I'm running into some difficulty running aufs over NFS using a 2.6.27.10 kernel and the latest CVS fetch of aufs. I have a Linux 2.6.27.10 server that does DHCP, PXE, and NFS for a diskless x86 Linux cluster. The nodes fetch their aufs-enabled kernel and initramfs over TFTP, mount the server's root filesystem over NFS, and then use aufs to provide a series of overlays above the root filesystem. Here's my branches:
1. The entire server root filesystem, mounted over NFS and set as a read-only branch. 2. A subdirectory for the entire cluster, mounted as a read-only branch. 3. A subdirectory for that specific node, based on its IP address, and mounted as a read-write branch. Branch #3 is the only read-write branch, so it's where all the writes end up. The problem is that about 75% of the time I boot a diskless node, it hangs between when the init scripts complete, and when the init process spawns a login prompt. Rebooting several times gets a node all the way to the init prompt. This is a stock Linux kernel with only aufs applied. I've also applied the lhash, splice, sec_perm, and ksize patches. I also use this same setup to provide diskless nodes with just a RAM disk (no NFS), and this works fine, so I suspect it's an NFS problem. Let me know what kind of additional information would be useful to help in debugging. Thanks in advance for any help, -- -- Skylar Thompson ([email protected]) -- http://www.cs.earlham.edu/~skylar/
signature.asc
Description: OpenPGP digital signature
------------------------------------------------------------------------------
