Only guessing here but for 2 of you to have the same/similar problems develop over the last few days... there must be a common cause.
Hi Dave, thanks for drilling down some options! It baffles me to, you would suspect a software upgrade to be the reason but things just don't point into that direction.
Rapman = same
Yes we have the same machine from the same batch afaik
Firmware = same
Yes BUT Bogdan tried uploading all previous versions of the firmware and the problem persisted. Rules out Firmware
SD Card = same or reformatted
I tried all my cards, and reformatted in Windows the one that I have used since the beginning. No change in symptoms, rules out SD cards
working environment = same
Belgium, Serbia... Yes same

No seriously. I did not change my environment, its in my office ±20 degrees. First I thought that it was from a electrical extension cord switch that was a bit broken. But the RM is on a different outlet and I did a stress test flicking the switch repeatedly while printing and also plugging and unplugging other appliances. Nothing happened, kept printing.
so.... Software?

have you both updated skeinforge recently?
No, I don't update anymore unless I have a serious issue with the current version. I can digg up the version tomorrow but I doubt Skeinforge being the problem but will double check
is this only happening to newly created g-code files?
does a file you created a couple of weeks ago print?
does the standard minimug g-code file work? (downloaded from BfB not one you created)
New files, old files, happens to all of them. And I must say that at the moment I cant print anymore at all because of this and it feels like the error pops up quicker and quicker. A small print might work in a few tries but anything longer then 15-20 min. will likely fail somewhere. I will double check with the BfB files and some of my first gcode files.
Have you checked through the g-code... is anything different to what you would normally expect?
Dave
Havent checked yet, the stops happen at random points in the file and both in new and old files. So pretty unlikely that it is in the Gcode.
So all together is seams unlikely that the reason is software. I can imagine something with the environment although that sounds far fetched but as far fetched as a simultaneous malfunctioning of our same gen. boards. There is the case of Apples first gen. Airport express base stations that were likely to fail after 12-13 months.