1

I was reading info on this Wiki. It says...

With the release of the Raspberry Pi 2 Model B and its ARMv7-based BCM2709 processor, it is now possible to run Ubuntu directly on the Raspberry Pi.

Note that the information on this page currently only applies to the Raspberry Pi 2 and 3, not the original Raspberry Pi. The Raspberry Pi 3 does not (yet) work with official Ubuntu images out of the box, but unofficial images are available.

That page is stamped

last edited 2016-08-09 13:18:45 by ssweeny

But the Ubuntu download page says

Ubuntu 16.04.1 LTS includes support for the very latest ARM-based server systems powered by certified 64-bit processors

I know there is an "Ubuntu MATE" OS that works on a Raspberry Pi, but I would rather use official Ubuntu images since they offer long term support. In the quote above, they reference an "ARMv7" but I just bought a Raspberry Pi 3 and it is a "64-bit quad-core ARMv8 CPU", which sounds like the official Ubuntu images should work.

My question is, does Ubuntu's latest download work on a Raspberry Pi 3 with a "64-bit quad-core ARMv8 CPU"?

EDIT: The reason I ask is because I would like to run MongoDB. Judging from this discussion (jira.mongodb.org/browse/SERVER-1811), recent support has been offered

ARM support will apply for Ubuntu 16.04 64-bit on ARMv8 or newer.

But it's only supported for the release of Ubuntu I highlighted above. So it sounds like I'm stuck waiting for one of the followin A) Raspberry Pi support for Ubuntu 16.04 64-bit on ARMv8 B) Raspberry Pi support for MongoDB v3.4

I tried the MongoDB v2.1.1 for the Raspberry Pi, but it's too out dated and doesn't offer the supporting functions I'm looking for. I should also note, if you haven't realized this already, I'm running a database server on the Pi so I don't use GUI or GPIO features on the pi.

4
  • IMO, I would just consider running Mongo in a Docker container.
    – thibmaek
    Commented Sep 6, 2016 at 21:55
  • @Thibmaekelbergh I'm not familiar with this option. What is a "Docker container"? Can I run the latest MongoDB versions? Will it work on a Raspberry Pi native OS, like Raspbian Jessie Lite? Commented Sep 6, 2016 at 22:10
  • It's a concept based off lcx (linux containers) that is becomming somewhat the defacto standard for seperating app services. It will run the latest MongoDB versions but you'd need to compile a new kernel for Raspbian AFAIK. resin.io currently runs Docker on Raspberry Pi but in Arch instead of Raspbian. Some good places to start would be docker.com and blog.hypriot.com (a precompiled kernel and image for using Docker on Raspberry Pi). This is a completely new approach to what you want though and might be slightly opinionated, but IMO it's the most performant and easiest solution.
    – thibmaek
    Commented Sep 6, 2016 at 22:17
  • @Thibmaekelbergh the resin.io base images have raspbian/debian, alpine linux, and fedora versions, can use any of them. Looking around, both fedora and alpine has mongodb 3.2.7 or newer, so that could work. On the other hand, it depends on the use case of ThatsRightJack, whether running things in docker would worth the trouble or not. If running a server, maybe a project like "ubuntu-pi-flavour-maker.org" would work too, though they are shipping with "old" mongo as well (v2.6.10 as I've checked now). I wonder what's the function/feature that ThatsRightJack is looking for?
    – Gergely
    Commented Sep 7, 2016 at 10:00

1 Answer 1

2

No! The standard install will NOT boot.

It may be possible to customise the installation to use the Pi firmware. This has a number of downsides:- you will be lacking support for GPIO access and the processor on the Pi will struggle with the demands of the GUI.

Even the standard MATE install (which is better matched to the Pi processor) will not work without Pi firmware.

The linked images are experimental, and if you want to try you won't lose much.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.