Ubuntu
torque package

torque-mom LXD mlockall() failure

Asked by LF

Hello

I am not sure I am taking the right approach, but I am trying to install the torque batch system inside an LXD container.

it appears that the packaged version of torque-mom uses mlockall() call which cannot be executed inside of
an un-privileged container. However, it appears that torque has a compile-time switch to replace mlockall() with something more robust.

I am not sure if the impact of changing the way how the memory is allocated but maybe torque-mom could detect the problem at run-time and switch the memory allocation mechanism on the fly?

If it is a stupid idea to run torque inside LXD than could someone advice me on this?

I have a one-CPU 10core system and I thought I will create one container as a user login/submit node, one for the main OS and the remaining 8 cores will be used inside container(s) as compute nodes.

Thank you very much for your help
LF

Question information

Language:
English Edit question
Status:
Expired
For:
Ubuntu torque Edit question
Assignee:
No assignee Edit question
Last query:
Last reply:
Revision history for this message
michael (yellupcm-gmail) said : 		#1

Check this site.

http://docs.adaptivecomputing.com/torque/5-1-1/Content/topics/hpcSuiteInstall/manual/1-installing/installingTorque.htm

Copied from above site.

2.3.1 Requirements
Supported Operating Systems
CentOS 6.x, 7.x
RHEL 6.x, 7.x
Scientific Linux 6.x, 7.x
SUSE Linux Enterprise Server 11, 12

CentOS 5.9, RHEL 5.9 and Scientific Linux 5.9 are supported, largely to continue support for clusters where the compute nodes operating systems cannot be upgraded. We recommend that the TORQUE head node run on the supported operating systems listed above.

Could not find any information about it working on Ubuntu.

Revision history for this message
Launchpad Janitor (janitor) said : 		#2

This question was expired because it remained in the 'Open' state without activity for the last 15 days.