Presentation is loading. Please wait.

Presentation is loading. Please wait.

Accountable systems or how to catch a liar? Jinyang Li (with slides from authors of SUNDR and PeerReview)

Similar presentations


Presentation on theme: "Accountable systems or how to catch a liar? Jinyang Li (with slides from authors of SUNDR and PeerReview)"— Presentation transcript:

1 Accountable systems or how to catch a liar? Jinyang Li (with slides from authors of SUNDR and PeerReview)

2 What we have learnt so far Use BFT replication to handle (a few) malicious servers Strengths: –Transparent (service goes on as if no faults) –Minimal client modifications Weaknesses: –Breaks down completely if >1/3 servers are bad

3 Hold (bad) servers accountable What is possible when >1/3 servers fail? –E.g. 1 out of 1 servers fail? Let us consider a bad file server –Can it delete a client’s data? –Can it leak the content of a client’s data? –Can it change a client’s data to arbitrary values? –Can it show client A garbage and claim it’s from B? –Can it show client A old data of client B? –Can it show A (op1, op2) and B (op2, op1)?

4 Hold (bad) servers accountable Lessons: –Cannot prevent a bad server from misbehaving at least once –We can do many checks to detect past misbehaviors! Useful? Fool me once, shame on you; fool me twice, shame on …

5 Case study I: SUNDR What’s SUNDR? –A (single server) network file system –Handle potential Byzantine server behavior –Can run on an un-trusted server Useful properties: –Tamper-evident Unauthorized operations will be immediately detected –Can detect past misbehavior If server drops operations, can be caught eventually

6 Ideal File system semantics Represent FS calls as fetch/modify operations – Fetch: client downloads new data – Modify: client makes new change visible to others Ideal (sequential) consistency: –A fetch reflects the sequence of modifications that happen before it –Impossible when the server is malicious A is only aware of B’s latest modification via the server Goal: –get as close to ideal consistency as possible

7 Strawman File System A Modify f2 sig3 A Modify f1 sig1 B Fetch f4 sig2 A Modify f2 sig3 B Fetch f2 sig4 A Modify f1 sig1 B Fetch f4 sig2 A Modify f1 sig1 B Fetch f4 sig2 A Modify f1 sig1 B Fetch f4 sig2 sig1 sig2 A Modify f2 sig3 File server A: echo “A was here” >> /share/aaa B: cat /share/aaa userA userB

8 Log specifies the total order A Modify f1 sig1 B Fetch f4 sig2 A Modify f2 sig3 B Fetch f2 sig4 A Modify f1 sig1 B Fetch f4 sig2 A Modify f2 sig3 The total order: Log A ≤ Log B iff Log A is prefix of Log B A’s latest log: B’s latest log: Log A Log B

9 Detecting attacks by the server A Modify f2 sig3 A Modify f1 sig1 B Fetch f4 sig2 B Fetch f2 sig3 A: echo “A was here” >> /share/aaa A Modify f1 sig1 B Fetch f4 sig2 A Modify f1 sig1 B Fetch f4 sig2 A Modify f1 sig1 B Fetch f4 sig2 A Modify f2 sig3 A B B: cat /share/aaa(stale result!) File server

10 A Modify f1 sig1 B Fetch f4 sig2 B Fetch f2 sig3b A Modify f1 sig1 B Fetch f4 sig2 A Modify f2 sig3a A’s log and B’s log can no longer be ordered: Log A ≤ Log B, Log B ≤ Log A A’s latest log: B’s latest log: Detecting attacks by the server Log A Log B sig1 sig2 sig3a

11 What Strawman has achieved  High overhead, no concurrency Tamper-evident A bad server can’t make up ops users didn’t do Achieves fork consistency A bad server can conceal users’ ops from each other, but such misbehavior could be detected later

12 Fork Consistency: A tale of two worlds File Server A’s view B’s view … …

13 Fork consistency is useful Best possible alternative to sequential consistency –If server lies only once, it has to lie forever Enable detections of misbehavior: – users periodically gossip to check violations – or deploy a trusted online “timestamp” box

