Hi guys, could you help me with the instructions to build an xcat image by using these files: http://mirror.centos.org/centos/8/BaseOS/x86_64/kickstart/ I have already downloaded this folder to the xc

For those who are managing OpenHPC clusters with OpenSUSE 15, may I ask what the status of the OFED stack is with this distribution? How is OFED installed with OpenHPC? I see that it is available as a

Hello fellow OpenHC users, I inadvertently deleted the /opt/xcat directory on the compute nodes of a cluster that I manage. I followed the instructions in the Install Guide (v2.0): CentOS8.2/x86 64 +

Steve, You can always point to the CentOS vault repos instead of using the rolling release repos by updating the URL in the yum.resos.d/XXX.repo files. For example the URL’s for the 3 main CentOS 8.2

Hi all, I notice that the openhpc wiki for 8.x [1] says that Centos 8.2 is the "compatible OS version", and I seem to remember a comment somewhere that it wasn't tested against 8.3 (although I can't f

Hello Guys I really thank you in advance for your help about any clues related to this error that I get when I try to boot my first comput node, I am following the guide "Install Guide (v2.0): CentOS8

Hi, this is an issue that we all know but I’ve never came across a good solution. If something went out of track in SLURM the logs are just flooded with repeated messages that does not help at all sol

Hello everyone! Not too long ago, I saw a presentation for using EasyBuild and thought "I should give that a try!". Today, I did. It didn't quite go how I thought it would. I've got a bunch of questio

Happy Friday eve everyone! We’ve been struggling trying to fill an open HPC system administrator position and I was wondering if anyone has suggestions on places to advertise the open position. We’ve

I'm at a bit of a loss on this one. I have 3 nodes running on CentOS7 in a stateful configuration. When they boot, I have no network. I can manually start the network and it works fine, but it will no

Hello I'm preparing a little HPC and using manual v1.3.9 openHPC CentOS Warewulf/Slurm. I have two nodes and both are down. I have changed manually the state to "idle" and "resume" with scontrol and a

---------- Forwarded message --------- From: jitendra Phull <jitendraphull@...> Date: Wed, Feb 3, 2021, 18:54 Subject: Need Help in provisioning HPC cluster To: <OpenHPC-users+help@groups.io> Dear Fri

Dear Users, I've tried deploying OpenHPC Both releases with AMD EPYC 7452 on CentOS 7 and 8 and unable to do stateful provisioning . Stateless works perfectly as it should be . Now when i do stateful

There are several changes you have to make in the provisioning config/partitioning to make this work with NVME. Look at the threads on the group and search for nvme stateful and you should find the de

Hello, We have setup an Ethernet-only OpenHPC cluster with 1 master node and 2 compute nodes according to below document: * "Install Guide (v2.0): CentOS 8.2/x86 64 + Warewulf + SLURM" * https://githu

Hello guys, I have a question for you if somebody can help me: Is there a way to customize the partitioning scheme by node using xcat with an environmental variable from xcat database in a partition f

Hello everyone, Does anyone know if there are future plans to support active/passive head nodes with OpenHPC? If so, would it be developed for version 1.x, or just version 2.x? Thanks in advance for y

FWIW, like many projects across the industry, MySQL dropped the master/slave vocabulary last July in favor of source/replica. In my view, none of us should be using those terms at all any more except

Having managed fault tolerant DHCP, DNS, MySQL, and HPC cluster tools. I can say that I've found the "fault tolerance" to be the biggest source of headaches and failures. For MySQL, Percona clusters d

Not such an odd request at all. The environment you have set up is where I'd like to be 6 months from now. So please share your experiences. Under CentOS6 and CentOS7 I use the "golden image" approach