2.4 KiB
Fuzzing LND
The fuzz
package is organized into subpackages which are named after the lnd
package they test. Each subpackage has its own set of fuzz targets.
Setup and Installation
This section will cover setup and installation of the fuzzing binaries.
- The following is a command to build all fuzzing harnesses:
⛰ make fuzz-build
-
This may take a while since this will create zip files associated with each fuzzing target.
-
The following is a command to run all fuzzing harnesses for 30 seconds:
⛰ make fuzz-run
go-fuzz
will print out log lines every couple of seconds. Example output:
2017/09/19 17:44:23 workers: 8, corpus: 23 (3s ago), crashers: 1, restarts: 1/748, execs: 400690 (16694/sec), cover: 394, uptime: 24s
Corpus is the number of items in the corpus. go-fuzz
may add valid inputs to
the corpus in an attempt to gain more coverage. Crashers is the number of inputs
resulting in a crash. The inputs, and their outputs are logged by default in:
fuzz/<package>/<harness>/crashers
. go-fuzz
also creates a suppressions
directory
of stacktraces to ignore so that it doesn't create duplicate stacktraces.
Cover is a number representing edge coverage of the program being fuzzed.
Options
Several parameters can be appended to the end of the make commands to tune the build process or the way the fuzzer runs.
run_time
specifies how long each fuzz harness runs for. The default is 30 seconds.timeout
specifies how long an individual testcase can run before raising an error. The default is 20 seconds.processes
specifies the number of parallel processes to use while running the harnesses.pkg
specifies thelnd
packages to build or fuzz. The default is to build and run all available packages (brontide lnwire wtwire zpay32
). This can be changed to build/run against individual packages.base_workdir
specifies the workspace of the fuzzer. This folder will contain the corpus, crashers, and suppressions.
Corpus
Fuzzing generally works best with a corpus that is of minimal size while achieving the maximum coverage. go-fuzz
automatically minimizes the corpus in-memory before fuzzing so a large corpus shouldn't make a difference.
Disclosure
If you find any crashers that affect LND security, please disclose with the information found here.