Package net.i2p.router.networkdb.kademlia


package net.i2p.router.networkdb.kademlia

The network database, including implementation of floodfills. Database lookups, stores, exploration, and replies. Management of locally-generated RouterInfos and Leasesets.

  • Class
    Description
    Cache of blinding data.
     
    Ask a connected peer for his RI.
    Override to not call failed() in setMessage(), as it will be called from runJob()
    Periodically search through all leases to find expired ones, failing those keys and firing up a new search for each (in case we want it later, might as well preemptively fetch it)
    Go through the routing table pick routers that are is out of date, but don't expire routers we're actively connected to.
    Search for a particular key iteratively until we either find a value, we run out of peers, or the bucket the key belongs in has sufficient values in it.
    Deprecated.
    unused, see comments in KNDF
    Build a HandleDatabaseLookupMessageJob whenever a DatabaseLookupMessage arrives
    Create a HandleDatabaseStoreMessageJob whenever a DatabaseStoreMessage arrives
    Simple job to monitor the floodfill pool.
    The network database
    This is where we implement semi-Kademlia with the floodfills, by selecting floodfills closest to a given key for searches and stores.
    Job to flood nearby floodfill routers with our RI.
    This extends StoreJob to fire off a FloodfillVerifyStoreJob after success.
    Send a netDb lookup to a floodfill peer - If it is found, great, but if they reply back saying they dont know it, queue up a store of the key to a random floodfill peer again (via FloodfillStoreJob)
     
    Mostly replaced by IterativeLookupSelector
    This is the timeout for the whole search.
    Uunused directly, replaced by IterativeSearchJob, but still extended by SingleSearchJob.
    Try sending a search to some floodfill peers, but if we don't get a successful match within half the allowed lookup time, give up and start querying through the normal (kademlia) channels.
    Count how often we have recently flooded a key This offers basic DOS protection but is not a complete solution.
    Handle a lookup for a key received from a remote peer.
    Receive DatabaseStoreMessage data and store it in the local net db
    Ask a single peer for a single key.
    Ask the peer who sent us the DSRM for the RouterInfos...
    Slightly modified version of FloodOnlyLookupSelector.
    A traditional Kademlia search that continues to search when the initial lookup fails, by iteratively searching the closer-to-the-key peers returned by the query in a DSRM.
    This is the timeout for a single lookup, not for the whole search.
    Kademlia based version of the network database.
    Pull the caching used only by KBucketImpl out of Hash and put it here.
    Count how often we have recently received a lookup request with the reply specified to go to a peer/TunnelId pair.
    Method and class for garlic encrypting outbound netdb traffic, and sending keys and tags for others to encrypt inbound netdb traffic, including management of the ElGamal/AES tags.
    A single key and tag, for receiving a single message.
    Wrapper so that we can keep track of the key and tags for later notification to the SKM
    Track lookup fails
    Mostly unused, see overrides in FloodfillPeerSelector
    Write out keys to disk when we get them and periodically read ones we don't know about into memory, with newly read routers are also added to the routing table.
    Go through all the routers once, after startup, and refetch their router infos.
    Run periodically for each locally created leaseSet to cause it to be republished if the client is still connected.
    Search for a particular key iteratively until we either find a value or we run out of peers Note that this is rarely if ever used directly, and is primary used by the ExploreJob extension.
    Check to see the message is a reply from the peer regarding the current search
     
    Data related to a particular search
    Called after a match to a db search is found Used only by SearchJob which is only used by ExploreJob
    Ask the peer who sent us the DSRM for the RouterInfos...
    Ask a single peer for a single key.
    Fire off search jobs for random keys from the explore pool, up to MAX_PER_RUN at a time.
    Stores through this always request a reply.
    Check to see the message is a reply from the peer regarding the current store
    Tracks the state of a StoreJob
    Stores in-memory only.
    Signature verification failed because the sig type is unknown or unavailable.