That is not correct.  v0.7 nodes accepted a 1MB block on testnet.  The issue was more complex then just the size of the block.  By the protocol the block which was rejected by some nodes SHOULD have been accepted.  The 250kb soft limit was simply a default for block construction.  Even with it in place nodes should have accepted properly constructed blocks up to the 1MB limit.  It also appears not all v0.7 nodes were affected.  Some accepted the problem block and some didn't.  The defect/limit in BDB wasn't documented and didn't occur in all versions/platforms.   It appears the number of transactions being changed as a result of the block validation crossed some "magic code" not in the Bitcoin source code but undocumented in some implementations of BDB.

v0.7 reliance on BDB caused it to be fundamentally broken/flawed. Its actions weren't consistent with the documented protocol, the higher level source code, or anyone's understanding/expectation of what should have happened.   Nobody was saying/thinking oh yeah if you make a block with more than x transaction it will abort, cause a rollback, and result in a rejection.  It was a landmine.