Checkbox Provider - Base

cpu-offlining failed on high core count sytem due to out of range error

Bug #1583628 reported by Jeff Lane 
14
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Checkbox Provider - Base
Opinion
Medium
Jeff Lane 

Bug Description

cpu topology ran on a system with 144 cores and failed on core 99 with this error:

Offlining cpu96
Offlining cpu97
Offlining cpu98
Offlining cpu99
/tmp/nest-ost1zdw_.aed22c82c1c75686f03b21e557a226092573a9b941959f278bde1cac5e8c5cd3/cpu_offlining: line 14: echo: write error: Numerical result out of range
ERROR: Failed to offline cpu99

Revision history for this message
Jeff Lane  (bladernr) wrote :

cpuinfo from system

tags: added: hwcert-server
Revision history for this message
Jeff Lane  (bladernr) wrote :

output of test

Revision history for this message
Po-Hsu Lin (cypressyew) wrote :

This error is quite interesting,
line 14 is:
    echo 0 > /sys/devices/system/cpu/$cpu_num/online

The number it tries to write into the "online" file shouldn't be changed.

Can you try to make it offline manually? Simply run:

echo 0 | sudo tee /sys/devices/system/cpu/cpu99/online

will be fine, or you can run the original command after switch to root.

Jeff Lane  (bladernr)
Changed in plainbox-provider-checkbox:
status: New → Confirmed
assignee: nobody → Jeff Lane (bladernr)
importance: Undecided → Medium
Revision history for this message
Jeff Lane  (bladernr) wrote :

Setting this to Opinion... I've been unable to recreate this and similar systems (up to 256 cores) have not demonstrated this bug.

Changed in plainbox-provider-checkbox:
status: Confirmed → Opinion
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Duplicates of this bug

Subscribing...

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.