Hallo,

I am trying to install lustre server with ldiskfs on a RHEL 8.10 kernel 
version: 4.18.0-553.5.1.el8_10.x86_64.

i have already successfully installed the lustre-ldiskfs-dkms module....


[#########~]$ sudo dnf list installed | grep lustre
kernel-debuginfo-common-x86_64.x86_64       4.18.0-553.5.1.el8_lustre           
      @lustre-server
lustre-ldiskfs-dkms.noarch                  2.15.5-1.el8                        
      @lustre-server

But when i am trying to install the lustre-2.15.5-1.el8 package i get the 
following error:


[########~]$ sudo dnf install lustre-2.15.5-1.el8
Error:
 Problem: package lustre-osd-zfs-mount-2.15.5-1.el8.x86_64 from lustre-server 
requires kmod-lustre-osd-zfs, but none of the providers can be installed
  - package lustre-all-dkms-2.15.5-1.el8.noarch from lustre-server conflicts 
with lustre-ldiskfs-dkms provided by lustre-ldiskfs-dkms-2.15.5-1.el8.noarch 
from lustre-server
  - package lustre-ldiskfs-dkms-2.15.5-1.el8.noarch from lustre-server 
conflicts with lustre-all-dkms provided by lustre-all-dkms-2.15.5-1.el8.noarch 
from lustre-server

I don't understand why we should get this error....since we are strictly trying 
install the lustre using dkms,

we also have the following kernel devel and kernel header packages :


kernel.x86_64                               4.18.0-553.5.1.el8_10
kernel-core.x86_64                          4.18.0-553.5.1.el8_10
kernel-debuginfo-common-x86_64.x86_64       4.18.0-553.5.1.el8_lustre
kernel-devel.x86_64                         4.18.0-553.5.1.el8_10
kernel-headers.x86_64                       4.18.0-553.5.1.el8_10



As far as I understood the Lustre Server 2.15.5 supports for kernel

  *   RHEL 8.10 (4.18.0-553.5.1.el8_10)

So there should not be any problem while installing the lustre-2.15.5-1.el8

Looking forward to your reply,

With sincerest regards


Abhilasha Pati

_______________________________________________
lustre-discuss mailing list
[email protected]
http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org

Reply via email to