I have seen no comment on the support list ( where it has been mentioned) from the developers about the apparent habit of recent Freenet builds of discarding all the contents of the datastore whenever fred is stopped and restarted. If it is deliberate, I would observe that it is a nuisance routing-wise, and could be catastrophic if, for any reason, a large proportion of major Freenet nodes had to be restarted on the same day. (We can all see how that might happen!)
Is there any possibility of developing a datastore repair tool? If each item of data is stored as a discrete, de-encryptable, contiguous record, it should be possible to develop a tool which checks the integrity of each item and reconstructs the store metadata around the undamaged items. This would deal not only with datastore bugs, but also the vagaries of user, OS and hardware, which are likely to quite often damage even a bugless store. If the re-initialisation of the store is only apparent (i.e. an artefact of the "environment" page) I apologise. If it is an accident, rather than a deliberate attempt to avoid DSBs, then perhaps it could be corrected. Many thanks for your attention, -- Roger Hayter _______________________________________________ devl mailing list [EMAIL PROTECTED] http://hawk.freenetproject.org/cgi-bin/mailman/listinfo/devl
