lucy24 - 2:50 am on Feb 3, 2012 (gmt 0)
Setting aside the possibility that google has just got the hiccups and everything will be back to normal tomorrow (which really does happen)...
Also, the robots.txt is called http://mysite.com/robots.txt.
OK, now you understand why the Forums rules insist on the specific form "example.com". Everything else turns into a clickable link and nobody can see what you typed. Or, in this case, figure out what you mean. I really hope you don't mean that the actual name of the file is, in full,
?! (and let's stipulate that the final . just sneaked in for grammatical reasons) Not that that would cause the problems you describe; it would just be as if robots.txt didn't exist at all.
But that brings up one easy thing to check: Have you looked in the "crawler access" area of GWT? That's where you can see the robots.txt they're currently working from, and feed in random URLs to see if the googlebot can get in.
Also look in the Parameters area, if the site uses them. It's in the same general area as "crawler access".
Whoops! Gotta go. When I went to gwt to make sure I was calling things by the right names, I found a batch of new probably-spurious links from a probably-garbage site near the top of the list. Off to investigate.