[RIPE Atlas Ambassadors] LED status for RIPE probes

Hi,

On 2015-12-25 8:35, Nishal Goburdhan wrote:
>> On 25 Dec 2015, at 14:53, Denys Stroebel <denyss at faircape.co.za> wrote:
>>
>> As a side note to reinforce random unplugging of the probes, I have done so with my first non-Ambassador probe and managed to brick it. The general advice I give when handing out is simply to leave the probes for an extended period of time and be patient.
> 
> +1. 

Something that can help here to understand what's going on:

When we prepare (initialise) the probes, we put the latest available
firmware on them. But firmware development is ongoing, so by the time a user
gets the probe -- whether from us or from an ambassador -- chances are we
have released a newer one.

Now, the very first thing that happens after a probe is plugged in (assuming
successful network connections) is that it's checking for firmware updates,
so in most cases it immediately upgrades itself. This, with all the steps
involved, can take a while: depending on network speeds, number of retries,
etc. it can be anything between 10-30 minutes. Then the probe needs to
reconnect, which can also be flakey in remote areas.

All in all, the first actual connection is not guaranteed to be immediate.

Now, we're going to add some features to the probe page to guide the users
about what's going on. At a minimum stuff like "DNS works" and such.
Hopefully, we can add more useful guidance later, like "your probe seems to
be plugged in now, stand by!" or "your probe is now upgrading its firmware".

>> For the power concern, we are usually lucky enough for participants to have USB ports on their home routers which works well.
> 
> ...and even when you think you have things running, its a great idea to still ping the atlas team if something does't look right.
> 
> the probe at my home is connected to the usb port on a 2011/12 mac mini
> and, after 3 months or so after initial setup, i started receiving random
> messages about the probe being offline (even when external pings showed
> otherwise).  some digging by the dev team (thanks robert!) showed that
> the mac usb port wasn't delivering enough power.  plugin the second port
> of the usb power cable, into another usb port on the mini, and its been
> silent ever since...

Indeed, unreliable power seems to be the number 1 issue nowadays. It's a
hard one to solve, and I'm beginning to think that we should advise *all*
users to use *both* type A ends of the cable if they can...

Regards,
Robert

> --n.
> _______________________________________________
> RIPE-Atlas-Ambassadors mailing list
> RIPE-Atlas-Ambassadors at ripe.net
> https://www.ripe.net/mailman/listinfo/ripe-atlas-ambassadors
>