Analyzers¶

Options¶

All the analyzers can have optional options.

analyzer: ...
query: ...
options:
  retry_times: ...
  retry_interval: ...
  retry_exponential_backoff: ...
  timeout: ...
  ignore_error: ...
  parallel: ...

Also the following analyzers can have pagination options.

options:
  pagination_interval: ...
  pagination_limit: ...

Retry Times¶

retry_times (integer) is a number of times of retry when something goes wrong. Optional. Defaults to 3. Configurable via RETRY_TIMES environment variable.

Retry Interval¶

retry_interval (integer) is an interval in seconds between retries. Optional. Defaults to 5. Configurable via RETRY_INTERVAL environment variable.

Retry Exponential Backoff¶

retry_exponential_backoff (bool) controls whether to do exponential backoff. Optional. Defaults to true. Configurable via RETRY_EXPONENTIAL_BACKOFF environment variable.

Timeout¶

timeout (integer) is an HTTP timeout in seconds. Optional.

Ignore Error¶

ignore_error (bool) controls whether to ignore an error or not. Optional. Defaults to false.

Mihari uses fail-fast approach. For example, if the first Shodan analyzer fails, the second Censys analyzer will not be triggered because Mihari will fail before it.

queries:
  - analyzer: shodan
    query: ip:1.1.1.1
  - analyzer: censys
    query: ip:8.8.8.8

You can set ignore_error option to make it fault-tolerant.

queries:
  - analyzer: shodan
    query: ip:1.1.1.1
    options:
      ignore_error: true
  - analyzer: censys
    query: ip:8.8.8.8

Parallel¶

parallel (bool) controls whether to allow parallel execution or not. Optional. Defaults to false. Configurable via ANALYZER_PARALLELISM environment variable.

Pagination Interval¶

pagination_interval (integer) is an interval in seconds between pagination. Optional. Defaults to 0. Configurable via PAGINATION_INTERVAL environment variable.

Pagination Limit¶

pagination_limit (integer) is an limit for pagination. Optional. Defaults to 100. Configurable via PAGINATION_LIMIT environment variable.

In the worst case, if something wrong with Mihari or a service, Mihari can drain API quota by doing pagination forever. pagination_limit is a safety valve for that. A number of pagination is limited as pagination_limit times.