From 339c63f0c8cd4374f6fa26484498eb6fa91b7bca Mon Sep 17 00:00:00 2001 From: Yawning Angel Date: Sun, 17 Aug 2014 17:11:03 +0000 Subject: Massive cleanup/code reorg. * Changed obfs4proxy to be more like obfsproxy in terms of design, including being an easy framework for developing new TCP/IP style pluggable transports. * Added support for also acting as an obfs2/obfs3 client or bridge as a transition measure (and because the code itself is trivial). * Massively cleaned up the obfs4 and related code to be easier to read, and more idiomatic Go-like in style. * To ease deployment, obfs4proxy will now autogenerate the node-id, curve25519 keypair, and drbg seed if none are specified, and save them to a JSON file in the pt_state directory (Fixes Tor bug #12605). --- common/replayfilter/replay_filter.go | 147 +++++++++++++++++++++++++++++++++++ 1 file changed, 147 insertions(+) create mode 100644 common/replayfilter/replay_filter.go (limited to 'common/replayfilter/replay_filter.go') diff --git a/common/replayfilter/replay_filter.go b/common/replayfilter/replay_filter.go new file mode 100644 index 0000000..95cc5d6 --- /dev/null +++ b/common/replayfilter/replay_filter.go @@ -0,0 +1,147 @@ +/* + * Copyright (c) 2014, Yawning Angel + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * + * * Redistributions of source code must retain the above copyright notice, + * this list of conditions and the following disclaimer. + * + * * Redistributions in binary form must reproduce the above copyright notice, + * this list of conditions and the following disclaimer in the documentation + * and/or other materials provided with the distribution. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" + * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE + * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR + * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF + * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS + * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN + * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) + * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE + * POSSIBILITY OF SUCH DAMAGE. + */ + +// Package replayfilter implements a generic replay detection filter with a +// caller specifiable time-to-live. It only detects if a given byte sequence +// has been seen before based on the SipHash-2-4 digest of the sequence. +// Collisions are treated as positive matches, though the probability of this +// happening is negligible. +package replayfilter + +import ( + "container/list" + "encoding/binary" + "sync" + "time" + + "github.com/dchest/siphash" + + "git.torproject.org/pluggable-transports/obfs4.git/common/csrand" +) + +// maxFilterSize is the maximum capacity of a replay filter. This value is +// more as a safeguard to prevent runaway filter growth, and is sized to be +// serveral orders of magnitude greater than the number of connections a busy +// bridge sees in one day, so in practice should never be reached. +const maxFilterSize = 100 * 1024 + +type entry struct { + digest uint64 + firstSeen time.Time + element *list.Element +} + +// ReplayFilter is a simple filter designed only to detect if a given byte +// sequence has been seen before. +type ReplayFilter struct { + sync.Mutex + + filter map[uint64]*entry + fifo *list.List + + key [2]uint64 + ttl time.Duration +} + +// New creates a new ReplayFilter instance. +func New(ttl time.Duration) (filter *ReplayFilter, err error) { + // Initialize the SipHash-2-4 instance with a random key. + var key [16]byte + if err = csrand.Bytes(key[:]); err != nil { + return + } + + filter = new(ReplayFilter) + filter.filter = make(map[uint64]*entry) + filter.fifo = list.New() + filter.key[0] = binary.BigEndian.Uint64(key[0:8]) + filter.key[1] = binary.BigEndian.Uint64(key[8:16]) + filter.ttl = ttl + + return +} + +// TestAndSet queries the filter for a given byte sequence, inserts the +// sequence, and returns if it was present before the insertion operation. +func (f *ReplayFilter) TestAndSet(now time.Time, buf []byte) bool { + digest := siphash.Hash(f.key[0], f.key[1], buf) + + f.Lock() + defer f.Unlock() + + f.compactFilter(now) + + if e := f.filter[digest]; e != nil { + // Hit. Just return. + return true + } + + // Miss. Add a new entry. + e := new(entry) + e.digest = digest + e.firstSeen = now + e.element = f.fifo.PushBack(e) + f.filter[digest] = e + + return false +} + +func (f *ReplayFilter) compactFilter(now time.Time) { + e := f.fifo.Front() + for e != nil { + ent, _ := e.Value.(*entry) + + // If the filter is not full, only purge entries that exceed the TTL, + // otherwise purge at least one entry, then revert to TTL based + // compaction. + if f.fifo.Len() < maxFilterSize && f.ttl > 0 { + deltaT := now.Sub(ent.firstSeen) + if deltaT < 0 { + // Aeeeeeee, the system time jumped backwards, potentially by + // a lot. This will eventually self-correct, but "eventually" + // could be a long time. As much as this sucks, jettison the + // entire filter. + f.reset() + return + } else if deltaT < f.ttl { + return + } + } + + // Remove the eldest entry. + eNext := e.Next() + delete(f.filter, ent.digest) + f.fifo.Remove(ent.element) + ent.element = nil + e = eNext + } +} + +func (f *ReplayFilter) reset() { + f.filter = make(map[uint64]*entry) + f.fifo = list.New() +} -- cgit v1.2.3