Ask Your Question

soft lockup cpu

asked 2015-01-22 15:33:58 -0500

Fishbone gravatar image

updated 2015-01-23 15:43:51 -0500

rbowen gravatar image

Hello I am running openstack icehouse under the three node architecture, utilizing neutron networking with the ml2 plugin.

Whenever I attempt to launch an instance it resolves as if it is running fine, however; I cannot view the console of the vm, the web page comes up with the novnc background and no console window. When viewing the hypervisor's console I can see many lines output, mostly referencing an error: "soft lockup - cpu#0 stuck for 22s [qemu-system-x86]". When attempting to kill the instance it lingers in the database even after restarting the hypervisor. The resource is freed from the hypervisor and the only way to clear the entry from the database is to manually remove it. The hypervisor itself is still accessible although it may be a system crash. I am running the entire architecture on Ubuntu 14.04 x64 servers with the 3.17 linux kernel. Please let me know if I may be overlooking something.

Thank you for your time.

edit retag flag offensive close merge delete


Hello, I am still unable to find a solution to this. Any ideas, could it be hardware inefficiency?

Fishbone gravatar imageFishbone ( 2015-02-03 12:04:57 -0500 )edit

2 answers

Sort by ยป oldest newest most voted

answered 2015-01-23 22:13:19 -0500

Did you google for soft lockup - cpu#0 stuck for 22s [qemu-system-x86] ?

You will see that its most likely a kernel bug.

Its referenced here :

edit flag offensive delete link more


My apologies for the late response. I had previously assessed the kernel issue in reference to this. The linux kernel is the recommended version 3.17 but I still receive this error. I had not noticed any other solutions that recommended another method. They all leaned toward updating to kernel 3.17.

Fishbone gravatar imageFishbone ( 2015-01-29 09:41:09 -0500 )edit

I am also not sure I completely understand the thread you led me to. I am unfamiliar with the ksm feature and do not believe that I am applying it in my deployment. I have literally just completed the setup of openstack entirely utilizing the openstack documentation, which lead to this error.

Fishbone gravatar imageFishbone ( 2015-01-29 09:53:12 -0500 )edit

answered 2015-05-28 15:59:04 -0500

Fishbone gravatar image

Hello all, turns out my processor was unable to handle the compute load required to run a vm. Replaced the server and all is well now.

edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools



Asked: 2015-01-22 15:33:58 -0500

Seen: 1,520 times

Last updated: May 28 '15