PipelineConf 2018: Poised To Adapt

Continuous Delivery's Relationship to Resilience Engineering

Transcript

1. Poised To Adapt Continuous Delivery’s Relationship to Resilience Engineering John Allspaw Principal, Adaptive Capacity Labs
2. example #1 rm -rf $PATHNAME
3. @@ -1,2 +1,2 @@ -<!-- Status: Ok --> +<!-- Status: OK --> Showing 1 changed file with 1 addition and 1 deletion. index.html example #2
4. all work is contextual
5. Adaptive Capacity Labs
6. Velocity Conference 2009
7. http://bitly.com/AllspawThesis
8. http://stella.report Year-long project Researchers analyzed 3 incidents, at: Six themes •Postmortems as re-calibration •Blameless v. sanctionless after action actions •Controlling the costs of coordination •Visualizations during anomaly management •Strange Loops •Dark Debt
9. What You Are In For 1. Resilience Engineering: a field and a community 2. A cognitive systems perspective of the CD/CI community 3. Poise/potential/capacity to adapt 4. Some (hopefully) thought-provoking questions
10. Resilience Engineering • A field of study that emerged largely from Cognitive Systems Engineering, early 2000s. • 7 symposia over 12 years
11. Resilience Engineering Community is largely made up of practitioners and researchers from…. working in these domains… Aviation/ATM Rail Maritime Space Surgery Power Plants Intelligence Agencies Law Enforcement Mining Construction Explosives Firefighting Anesthesia Pediatrics Power Grid & Distribution Military Agencies Software Engineering Human Factors & Ergonomics Cognitive Systems Engineering Cybernetics Complexity Science Engineering* Psychology Sociology Ecology Safety Science
12. Some of the cast of characters David Woods CSEL/OSU Shawna Perry Univ of Florida Emergency Medicine Dr. Richard Cook Anesthesiologist Researcher Ivonne Andrade Herrera SINTEF Erik Hollnagel Univ of S. Denmark Anne-Sophie Nyssen University de Liege Johan Bergström Lund University Sidney Dekker Griffith University Asher Balkin CSEL/OSU Laura Maguire CSEL/OSU
13. Sample of Research Experiences in Fukushima Dai-ichi nuclear power plant in light of resilience engineering Unmanned Aircraft Systems in (Inter)national Airspace: Resilience as a Lever in the Debate Sociotechnical Networks for Power Grid Resilience: South Korean Case Study Limits on adaptation: Modeling Resilience and Brittleness in Hospital Emergency Departments
14. Books
16. externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results
17. externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results macro descriptions externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results
18. code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results
19. code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools system system framing doing code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results
20. deploy organization/ “monitoring” Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing code deploy organization/
21. code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results The Work Is Done Here Your Product Or Service The Stuff You Build and Maintain With
22. code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results
23. code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results Copyright © 2016 by R.I. Cook for ACL, all rights reserved ack: Michael Angeles http://konigi.com/tools/ What matters. Why what matters matters. code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools above the line below the line Why is it doing that? What needs to change? What does it mean? How should this work? What’s it doing? What does it mean? What is happening? What should be happening What does it mean? Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results goals purposes risks cognition actions interactions speech gestures clicks signals representations artifacts the line of representation individuals have unique models of the “system” Copyright © 2016 by R.I. Cook for ACL, all rights reserved ack: Michael Angeles http://konigi.com/tools/ What matters. Why what matters matters. code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools above the line below the line Why is it doing that? What needs to change? What does it mean? How should this work? What’s it doing? What does it mean? What is happening? What should be happening What does it mean? Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results goals purposes risks cognition actions interactions speech gestures clicks signals representations artifacts the line of representation individuals have unique models of the “system” observing inferring anticipating planning troubleshooting diagnosing correcting modifying reacting
