The FAQ entry you cite[1] refers to behavior that existed 13 years ago, regarding to metadata updates related to data which was not flushed to disk; this behavior was fixed per the details in the

I am not in a position to make a statement about licensing or legal issues related to redistributing xfsprogs code or binaries along with the Warewulf project. Unlike Bradley Kuhn, I'm not a lawyer

This wasn't recovery speed. Every open file on the filesystem was 0 size when the filesystem came back up (could trace it back to jobs that were running at the time, libraries in use, data files

Hi Sandeen, Darrick I have sumbitted a PR to warewulf[1](Warewulf is a scalable systems management suite originally developed to manage large high-performance Linux clusters.) to support xfs

Hi all, The OpenHPC roadmap shows a 2.0 release for Centos 8 expected in 2020 Q1. Is there any update on when this is expected to land? many thanks Steve Brasier http://stackhpc.com/ Please note I

Dear Mr. Hearns, thank you, I think so too... :-) I do not have the necessary hardware yet, I will decide in the near future if and what I want to buy....it is still a hobby... :-) Thank you

H Mario. I think you are getting very good advice here. On your OpenHPC cluster there will be a batch or scheduling system. It this named Slurm? I would convert the Docker container to Singularity

Dear Mr. Brayford, big thank you for answer also. I read about converting a docker to Singularity just yesterday. I will have that in mind also. Thank you. Do you guys mind if I link this thread to

It would be possible to convert the Docker image to a Charlecloud or Singularity image and be able to run across multiple nodes with MPI. I've done this successfully with Charliecloud. A quick

Dear Mr. Reber! Thanks for your ultra-quick answer. Yes, everything is done inside the container. That's why it is so convenient for me... :-) I will digest your answer and do some more digging.

I would say it is possible, but probably not as easy as you hope. From what you describe, it sounds like everything is running in your container. mpirun and the process mpirun starts. The difference

Hello, first of all, I must say I am a total newbie to OpenHPC and not a software engineer. So please excuse any dumb question. However, I was not able to find any useful information to my questions

Hi - This doesn't have anything to do with OpenHPC. What version of NFS are you trying to use? If >= v.4, you will need to export an NFS root under which your exports are bind-mounted. Sadly, the

Padma, The answer we need to make any definitive recommendation: Is the remote “nfs-home” share available to mount on the compute nodes directly? If the answer is no, a fix is more complex.

I am getting this issue as well. Was there ever any resolution to this?

Dear Padma, I agree with Richard, you may directly mount the /home folder (from NFS server) to all the compute nodes as you have already done for the head node. Best regards, Raja -- Rajendra

No Elegant way to do this, although I have had to do it for a really "special" case, but would not do it for a "home" export. Option 1 mount the /home from each "compute" resource individually, if

Sometimes it helps to find a whiteboard. Draw on the whiteboard the cluster head node, the NFS server, the compute nodes. Draw on the cluster ethernet switch and the infiniband switch. Then draw on

We also have a NFS storage server installed with - CentOS Linux release 8.1.1911 (Core). Network – Ethernet and Infiniband. Does the NFS storage server have a physical cable connected to the

Historically, the answer is "You can't re-export NFS shares". There's a bunch of info on google about that, as well as this possibly interesting RHEL article