We are all assuming it's the Met Office to blame.

The basic source for the "mirrors" is NOAA. To the best of my
knowledge, H+108 and H+132 were never available on internet direct
from the MetOffice, only via this tortuous route:

http://weather.noaa.gov/pub/fax/

Now NOAA hasn't been updating with the PPVN89 (the H+108 chart) and
the PPVP89 (the H+132) since 4th March. So maybe that is the problem,
not the MetOffice.

Jack