During the eclipse (and in fact already a few days in advance), I've
been logging signal strengths over the entire 0 to 29 MHz spectrum
using my WebSDR system in Enschede, The Netherlands.
As was to be expected, the eclipse's effect was very clear on medium
wave broadcast signals, and also on longwave, with, as Roelof already
pointed out, Iceland's broadcast coming in very strongly.
Surprisingly, on one broadcast station I saw a dip in propagation rather
than enhancement, namely the 216 kHz longwave transmitter in southern
France. Could that be due to destructive interference between skywave
Preliminary results, including a wideband waterfall display, are now on
http://wwwhome.cs.utwente.nl/~ptdeboer/ham/eclipse2015/ ; but more
analysis still needs to be done.