You may find helpful information here https://openhpc.community/support/faqwiki/ I personally would avoid centos8, as the "stream" model is more suited to a development world than a production world.

Hi Recently, I have changed the ssh port 22 of the master node to another number. When I reboot the compute nodes, it came out with: [root@nucimp wwmkchroot]# pdsh -w nuc[1-2] uptime nuc2: ssh: connec

Dear Community, OpenHPC is currently accepting applications for a new mentorship program that aims to give students practical experience contributing to an open-source project and an opportunity to wo

I have a weird problem on a recent stateful deployment on centos 7.9 ohpc-release 1.3-1.el7. Deployment goes well and the system is bootable, however, the kernel parameters set in kargs do not show up

I have read a few blurbs here about pmix changing the way it deploys. And the package in the docs doesn't even exist anymore. For the time being I have my users running with mpirun, but is pmix broken

Hi, I'm buiding a cluster with openHPC (openSUSE 15.3/openHPC 2.4) to test if it can replace our Rocks cluster. I made my master, an image for a few nodes, I can ssh them. But when I try to build a ba

Dear Community, The OpenHPC Technical Steering Committee (TSC) is approaching completion of its sixth year of operations and would like to invite nominees interested in volunteering on the TSC for the

Hi, I am Sr. Systems Engineer in Columbia University New York. I am new to OpenHPC community. We are planning to build new cluster with OpenHPC in our Data Center. Can anyone recommend any documentati

Hi All, I have a script that successfully images a variety of servers. Imaging a superrmicro server with intel xeon E5-2620 v4 (broadwell cpus) server with SMS eth0 worked fine without supplying kargs

Hi everyone I am following the instruction for installing openhpc 2.4.When I moved to " # Enable internal interface for provisioning ", it came up: " [root@nucimp liujj]# vim /etc/warewulf/provision.c

Dear experts, I setup a head node running OpenHPC 2.4 with Leap 15.3 warewulf 3 for 4x compute nodes with Intel CPU. It works fine. I am to add 4x compute node with AMD CPU to run CentOS8.5 ( or RHEL8

Hi everyone I have a smaller cluster, one adm node and two compute nodes running red hat 8.5. A hard disk array is attached with adm node by SAS. I have a crazy thought about expanding our storage by

Hello. please help i have changed the public ip of a master server, and a got the problem that TFTP service based on xcat is not running, i have restarted de xcatd, but the problem continues. that cau

I think it’s extremely premature to ask for EL9 support. EL9 is out for 6 days now. Keep in mind that OpenHPC is free software. If you look back to when EL8 was released OpenHPC only added support whe

As a user of ohpc, I suggest upgrading the OS after two minor versions, and for now, adapting rocky8.6 is a better choice. 发件人: OpenHPC-users@groups.io <OpenHPC-users@groups.io> 代表 rusholme+ohpc via g

Hi, Is there a timeline yet for OpenHPC on the released RHEL 9 and clones? Is it dependent on Rocky9? Thanks, Ben

While golden imaging I ran into this issue (ohpc 2.4, cheetah 15k 300g disk, which works as it still boots centos 6 on it). Falling back to stateful, same problem. https://github.com/warewulf/warewulf

Hello, Is anyone using ssh certificates? We are looking at them for our head node access. It's a university cluster and each campus has their own authentication method, there isn't a central authority

Hello all. For those running SLURM with login nodes, how are you guys retrieving sacct info? My setup uses stateless configuration and the login nodes queries the slurmctld daemon with DNS SRV entries

Hello, With the end of CentOS, the HPC world, from what I have been able to read, has been focused on two distributions to replace it: AlmaLinux and Rocky. Both distributions have similar behaviors an