Fuzzing or fuzz testing is an automated software testing technique that involves providing invalid, unexpected, or random data as inputs to a computer program. A fuzzing target is defined for running the tests. These targets are defined in the test/fuzz directory and run using the instructions given below:

• Install the package from https://github.com/mdempsky/go114-fuzz-build using go get command. This is the link for the fuzzing tool that uses libfuzzer for its implementation purposes. I set it up and made a sample function to run the fuzzing tests. In these steps following are the important steps

  a.  Run below command to build the file for the function by giving its location.

  mkdir output
  cd output
  go114-fuzz-build -o yaml_FuzzNewIPPoolManager.a -func FuzzNewIPPoolManager ../test/fuzz/

  b. Run below command to make C binary files for the same function.

  clang -o yaml_FuzzNewIPPoolManager  yaml_FuzzNewIPPoolManager.a -fsanitize=fuzzer

  c. Run the fuzzer as below

  ./yaml_FuzzNewIPPoolManager

• Execution: Here is an example for execution of some test outputs:

  [63 32 45 58 119 50 9 119 50 9 119 50 9 45 58
  50 9 119 9 119 50 9 91 13 45 58]
  &{TypeMeta:{Kind: APIVersion:}
  ObjectMeta:{Name: GenerateName: Namespace: SelfLink: UID:
  ResourceVersion: Generation:0
  CreationTimestamp:0001-01-01 00:00:00 +0000 UTC DeletionTimestamp:<nil> DeletionGracePeriodSeconds:<nil>
  Labels:map[] Annotations:map[] OwnerReferences:[]
  Finalizers:[] ManagedFields:[]} Spec:{ClusterName:<nil> Pools:[]
  PreAllocations:map[] Prefix:0 Gateway:<nil> DNSServers:[]
  NamePrefix:} Status:{LastUpdated:<nil> Allocations:map[]}}
  #1028919 REDUCE cov: 4365 ft: 18232 corp:
  2348/250Kb lim: 1070 exec/s: 3535 rss: 463Mb L:
  118/1060 MS: 1 EraseBytes-
  

  Here first line contains the byte array provided by libfuzzer to the function we then typecast it to our struct of IPPool using JSON. Unmarshal and then run the function. It keeps on running until it finds a bug.