Is it worth while writing creating a IPv6 stack for embedded systems?
worth while IPv6
Most people would tell you it's not worth while writing or creating your own IPv4 stack ...
I'm talking about an IPv6 not IPv4
I think you missed the sarcasm!
How in the hell did I, I'm Irish
Any input would be appreciated guys...
Any relevant input?
I'll attempt
IPv6 is already in some frameworks and RTOS.
8b and 16b MCU in deeply embedded systems can implement standard I/O to a 32b MCU or MPU that has IPv6.
Microchip
MPLAB Harmony
http://www.microchip.com/mplab/mplab-harmony
...
TCP/IP Network Stack and WiFi support
...
The BSD layer compatibility has been greatly improved, new dynamic socket options have been added as well as IPv6 support.
...
The Contiki OS
Contiki: The Open Source OS for the Internet of Things
...
Full IP Networking
...
The Contiki IPv6 stack, developed by and contributed to Contiki by Cisco, is fully certified under the IPv6 Ready Logo program.
...
I have friends in the internet business who claim that ipv6 is important and necessary, especially for things like IoT where there are potentially a trillion devices.
Personally, I'm not convinced - the workarounds (NAT, etc) in IPv4 seem to work really well.
How in the hell did I, I'm Irish
well that makes it official then... even Irish have off days
Because of NAT I too doubt the need for IPv6.
Think about all your own uses of IP in your house or office? Are you really messing with anything but four part IPV4 addressing? In fact I'd go as far as to say that almost everything you do/configure is probably somewhere under 192.168.x.x
Sure you may set up some DNS servers (probably 8.8.8.8 and 8.8.4.4 in fact!) with a more global IP range but even then it's still IPv4 not v6.
Maybe the Cisco's of this world and the others that provide the "backbone" of the internet concern themselves with IPv6 - but domestic use? Even more so, "embedded micro" domestic use ? I don't think so.
I have friends in the internet business who claim that ipv6 is important and necessary, especially for things like IoT where there are potentially a trillion devices.
The premise for this argument is that every single "Thing" in the Internet of "Things" needs to have its own, globally-unique, globally-addressable IP address.
Personally, I'm not convinced
Neither am I.
I think the basic premise is flawed.
I don't see the point in being able to directly address every single lightbulb in my from anywhere in the world.
It seems far more sensible that I address my house, and then the things within it are just "sub addresses" of my house - which is what NAT effectively gives us.
I don't see the point in being able to directly address every single lightbulb in my from anywhere in the world.
More to the point, I see a need to >>prevent<< direct access to every single light bulb from anywhere in the world. Do you trust the light bulb manufacturer to get security right? Or do you do the sensible thing and put it behind a routing/NATting firewall? Will you have a trillion light bulbs behind that firewall?
Or do you do the sensible thing and put it behind a routing/NATting firewall?
embedded.com
IoT security: Putting trust in the edge
October 30, 2017
...
... ARM has drawn a line in the sand for IoT security – a line that passes straight through the edge that separates IoT terminal nodes from the cloud.
IoT edge devices are emerging as a kind of catch-all device, defined variously as local data aggregators, device controllers, data/state buffers, fog computing appliances, cloud gateways, and more.
...
Designed to provide developers with a concrete roadmap to IoT security, the ARM PSA [Platform Security Architecture] combines a security use-case analysis with hardware/firmware specifications and an open-source reference implementation ...
...
FreedomBox combines firewall and some local device access with (planned) mesh networking :
Debian
FreedomBox
Introduction
https://wiki.debian.org/FreedomBox/Introduction
...
Advanced usage: Smart Home Router
...
FreedomBox provides a VPN server ... to securely access various devices at home.
...
In the future, FreedomBox intends to deliver support for alternative ways of connecting to the Internet such as Mesh networking.
...
Feck it! I think I will implement IPv6. Only problem is finding tools to test and debug development.
Thanks for the input lads.
Wm.
Is it worth while to replace the Internet?
Yes
RT
‘By the people, for the people’: Kim Dotcom to launch alternative internet
Published time: 22 Nov, 2017 10:56 Edited time: 23 Nov, 2017
https://www.rt.com/news/410606-kim-dotcom-meganet-internet/
...
Megaupload founder Kim Dotcom says he will help facilitate an unobstructed internet, free from prying eyes, through MegaNet, which will operate without IP addresses.
...
f
Zephyr Project
What is the Zephyr Project?
https://www.zephyrproject.org/what-is-zephyr/
...
Connected
...
Supports standards like 6Lowpan, CoAP, IPv4, IPv6, and NFC
...
Zephyr's Microchip and related boards are Arduino Due, SAM4S Xplained, and SAM E70 Xplained :
http://docs.zephyrproject.org/boards/boards.html
Mouser Electronics
Open Source RTOS Comes to IoT
On September 6, 2016 in Dev Tools, IoT by Lynnette Reese
https://www.mouser.com/blog/open-source-rtos-comes-to-iot
...
Recently, Intel released WindRiver’s “nano-kernel” RTOS, formerly called “Rocket,” to The Linux Foundation (LF) and was renamed “Zephyr.”
...
Edit: typo
Substratum
Decentralizing the Web
WE ARE THE FOUNDATION OF THE DECENTRALIZED WEB.
Substratum is an open-source network that allows anyone to allocate spare computing resources to make the internet a free and fair place for the entire world.
...
https://substratum.net/why-substratum/
I think that it is very useful to make systems support ipv6. Not everybody wants to provision ipv4 on every network, it's additional overhead.
While you might think there are enough private ipv4 addresses for anybody's intranet, that doesn't always happen in practice. Company mergers are always interesting, when the companies are using exactly the same address space in their internal networks.
Someone either has to change, or it all just moves to v6.
Also, v6 is more automatic, if you only use link-local scope, or if you have v6 routers, there is no need for (messy, kludgy) dhcp.
Still thinking about doing an embedded IPv6 stack.
did a brief search in it though was unsuccessful in locating what MCU the ones at Microchip are targeting for Zephyr Project.
was unsuccessful in locating what MCU the ones at Microchip are targeting for Zephyr Project.
Yes, true zephyrs have not been seen in the wild for some generations, and are generally thought to be extinct.
1939
1950ish
1980ish
When I finish my current projects I'll be starting a IPv6 embedded stack.
While i'm impressed with your ambitions
I'd say , given your current line of questions here.
A fully working ipv6 stack might be aiming a bit high At the moment.
But best of luck
/Bingo
Ps:
Have a look here
http://contiki.sourceforge.net/d...
Or if you have the ram (Arm)
https://savannah.nongnu.org/proj...
I've already done a IPv4 stack, full implementation DHCP/DNS/ARP/ICMP/UDP/TCP/IP fragmentation.
Or if you have the ram (Arm)
https://github.com/atmelcorp/atmel-software-package/tree/master/lib/lwip
not an IP replacement, an HTTP replacement
[peer-to-peer]
[archival]
[resilient]
[persistence]
IPFS aims to replace HTTP and build a better web for all of us.
...
via IPFS is the Distributed Web