netcheck: network diagnoses from blackbox traces yanyan zhuang, eleni gessiou nyu poly, university...

13
NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

Upload: thomasine-martin

Post on 11-Jan-2016

222 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

NetCheck: Network Diagnoses from Blackbox Traces

Yanyan Zhuang, Eleni GessiouNYU Poly,

University of British Columbia

Page 2: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

Motivation

• Ping– Only reachability

• Wireshark– Applicationor network-specific knowledge

• Network Config Analysis– Detailed network knowledge – HW + config

• …

Page 3: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

Goal

• Diagnose network problems in large and complex applications

• Without modifying the original application

Page 4: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

Syscall trace from strace

Page 5: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

Challenges

• Accuracy: ambiguity in order reconstruction

• Efficiency: exploring an exponential space of possible orderings

• Network complexity: diagnosing issues in real networks

Page 6: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

NetCheck Overview

Finding deviations from the model of the network (Deutsch’s Fallacies)network is reliable, latency is zero, etc.

Page 7: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

Priority & Dependency of syscalls

Page 8: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

e.g.

Page 9: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

Fault diagnoses

Page 10: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

Rules summary

Example of rule (1): when a client is behind a NAT, (i) the client uses a private IP, (ii) the peer socket address in server’s accept is not the client’s IP

Page 11: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

Evaluation

• Reproduce reported bugs from bug trackers (Python, Apache, Ruby, Firefox, etc.)– A total of 71 bugs– Correct analysis of 95.7% bugs

• Twenty faults observed in practice on a live network(Seattle Testbed)– 90% of cases correctly detected

Page 12: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

Runtime performance overhead

Best: O(l)

Worst: O(nl)

Page 13: NetCheck: Network Diagnoses from Blackbox Traces Yanyan Zhuang, Eleni Gessiou NYU Poly, University of British Columbia

Conclusion

• Derives a plausible global traces ordering as a proxy for the ground truth

• Uses a model of expected and simple network behavior to identify and diagnose unexpected behavior