I imagine the entire generation of the block, along with the transaction should be available in the debug.log held in the datadir?

I have mined for over 24 hours, and not seen that issue.   The nodes in the default dimecoind.conf did seem unavailable last night.   I wonder if that was due to load, poor maintenance timing (so soon after a launch and concurrently with other nodes, or perhaps was related to a DOS attack of sorts.

The seednodes went offline yesterday for a time, people were complaining of no block chain.    Does anyone know - what happens if that occurs?   You would think if IRC is configured (it wasnt in the config by default) then all the clients have a place to check in and get more p2p peers.    I didnt notice that the windows client opened uPNP ports either... so wondering how viable connectivity from MOST Windows QT clients were as p2p nodes for inbound connections.    Because most people probably NOT setting up port-forwarding that early in the coins launch...    So if seed nodes are down, and cannot create inbound connections to other clients.... new clients... then you have "a copy" of the block chain... then you are in an island - and perhaps with other users... (maybe on your own).    At the point you are in network with transactions flying around it - would blocks not start getting generated?   and meanwhile in another island - more blocks (duplicate) are being created in that island..?

Then the question is if block 10,000 is created in island A and block 10,000 is created in island B.... then what happens when island A and B are merged?   Is it feasible this occurred?

Just speculation - not sure HOW the protocol is established to handle this split brain scenario - what weird things might happen...

-d