Not only are various ill-conceived broadband initiatives a waste of money, and not only are they actively delaying the deployment of sensible access technologies, but they also put peoples lives at risk.
If you plan to wait for IPv6 to become “necessary” and then deploy it only on your external interfaces, usually your mail and web servers, you are likely in for a whole range of nasty surprises.
This final episode of the diskless network rescue/install system takes a look at how to customize newly installed systems on first reboot—and the surprises I’ve run into while trying to resize partitions and file systems to make use of the entire physical disk installed to.
With the diskless installer all set up, what’s missing now are installation images. In this episode I explain how to create them, what to watch out for and how to ensure they are efficient and convenient to use.
With the rescue system from the previous two episodes in place, it is almost straightforward to extend it such that it caters for the installation of pre-built disk images on the target machines.
While trying to add a disk image install feature to the PXE rescue system, I ran into a few minor problems that unexpectedly turned out to be quite a can of worms.
Setting up a diskless boot rescue system may appear complex, but it is a valuable tool in any Unix-centric environment as well as a prerequirement for network installation.
What if you need more than one Ethernet interface on a notebook? A IEEE 802.1Q tagged VLAN capable switch solves the problem.
If you think that running your Unix (/Linux, if you insist) system forever to get an uptime you can brag about, then think again: What will happen next time you actually have to reboot that machine?
In the previous episode I’ve claimed that there is a business case for IPv6. But since you are hopefully paranoid enough not to believe every claim someone makes on YouTube, here are the details and the rationale.