24. Copyright © 2016 by R.I. Cook for ACL, all rights reserved ack: Michael Angeles http://konigi.com/tools/ What matters. Why what matters matters. code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools above the line below the line Why is it doing that? What needs to change? What does it mean? How should this work? What’s it doing? What does it mean? What is happening? What should be happening What does it mean? Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results goals purposes risks cognition actions interactions speech gestures clicks signals representations artifacts the line of representation individuals have unique models of the “system” Copyright © 2016 by R.I. Cook for ACL, all rights reserved ack: Michael Angeles http://konigi.com/tools/ What matters. Why what matters matters. code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools above the line below the line Why is it doing that? What needs to change? What does it mean? How should this work? What’s it doing? What does it mean? What is happening? What should be happening What does it mean? Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results goals purposes risks cognition actions interactions speech gestures clicks signals representations artifacts the line of representation individuals have unique models of the “system” observing inferring anticipating planning troubleshooting diagnosing correcting modifying reacting
25. What matters. Why what matters matters. code deploy organization/ encapsulation “monitoring” Why is it doing that? hat needs to change? What does it mean? How should this work? What’s it doing? What does it mean? What is happening? What should be happening What does it mean? Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing go purp ris cogn act intera spe ges cli sig represe What matters. Why what matters matters. code deploy organization/ encapsulation “monitoring” Why is it doing that? hat needs to change? What does it mean? How should this work? What’s it doing? What does it mean? What is happening? What should be happening What does it mean? Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing go purp ris cogn act intera spe ges cli sig represe observing inferring anticipating planning troubleshooting diagnosing correcting modifying reacting
26. code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results Time …and things are changing here things are changing here…
27. Resilience is something that a system does, not what a system has.
28. “Resilience is an expression of how people, alone or together, cope with everyday situations – large and small – by adjusting their performance to the conditions. An organisation’s performance is resilient if it can function as required under expected and unexpected conditions alike (changes/disturbances/opportunities).” Hollnagel, Erik. Safety-II in Practice: Developing the Resilience Potentials
29. –David Woods (2015) “Resilience is sustained adaptive capacity.”
30. Resilience is the story of the outage that didn’t happen.
31. Continuous Delivery?
32. small and frequent? risk? shorten lead time?
33. #include <stdio.h> int main(void) { printf("Hello World\n"); return 0; } 6 lines, 79 characters tr G-t F-s<<<Ifmmp\ Xpsme 1 line, 26 characters
34. • canaries are different than feature flags • feature flags are different than ramp-ups • ramp-ups are different than db schema changes • network, hardware, “front-end” “fault injection/chaos” • etc. etc. etc. a change is not a change is not a change
35. Hints that the situation is more complicated than usually described… Deploys get different evaluations based on their perceived risk. Freedom to deploy is sometimes restricted. Risk hedging is common. Conventions are everywhere. Dr. Richard Cook, Velocity Conf 2016 Santa Clara, CA
36. all work is contextual
37. code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools Adding stuff to the running system Getting stuff ready to be part of the running system architectural & structural framing keeping track of what “the system” is doing code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results Context Is Constructed Here
38. Poised To Deploy 1. Knowing what the platform is supposed to do 2. Knowing how the platform works 3. What the platform’s behavior means 4. Being able to devise a change that addresses 1, 2, & 3 5. Being able to predict the effects of that change 6. Being able to force the platform to change in that way 7. Being prepared to deal with the consequences Dr. Richard Cook, Velocity Conf 2016 Santa Clara, CA
39. Stuff you can test Unknown ∞ ∞ Known complete testing is impossible credit: Noah Sussman
40. How does our software work, really? How does our software break, really? What do we do to keep it all working?
41. “it depends”
42. how to discover what happens “above the line”?
43. incidents (outages, degradations, breaches, accidents, near-misses, “glitches”, untoward/unexpected events, etc.)
44. what makes incidents interesting & valuable?
