5 Patterns to Improve Application Robustness

A presentation by @stuherbert  for @GanbaroDigital Five Patterns For Improving
Application Robustness

@GanbaroDigital A Little Bit Of Personal History

@GanbaroDigital Where am I coming from? Why do I care
about robustness?

@GanbaroDigital https://ﬂic.kr/p/fQXzyo NQS

@GanbaroDigital https://ﬂic.kr/p/a6Yf7a CPMS

@GanbaroDigital https://laracasts.com/ Visual Debt

@GanbaroDigital Lessons I’ve learned from nearly a quarter of a
century of making things robust.

@GanbaroDigital I am still learning.

@GanbaroDigital What Is Robustness?

@GanbaroDigital Let’s make sure we’re all talking about the same
thing.

@GanbaroDigital ?? ?? What do you think robustness is?

@GanbaroDigital ?? ?? Have you heard of “happy path” programming?

@GanbaroDigital

@GanbaroDigital The “happy path” only deals with correctness under perfect
conditions.

@GanbaroDigital This is a world without robustness.

@GanbaroDigital “ Correctness bugs are logic ﬂaws that produce unacceptable
results.

@GanbaroDigital The happy path works when correctness bugs have been
found and resolved.

@GanbaroDigital “ Robustness code deals with all the things that
shouldn’t happen, but inevitably do.

@GanbaroDigital

@GanbaroDigital x

@GanbaroDigital “ Robustness bugs stop perfectly-good logic from working.

@GanbaroDigital “ Without robustness, your code runs on “There But
For The Grace of God”

@GanbaroDigital

@GanbaroDigital “ Add robustness up-front or deal with the consequences
in production.

@GanbaroDigital ?? ?? How do you make your applications more
robust?

@GanbaroDigital Here are ﬁve patterns that have been very successful
for me.

@GanbaroDigital These patterns work best when used together.

@GanbaroDigital I’m going to illustrate each pattern using very simple
examples.

@GanbaroDigital “ Robustness patterns are micro-patterns.

@GanbaroDigital Pattern 1

@GanbaroDigital ?? ?? What do you use returned values for?

@GanbaroDigital

@GanbaroDigital “ Nobody checks returned data 100% reliably.

@GanbaroDigital “ Nobody does anything 100% reliably.

@GanbaroDigital Sometimes, they don’t check it safely.

@GanbaroDigital if (!$position) vs if (false === $position)

@GanbaroDigital Sometimes, they don’t check it at all.

@GanbaroDigital Eliminate the need to check return values.

@GanbaroDigital

@GanbaroDigital ?? ?? What do you do when a robustness
problem has been detected?

@GanbaroDigital

@GanbaroDigital Use exceptions when normal ﬂow cannot continue.

@GanbaroDigital ?? ?? What about returning NULL?

@GanbaroDigital NULL is not an error report

@GanbaroDigital Stop using true, false and NULL as error responses

@GanbaroDigital Return NULL only when “no data” is a valid
response.

@GanbaroDigital Throw an exception when “no data” is not a
valid response.

@GanbaroDigital Succeed, Except When You Fail

@GanbaroDigital 1. Build APIs that succeed 2. Throw exceptions when
they can’t 3. Use NULL only when it’s a valid data value

@GanbaroDigital ?? ?? What breaks perfectly-good logic?

@GanbaroDigital

@GanbaroDigital “ Type mismatches break perfectly good logic.

@GanbaroDigital Standard solution: check incoming data before using it.

@GanbaroDigital PHP7 extended type-hinting to help.

@GanbaroDigital

@GanbaroDigital Type-hinting isn’t a silver bullet.

@GanbaroDigital Type-hinting only tells you if something is the right
shape. Type-hinting does not validate your data.

@GanbaroDigital

@GanbaroDigital ?? ?? What would you do about this?

@GanbaroDigital Checks and Requirements can help.

@GanbaroDigital

@GanbaroDigital Unsupported outputs can break logic.

@GanbaroDigital

@GanbaroDigital Unsupported state can break logic.

@GanbaroDigital

@GanbaroDigital Unhandled errors and uncaught exceptions  ultimately crash code.

@GanbaroDigital x

@GanbaroDigital “ Your code can crash and burn, or it
can fail gracefully. Pick one.

@GanbaroDigital Step 1: Know that you need to fail.

@GanbaroDigital Step 2: Fail deliberately.

@GanbaroDigital Fail Deliberately

@GanbaroDigital 1. Check everything before you use it 2. Check
both type and condition 3. Don’t propagate bad data

@GanbaroDigital ?? ?? When a method cannot continue, what do
you do?

@GanbaroDigital

@GanbaroDigital ?? ?? How often do you catch exceptions in
your code?

@GanbaroDigital “ Nobody does anything 100% reliably.

@GanbaroDigital ?? ?? How often do you catch exceptions thrown
by someone else’s code?

