Hello, my cluster, made of a front-end server + a number of diskless nodes, was no longer working last week after an upgrade to Slackware 14.2. The diskless nodes were unable to load the root file system via NFS. It turned out that the problem was due to a mismatch between the default behaviour of the user-space part of the NFS stack (nfs-utils) and the kernel-space part (linux/fs/nfs/nfsroot.c). The former, by recently changed default, runs protocols V3 and V4 but no longer V2, whereas the latter still uses V2 as the default. Solved by changing /etc/dhcpd.conf so as to advice diskless nodes to user V3 instead of the default. I think it is not a good idea, that the two parts of the NFS stack use incompatible defaults. So I'd suggest to fix this. Regards, Giuseppe Ciaccio http://www.dibris.unige.it/ciaccio-giuseppe DIBRIS - Universita' di Genova via Dodecaneso 35 16146 Genova, Italy giuseppe.ciaccio@xxxxxxxx phone +390103536663 fax +390103536699 --------------------------------------------------------------------------- -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