14 SUNDR’s tricks to make strawman practical 1.Store FS as a hash tree –No need to reconstruct FS image from log 2.Use version vector to totally order ops –No need to fetch entire log to check for misbehavior

15 Trick #1: Hash tree h0h0 h1h1 h3h3 h2h2 h4h4 h6h6 h5h5 h7h7 h9h9 h8h8 h10h11 h12 D0 D1 D2 D3 D4 D5D6 D7D8 D9 D10D11 D12 Key property: h0 verifies the entire tree of data

16 Trick #2: version vector Each client keeps track of its version # Server stores the (freshest) version vector –Version vector orders all operations E.g. (0, 1, 2) ≤ (1, 1, 2) A client remembers the latest version vector given to it by the server –If new vector does not succeed old one, detect order violation! E.g. (0,1,2) ≤ (0,2,1)

17 SUNDR architecture block server stores blocks retrievable by content-hash consistency server orders all events Untrusted Network Untrusted Network consistency server block server SUNDR server-side SUNDR client SUNDR client userA userB

18 SUNDR data structures: hash tree Each file is writable by one user or group Each user/group manages its own pool of inodes – A pool of inodes is represented as a hash tree

19 Hash files Blocks are stored and indexed by hash on the block server data1 Metadata H(data1) H(data2) H(iblk1) data2 data3 data4 H(data3) H(data4) iblk1 20-byte File Handle i-node

20 Hash a pool of i-nodes Hash all files writable by each user/group From this digest, client can retrieve and verify any block of any file 220-byte3 4 i-table i-node 2 i-node 3 i-node 4 20-byte digest i-num

21 SUNDR FS 220-byte 3 4 digest 220-byte 3 4 digest Superuser: UserA: SUNDR State How to fetch “/share/aaa”? /: Dir entry: (share, Superuser, 3) Lookup “/” /share: Dir entry: (aaa, UserA, 4) Lookup “/share” Fetch “/share/aaa” digest UserB: … 2 3 4 digest GroupG:

22 SUNDR data structures: version vector Server orders users’ fetch/modify ops Client signs version vector along with digest Version vectors will expose ordering failures

23 Version structure Each user has its own version structure (VST) Consistency server keeps latest VSTs of all users Clients fetch all other users’ VSTs from server before each operation and cache them We order VST A ≤ VST B iff all the version numbers in VST A are less than or equal in VST B VST A Signature A A Digest A A - 1 B - 1 G - 1 VST B ≤ Signature B B Digest B A - 1 B - 2 G - 2

24 Update VST: An example Consistency Server B A A-0 B-0 A: echo “A was here” >> /share/aaa B: cat /share/aaa Dig A A A-1 B-1 Dig A A A-1 B-1 Dig A A A-1 B-1 Dig A A A-0 B-1 Dig B B A-1 B-2 Dig B BA-1 B-2 Dig B BA-0 B-1 Dig B B VST A ≤ VST B

25 Detect attacks Consistency Server B A: echo “A was here” >> /share/aaa B: cat /share/aaa (stale!) A A-1 B-1 Dig A A A-0 B-0 Dig A A A-0 B-1 Dig A B A-1 B-1 Dig A A A-0 B-1 Dig B B A-0 B-2 Dig B B A-0 B-2 Dig B B A’s latest VST and B’s can no longer be ordered: VST A ≤ VST B, VST B ≤ VST A ≤ A-0 B-0 Dig A A

26 Support concurrent operations Clients may issue operations concurrently How does second client know what vector to sign? –If operations don’t conflict, can just include first user’s forthcoming version number in VST –But how do you know if concurrent operations conflict? Solution: Pre-declare operations in signed updates –Server returns latest VSTs and all pending updates, thereby ordering them before current operation –User computes new VST including pending updates –User signs and commits new VST

