Bug 34132 - System is unresponsive while using dd to copy DVD ISO to USB stick/key
Summary: System is unresponsive while using dd to copy DVD ISO to USB stick/key
Status: RESOLVED OBSOLETE
Alias: None
Product: IO/Storage
Classification: Unclassified
Component: Other (show other bugs)
Hardware: All Linux
: P1 high
Assignee: io_other
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2011-04-30 16:12 UTC by Jure Repinc
Modified: 2015-02-08 01:35 UTC (History)
2 users (show)

See Also:
Kernel Version: 2.6.38.2
Subsystem:
Regression: No
Bisected commit-id:


Attachments
dmesg2 (195.94 KB, text/plain)
2011-04-30 16:13 UTC, Jure Repinc
Details
lspci2 (9.54 KB, text/plain)
2011-04-30 16:15 UTC, Jure Repinc
Details
config2 (67.21 KB, text/plain)
2011-04-30 16:15 UTC, Jure Repinc
Details

Description Jure Repinc 2011-04-30 16:12:13 UTC
When I use

dd if=<some_iso_file> of=<device_of_usb_key>

The system becomes very unresponsive. X is almost completely unusable, maybe you can move a mouse for some time but click do nothing. Even togling the Caps Lock doesn't do anything. Trying to SSH is also almost impossible during this. System is like this all the time while using the dd command. After it finishes system again becomes responsive. I've noticed this on two of my computers:
1: AMD Athlon 64 3000+, 2 GB of memory, kernel 2.6.38.2
2: AMD Athlon II P320 Dual-Core, 4 GB of memory, kernel 2.6.39-rc2
The USB key is this one: http://www.takems.com/products.php?categ=usb&prod=MEM-Drive_Easy_II

What I expected was that the doing dd wouldn't have a noticable impact on responsivness of the rest of the system.
Comment 1 Jure Repinc 2011-04-30 16:13:52 UTC
Created attachment 55962 [details]
dmesg2
Comment 2 Jure Repinc 2011-04-30 16:15:00 UTC
Created attachment 55972 [details]
lspci2
Comment 3 Jure Repinc 2011-04-30 16:15:38 UTC
Created attachment 55982 [details]
config2
Comment 4 Alan 2012-08-23 13:46:24 UTC
This has been somewhat addressed in later kernels
Comment 5 Simon Ruggier 2015-02-08 01:35:55 UTC
Alas, I'm seeing this behaviour on a 3.16 kernel (although, it's a distribution kernel from Debian testing).

$ uname -a
Linux SPR-T500 3.16.0-4-amd64 #1 SMP Debian 3.16.7-ckt2-1 (2014-12-08) x86_64 GNU/Linux

Running `echo never > /sys/kernel/mm/transparent_hugepage/defrag`, seems to make the general responsiveness problem go away, while running `sysctl vm.dirty_background_ratio=1 vm.dirty_ratio=2` seems to help reduce the slowdown for individual processes that try to fsync. I confirmed that reverting these changes causes the problem to come back.

Should I open a new bug, or should this one be reopened?

Note You need to log in before you can comment on or make changes to this bug.