Ole Holm Nielsen wrote: > Patrick Lists wrote: >>>> I fail to see how your repository problem is related to my Kickstart PXE/NFS >>>> problem. We can install CentOS 6 (no Kickstart) without problems using PXE. >>>> It's the NFS-mounting of the Kickstart-file which fails. What I'd love to learn >>>> is how others have made NFS/Kickstart work... >>> Is CentOS 6 assuming NFSv4 by default perhaps? >> According to the RHEL 6.0 Release Notes it is: >> >> "Mounting a file system via NFS now defaults to NFSv4." > > Yes, this sounds like the correct explanation! Our RHEL5.6 PXE/TFTP/NFS > server is certainly configured for NFSv3. I haven't explored the NFSv4 > approach though. > > How can we tweak the CentOS6 Kickstart installation to explicitly > request an NFSv3 service in the PXE config file: > > kernel CentOS-6-i386/vmlinuz > append load_ramdisk=1 initrd=CentOS-6-i386/initrd.img network > ks=nfs:130.226.86.4:/u/rpm/kickstart/ks-centos-6-clean-i386.cfg > > To rephrase my question: Where do we find documentation for the vmlinuz > append flags shown above, in particular documentation of the ks=... flags? > > Hopefully there'll be some way to force an NFSv3 mount in stead of the > default NFSv4? Or perhaps we need to use http: and stop using nfs: with > CentOS6? I would suggest using NFSv4 on all of your servers. It needs only one port open on the firewall, tcp 2049 and I think it is better solution security wise (if I remember correctly). I started to use NFSv4 from the start on the CentOS 5.x and never had any problems. Why not upgrade when opportunity presented it self. -- Ljubomir Ljubojevic (Love is in the Air) PL Computers Serbia, Europe Google is the Mother, Google is the Father, and traceroute is your trusty Spiderman... StarOS, Mikrotik and CentOS/RHEL/Linux consultant