27 Concurrent update of VSTs Consistency Server A B update: B-2 update: A-1 A: echo “A was here” >>/share/aaa B: cat /share/bbb A-0 B-0 Dig A A A-1 A-0 B-1 Dig B B A-0 B-1 Dig B B A-1B-2 A-0 B-0 Dig A A VST A ≤ VST B A-1 B-1 Dig A A A-1 B-1 Dig A A A-1 B-2 Dig B B A-1B-2 A-1 B-2 Dig B B A-1B-2

28 SUNDR is practical Seconds

29 Case study II: PeerReview [SOSP 07 ]

30 Motivations for PeerReview Large distributed systems consist of many nodes Some nodes become Byzantine –Software compromise –Malicious/careless administrator Goal: detect past misbehavior of nodes –Apply to more general apps than FS

31 Challenges of general fault detection How to detect faults? How to convince others that a node is (not) faulty?

32 Overall approach Fault := Node deviates from expected behavior Obtain a signature for every action from each node If a node misbehaves, the signature works a proof-of-misbehavior against the node

33 Can we detect all faults? No –e.g. Faults affecting a node's internal state Detect observable faults –E.g. bad nodes send a message that correct nodes would not send A A X C C 1001010110 0010110101 1100100100 0

34 Can we always get a proof? No –A said it sent X B said A didn‘t C: did A send X? Generate verifiable evidence: –a proof of misbehavior (A send wrong X) –a challenge (C asks A to send X again) Nodes not answering challenges are suspects A A X B B C C ? I sent X! I never received X! ?!

35 Treat each node as a deterministic state machine Nodes sign every output message A witness checks that another node outputs correct messages –using a reference implementation and signed inputs PeerReview Overview

36 M PeerReview architecture All nodes keep a log of their inputs & outputs –Including all messages Each node has a set of witnesses, who audit its log periodically If the witnesses detect misbehavior, they –generate evidence –make the evidence avai-lable to other nodes Other nodes check evi- dence, report fault A's log B's log A A B B M C C D D E E A's witnesses M

37 PeerReview detects tampering A B Message Hash chain Send(X) Recv(Y) Send(Z) Recv(M) H0H0 H1H1 H2H2 H3H3 H4H4 B's log ACK What if a node modifies its log entries? Log entries form a hash chain Signed hash is included with every message  Node commits to having received all prior messages Hash(log)

38 PeerReview detects inconsistency What if a node –keeps multiple logs? –forks its log? Witness checks if signed hashes form a single chain H3'H3' Read X H4'H4' Not found Read Z OK Create X H0H0 H1H1 H2H2 H3H3 H4H4 OK "View #1" "View #2"

39 Module B PeerReview detects faults Witness audits a node : –Replay inputs on a reference implementation of the state machine –Check outputs against the log Module A Module B =? Log Network Input Output State machine if ≠ Module A

40 PeerReview‘s guarantees Faults will be detected (eventually) –If a node if faulty: Its witness has a proof of misbehavior Its witness generates a challenge that it cannot answer –If a node is correct No proof of misbehavor It can answer any challenge

41 PeerReview applications App #1: NFS server Tampering with files Lost updates App #2: Overlay multicast Freeloading Tampering with content App #3: P2P email Denial of service Dropping emails Metadata corruption Incorrect access control

42 PeerReview’s performance penalty Cost increases w/ # of witnesses per node W Baseline 1 2 3 4 5 100 80 60 40 20 0 Avg traffic (Kbps/node) Number of witnesses Baseline traffic

43 What have we learnt? Put constraints on what faulty servers can do –Clients sign data, bad SUNDR server cannot fake data –Clients sign version vector, bad server cannot hide past inconsistency Fault detection –Need proof of misbehavior (by signing actions) –Use challenges to distinguish slow nodes apart from bad ones


Download ppt "Accountable systems or how to catch a liar? Jinyang Li (with slides from authors of SUNDR and PeerReview)"

Similar presentations


Ads by Google