45. code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools above the line below the line externally sourced code (e.g. DB) results the using world delivery technology stack internally sourced code results code repositories macro descriptions testing/validation suites code code stuff meta rules scripts, rules, etc. test cases code generating tools testing tools deploy tools organization/ encapsulation tools “monitoring” tools above the line below the line externally sourced code (e.g. DB) results delivery technology stack internally sourced code results incidents as… drivers of software design - “incidents of yesterday inform the architectures of tomorrow” - incidents “below the line” drive changes “above the line" - staffing, budgets, planning, roadmaps, etc. - shape the design of new components, subsystems, architectures
46. 5/6/2010 - “Flash Crash” - loss of $1 trillion in market value in <10min 3/23/2012 - BATS IPO - systems issue halted the exchange’s own IPO 5/23/2012 - Facebook IPO - systems issue delayed IPO trading 8/1/2012 - Knight Capital - $461 million in 45 minutes “Regulation SCI” - tend also to give birth to new forms of regulations, policies, norms, compliance requirements, explosion of documentation, auditing, constraints, etc. - “incidents of yesterday inform the rules of tomorrow” - influence staffing, budgets, planning, roadmaps, etc. PCI-DSS 1988-1998, Visa and MasterCard reported credit card losses due to fraud of $750 million incidents as… motivators for policy
47. incidents tend to focus our attention on what matters
48. incidents help us gauge the delta between how the system works how we think the system works Δ { almost always greater than we imagine
49. What is it doing?! Why is it doing that?! What will it do next? How did it get into this state? WTF is happening? If we do Y, will it help us figure out what to do? Is it getting worse? It looks like it’s fixed…but is it…? If we do X, will it prevent it from getting worse…or make it worse? Who else should we call that can help us? Is this OUR issue, or are we BEING ATTACKED?!
50. “…nonroutine, challenging events, because these tough cases have the greatest potential for uncovering elements of expertise and related cognitive phenomena.” (Klein, Crandall, Hoffman, 2006) A family of well-worn methods, approaches, and techniques Cognitive task/work analysis Process tracing Conversation analysis Critical decision method Critical incident technique more… research validates these opportunities
52. incident 54 minutes start resolve
53. 12 minutes 54 minutes start resolve detect incident
54. 20 minutes 73 minutes 12 minutes 54 minutes start resolve detect start detect resolve incidents
55. 12 minutes 54 minutes start resolve detect 20 minutes 73 minutes start detect resolve 5 25 minutes start detect resolve incidents
56. incidents 12 minutes 54 minutes start resolve detect 20 minutes 73 minutes start detect resolve 5 25 minutes start detect resolve 135 minutes 100 minutes start detect resolve
57. incidents 12 minutes 54 minutes start resolve detect 20 minutes 73 minutes start detect resolve 5 25 minutes start detect resolve 135 minutes 100 minutes start detect resolve minutes
58. incidents minutes
59. incidents minutes jan feb mar apr may jun
60. incidents minutes jan feb mar apr may jun
61. incidents minutes jan feb mar apr may jun
62. incidents minutes jan feb mar apr may jun
63. “Resilience is an expression of how people, alone or together, cope with everyday situations – large and small – by adjusting their performance to the conditions. An organisation’s performance is resilient if it can function as required under expected and unexpected conditions alike (changes/disturbances/opportunities).”
64. “Resilience is sustained adaptive capacity.”
65. incidents minutes jan feb mar apr may jun
66. What is it doing?! Why is it doing that?! What will it do next? How did it get into this state? WTF is happening? If we do Y, will it help us figure out what to do? Is it getting worse? It looks like it’s fixed…but is it…? If we do X, will it prevent it from getting worse…or make it worse? Who else should we call that can help us? Is this OUR issue, or are we BEING ATTACKED?!
67. incidents provide calibration about… how decisions are focused how attention flows how work is coordinated how escalation manifests the weight of time pressure the effects of uncertainty the impact of ambiguity what consequences are consequential
68. What can we learn about these… how decisions are focused how attention flows how work is coordinated how escalation manifests the weight of time pressure the effects of uncertainty the impact of ambiguity what consequences are consequential …from these? (M)TTR? (M)TTD? Frequency of incidents? Severity of incidents? Customer impact? Number of deploys? “…while there is value in the items on the right, we value the items on the left more.”
69. Thought Food • We cannot comprehensively understand how our systems behave - we continually build and revise our understandings based on (relatively sparse) signals our tech sends us. • Applying CD approaches/techniques is an implicit acknowledgement of this sparse and fleeting understanding, and represent coping strategies for this state of affairs. • Understanding activities “above the line” are basically unexplored or ignored in our industry, and this needs to change.
70. Taking Human Performance Seriously This discussion is happening… in nuclear power in air traffic control in firefighting in medicine … We need to do more than just acknowledge this - we need to embrace it.