By using this site, you agree to our updated Privacy Policy and our Terms of Use. Manage your Cookies Settings.
429,077 Members | 2,160 Online
Bytes IT Community
+ Ask a Question
Need help? Post your question and get tips & solutions from a community of 429,077 IT Pros & Developers. It's quick & easy.

Question about Making a Diskless Python Enviroment

P: n/a
Hi everyone
We are making a disk less workstation environment which support
python.We use PXE DHCP TFTP to boot up disk less system.There are
three choices as i know to set up a python environment for diskless
system.

First Solution, Use a initrd image with everything inside as the root
file system.Diskless computer boots up with a kernel and this rootfs
image.If everything goes fine.We could use python in the ram disk.In
my situation,Target computers have 2G memory and 100M network,so the
we don't need to worry about the size of initrd image.MY QUESTIONS are
what files and libraries do i need to put into initrd image.Is there
any How-to file to help me install python into a initrd image or any
available initrd image with python installed .

Second Solution.Use a small initrd to boot up the system and use NFS
to mount those directories on Diskless server.I guess we could use /
etc/fstab file to make it.Though we boot up the system form
network ,we need to mount its local hard disk to directory tree.MY
QUESTION ARE How to mount local devices in a diskless system?Is that
done by kernel automatically or i have to configure it in fstab?

Third Solution.Use NFS ROOT file system to boot up .In this
solution.We don't need initrd any more.What we are going to do is
recompile the kernel and build those important modules inside.It is
weird that my kernel couldn't get ip form dhcp server(it can use a
static ip form kernel command line).I notice that pxe could get ip
form dhcp so it must be something wrong with the Kernel.We stop here.

The three solution above are all i could think about. Any idea about
my questions is welcomed

Thanks

Best Regard

Jul 6 '07 #1
Share this question for a faster answer!
Share on Google+

This discussion thread is closed

Replies have been disabled for this discussion.