Thomas White:
> Not sure if anyone knows of this, but the following might be a good
> solution to scale the current mirror system and is used by a number of
> other open source projects.
>
> http://mirrorbrain.org
Thanks for your input,
We already considered using mirrorbrain and it is already mentioned on
our blueprint for a new mirror infrastructure:
https://tails.boum.org/blueprint/HTTP_mirror_pool
Our current plan is to write a very minimal custom PHP script.
The downside of mirrorbrain here is that it's quite big (probably too
big for what we want to do) and it's not in Debian, not really
authenticated, audited, etc.
Still, if we want to further improve our custom script in the future,
for example to include load balancing or geolocation, we might end up
reinventing partly the wheel when more complex tools such as mirrorbrain
already exist.
Anyway, I don't feel like I'm the right person to best argue or decide
on this.
--
sajolida