@GanbaroDigital “ No-one catches exceptions they didn’t throw themselves.

@GanbaroDigital ?? ?? When code must fail, can a library
know what must happen next?

@GanbaroDigital A library, by its nature, cannot know the context
it is being used in.

@GanbaroDigital A library can decide that it must fail. Failure
is a technical situation.

@GanbaroDigital The app must decide what to do on failure.
What to do is a matter of policy.

@GanbaroDigital “ Policy belongs in application code, not libraries or
packages.

@GanbaroDigital

@GanbaroDigital ?? ?? What if we injected policy from the
app into the library?

@GanbaroDigital

@GanbaroDigital Inversion of Control: library decides if to fail, app
decides how to fail.

@GanbaroDigital For example: throw an exception containing the HTTP response

@GanbaroDigital

@GanbaroDigital $onFailure can provide a default RETURN value instead.

@GanbaroDigital Tell Me How To Fail

@GanbaroDigital 1. Support a $onFailure callback 2. Let it throw
any exceptions 3. Let it produce a return value

@GanbaroDigital ?? ?? What state are things left in when
your code fails?

@GanbaroDigital “ Your code can crash and burn, or it
can fail gracefully. Pick one.

@GanbaroDigital Step 1: Know that you need to fail.

@GanbaroDigital Step 2: Fail deliberately.

@GanbaroDigital Step 3: Don’t leave a mess behind when you
fail.

@GanbaroDigital If you don’t do this, your application stops working
until someone manually repairs the damage done.

@GanbaroDigital ?? ?? What needs to be considered?

@GanbaroDigital Basic principle: reset as if the operation never happened.

@GanbaroDigital Post-Failure Reset 1. PHP session contents 2. Changes to
your databases / data stores 3. Changes on remote systems

@GanbaroDigital ?? ?? How do you avoid a mess at
the caller’s end?

@GanbaroDigital Tell them that something went wrong, and it’s not
their fault. HTTP 5xx!

@GanbaroDigital Make it safe to retry actions that produce HTTP
5xx errors.

@GanbaroDigital “Idempotent actions can be safely attempted multiple times, but
only take effect once.

@GanbaroDigital “ Idempotent is HARD.

@GanbaroDigital Don’t Crash Land

@GanbaroDigital 1. Reset state after a robustness failure 2. Return
an error indicator to your API caller 3. Make operations idempotent

@GanbaroDigital ?? ?? How do you know that you have
a robustness problem in production?

@GanbaroDigital Write Errors and Exceptions to your log ﬁles.

@GanbaroDigital ?? ?? How do you know what the problem
is?

@GanbaroDigital I’ve lost count of the number of problems that
had to be debugged in Production.

@GanbaroDigital Make sure your logs tell you what you need
to ﬁx the bug.

@GanbaroDigital “ If you can’t solve a Production issue,  ship
extra checks and logging to gather better evidence.

@GanbaroDigital Logs don’t write themselves. Install a default error handler.
Install a default exception handler.

@GanbaroDigital Fix the bugs that appear in your logs. Wipe
them out. All of them.

@GanbaroDigital You never know what’s hiding behind the bugs that
you can see.

@GanbaroDigital Know That   Code Has Failed

@GanbaroDigital 1. Log errors and exceptions 2. Logging must be
production-safe 3. Logs must be useful! 4. Fix every bug that appears in your logs

@GanbaroDigital In Summary

@GanbaroDigital 5 Patterns To Improve Robustness 1. Succeed, Except When
You Fail 2. Fail Deliberately 3. Tell Me How To Fail 4. Don’t Crash Land 5. Know That Code Has Failed

@GanbaroDigital A Revisit …

@GanbaroDigital https://laracasts.com/ Visual Debt

@GanbaroDigital It must have been created for a genuine reason.

@GanbaroDigital The tips in the original video reduce immediate cognitive
load. That must be attractive to programmers at the beginning of their experience.

@GanbaroDigital In my experience: they also reduce or eliminate proven
robustness and correctness practices.

@GanbaroDigital Those problems may not appear in small, disposable apps.

@GanbaroDigital Those problems will appear in larger, or longer-lived projects.

@GanbaroDigital Larger projects require code from different people to work
together. That never happens perfectly.

@GanbaroDigital Longer-lived projects have to be resilient to changes to
code and changes to the team.

@GanbaroDigital “ Resilience against change and time doesn’t happen accidentally.

@GanbaroDigital On larger projects, both robustness and correctness save time,
money, and reputation.

@GanbaroDigital “ No-one can afford to produce zero-defect code.

@GanbaroDigital I’ll be looking at that @GlasgowPHP in July 2017.

Thank You Any Questions? A presentation by @stuherbert  for @GanbaroDigital

5 Patterns to Improve Application Robustness

5 Patterns to Improve Application Robustness

More Decks by Stuart Herbert

Other Decks in Programming

Featured

Transcript