file from my computer's SSD (GPT, ext4) -> The whole GUI will become unresponsive. No input devices work.
  • Example 2: Start a 40GB VM with Virtualbox -> The whole GUI will become unresponsive. No input devices work.

Using iotop I can see that these disk read/write operation took up 100% CPU.

How can I avoid an unresponsive computer while moving large data?

2
  • I have not seen this happen before but if it did the first thing I'd check is weather this is GUI related or not exiting your WM and trying to reproduce the problem from CLI. That should let you determine if the WM has anything to do with it.
    – krowe
    Jul 5, 2014 at 18:49
  • @krowe Thank you for your reply. My question was already answered, but the original poster removed his answer. So I posted my own version now. Jul 5, 2014 at 19:01

1 Answer 1

3

For unknown reasons the I/O scheduler has been changed from the default configuration (cfq - completely fair queuing) to deadline.

Since I have the tlp tools installed to reduce the power consumption, my solution was to set the scheduler for all drives back to the default value in /etc/default/tlp:

...
# Select io scheduler for the disk devices: noop/deadline/cfq (Default: cfq)
# Separate values for multiple devices with spaces.
#DISK_IOSCHED="cfq cfq"
...

In this part you uncomment the last line to:

DISK_IOSCHED="cfq cfq"

Alternatively you can do the same with the following command:

sudo sed -i 's/#DISK_IOSCHED/DISK_IOSCHED/' /etc/default/tlp
1

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .