On bitcoin testing

Copying from a response posted to "Bitcoin software testing effort" https://bitcointalk.org/index.php?topic=117487.0 as it is relevant to a recent thread here...

Any level of testing is useful and appreciated. Various types of testing that are helpful:

* "it works" testing: Simply run the latest Release Candidate (or latest version, if released). Make sure all the basics work (for whatever definition of "basics" you desire). This is the level most accessible to casual users. * Major features testing: Develop a short checklist of must-work features, and organize volunteers to work together and go through that checklist, item by item. Test each major feature on each major platform. * Stress and fuzz testing: Attempt to "stress" the system somehow, or randomly corrupt bits of data. See what breaks. * Regression testing: Record bugs fixed, and develop automated test cases that successfully reproduce the bugs on older versions, and verify newer versions remain fixed. * Unit function testing: Rigorously exercise each C++ class to ensure it behaves as expected at a micro level. * Full peer automated testing: Automated testing of RPC and P2P functions is non-existent, because of the difficulty in doing so. Find a solution to this problem. * Data-driven tests: If possible, write software-neutral, data-driven tests. This enables clients other than the reference one (Satoshi client) to be tested. Embed tests in testnet3 chain, if possible.

The community at large can be a big help simply by doing the first item: download and run the Release Candidates and the latest version, and report any problems. Even reporting success is fine by me, for example: "Version 0.7.1 works for me on Windows 7/32-bit" posted on a forum thread.

It is always very difficult to organize any sort of testing regime with open source volunteers that come and go. Each volunteer chooses their level of involvement. Any amount of testing and test-case writing, large or small, is helpful to bitcoin.

thanks for the great reply jeff. i'm going to get a virtual machine set up on my system later tonight so at the very least, i myself can start testing.

steve - haven't heard from you in almost a week. I'd still really like to get a look at the test cases and such you set up.

Arklan

---------- As long as there is light, the darkness holds no fear. And yet, even in the deepest black, there is life. - Arklan Uth Oslin

I want to leave this world the same way I came into it: backwards and on fire. - Arklan Uth Oslin

On Tue, Oct 9, 2012 at 5:12 PM, Jeff Garzik <jgarzik@exmulti•com> wrote:

> Copying from a response posted to "Bitcoin software testing effort"
> https://bitcointalk.org/index.php?topic=117487.0 as it is relevant to
> a recent thread here...
>
> Any level of testing is useful and appreciated.  Various types of
> testing that are helpful:
>
> * "it works" testing:  Simply run the latest Release Candidate (or
> latest version, if released).  Make sure all the basics work (for
> whatever definition of "basics" you desire).  This is the level most
> accessible to casual users.
> * Major features testing:  Develop a short checklist of must-work
> features, and organize volunteers to work together and go through that
> checklist, item by item.  Test each major feature on each major
> platform.
> * Stress and fuzz testing:  Attempt to "stress" the system somehow, or
> randomly corrupt bits of data.  See what breaks.
> * Regression testing:  Record bugs fixed, and develop automated test
> cases that successfully reproduce the bugs on older versions, and
> verify newer versions remain fixed.
> * Unit function testing:  Rigorously exercise each C++ class to ensure
> it behaves as expected at a micro level.
> * Full peer automated testing:  Automated testing of RPC and P2P
> functions is non-existent, because of the difficulty in doing so.
> Find a solution to this problem.
> * Data-driven tests: If possible, write software-neutral, data-driven
> tests.  This enables clients other than the reference one (Satoshi
> client) to be tested.  Embed tests in testnet3 chain, if possible.
>
>
> The community at large can be a big help simply by doing the first
> item:  download and run the Release Candidates and the latest version,
> and report any problems.  Even reporting success is fine by me, for
> example: "Version 0.7.1 works for me on Windows 7/32-bit" posted on a
> forum thread.
>
> It is always very difficult to organize any sort of testing regime
> with open source volunteers that come and go.  Each volunteer chooses
> their level of involvement.  Any amount of testing and test-case
> writing, large or small, is helpful to bitcoin.
>
> --
> Jeff Garzik
> exMULTI, Inc.
> jgarzik@exmulti•com
>
>
> ------------------------------------------------------------------------------
> Don't let slow site performance ruin your business. Deploy New Relic APM
> Deploy New Relic app performance management and know exactly
> what is happening inside your Ruby, Python, PHP, Java, and .NET app
> Try New Relic at no cost today and get our sweet Data Nerd shirt too!
> http://p.sf.net/sfu/newrelic-dev2dev
> _______________________________________________
> Bitcoin-development mailing list
> Bitcoin-development@lists•sourceforge•net
> https://lists.sourceforge.net/lists/listinfo/bitcoin-development
>

On Tue, Oct 9, 2012 at 7:12 PM, Jeff Garzik <jgarzik@exmulti•com> wrote:

> * Data-driven tests: If possible, write software-neutral, data-driven
> tests.  This enables clients other than the reference one (Satoshi
> client) to be tested.  Embed tests in testnet3 chain, if possible.

The mention of testnet3 here reminds me to make a point: Confirmation bias is a common problem for software testing— people often over test the success cases and under-test the failure cases. This is certainly the case in Bitcoin: For example, testnet3+the packaged tests test all the branches inside the interior script evaluation engine _except_ the rejection cases.

For us failure cases can be harder to package up (e.g. can't be placed in testnet) but Matt's node-simulation based tester provides a good example of how to create a data driven test set that tests both failure cases and dynamic behavior (e.g. reorgs).

Testing of failure cases is absolutely critical for testing of implementation compatibility: The existence of a difference in what gets rejected in a widely deployed alternative node could result in an utterly devastating network split.

Generally every test of something which must succeeded should be matched by a test of something that must fail. Personally, I like to test the boundary cases— e.g. if something has an allowed range of [0-8], I'll test -1,0,8,9 at a minimum. Though reasoning trumps rules of thumb.

Confirmation bias is another reason why it's important to have a more diverse collection of testers than the core developers. People who work closely with the software have strong expectations of how the software should work and are less likely to test crazy corner cases because they "know" the outcome, sometimes erroneously.

To reinforce Jeff's list of different approaches: I've long found that each mechanism of software testing has diminishing returns the more of it you apply. So you're best off using as many different approaches a little rather than spending all your resources going as deep as possible with any one approach.

There are also some kind of testing which are synergistic: Almost all testing is enhanced enormously by combining it with valgrind because it substantially lowers the threshold of issue detection substantially (e.g. detecting bogus memory accesses which are _currently_ causing a crash for you but could). If I could only test one of "with valgrind" or "without" I'd test with every time. Sadly valgrind doesn't exist on windows and it's rather slow. Dr. Memory (http://code.google.com/p/drmemory/) may be an alternative on Windows, and there is work to port ASAN to GCC so it may be possible to mingw ASAN builds in not too long.

I've also found that any highly automatable testing (coded data driven, unit, and fuzz testing) combines well with diverse compilation, e.g. building on as many system types and architectures— including production irrelevant ones— as possible in the hopes that some system specific quark make a bug easier to detect.