Skip to content
This repository has been archived by the owner on Jun 29, 2022. It is now read-only.

Latest commit

 

History

History
190 lines (143 loc) · 7.47 KB

command_syntax_usage.rst

File metadata and controls

190 lines (143 loc) · 7.47 KB

Command Syntax & Usage

The basic syntax is:

rho command subcommand [options]

There are four rho commands:
  • auth - for managing auth entries
  • profile - for managing profile entries
  • scan - for running scans
  • fact - to show information about the facts rho can collect
auth and profile both have five subcommands:
  • add - to create a new entry
  • edit - to modify an existing entry
  • clear - to remove any or all entries
  • show - to display a specific entry
  • list - to display one or more entries
fact has two subcommands:
  • list - to display the list of facts that can be scanned
  • hash - to hash sensitive facts within report

The complete list of options for each command and subcommand are listed in the rho manpage with other usage examples. The common options are listed with the examples in this document.

Auth Entries

The first step to configuring rho is adding auth credentials to use to connect over SSH. Each authentication identity requires its own auth entry.

rho auth add --name=server1creds --username=rho-user --sshkeyfile=/etc/ssh/ssh_host_rsa_key

Note: --password not being passed or passed as empty are considered the same thing.

SSH Key with a Passphrase

There are two ways to handle ssh keys with passphrases. If a passphrase-protected ssh key is specified in an auth and the key is not present in the ssh agent, then Rho will prompt the user for the passphrase each time it needs to use the key.

To avoid repetitive prompts, users can add their keys to the local ssh agent (as documented at http://docs.ansible.com/ansible/latest/intro_getting_started.html), which will decrypt them once and make them available to Rho.

Profiles

Then, create the profile to use for the scan. This should include a list of IP addresses or ranges, and the auth identity to use.

rho profile add --name=profile1 --hosts 1.2.3.[0:255] --auth server1creds server2creds

The hosts can be passed in as a file with all the ranges listed in newlines. Example below:

rho profile edit --name=profile1 --hosts hosts_file --auth server1creds server2creds

where hosts_file contains the ip address or ranges separated by newlines:

1.2.3.1
1.2.3.14
1.2.4.34

Scanning

The options required for a scan are the profile to use and the file path for the report. Optionally we can pass the number of Ansible forks and the facts to be collected. Finally the cache option tells rho that the profile you are providing has already been processed for connection mappings.

rho scan --profile=profile1 --reportfile=report.csv

Since rho collects the successful host auth mappings from a full scan the user doesn't have to worry about iterating through multiple auths and hosts again and again in the same profile as long as the profile hasn't changed when utilizing the cache option. For purposes of record keeping the host mapping files are always written to whenever a scan is done using a profile. When profiles are deleted the host auth mappings corresponding to the profile are renamed with prefix (DELETED PROFILE) so that they are recognizable. Every scan has a timestamp in the mappings.

As far as the auths used by the inventory of a particular scan is concerned, it is important to note that the order of the auths passed into a profile matters. A profile by definition takes in one ordering of auths and all the hosts in the profile are tested in that order with the auths before the first auths to work are picked to run the fact collection. Therefore, it's up to the user to pass in auths as per the priority they deem fit for a profile. If a non root auth is better tried first and then the root auth then the user has to pass in the auths in the order as --auth <nonroot_1> <nonroot_2> <root_1> <root_2> etc.

The output of the Ansible process is saved to $XDG_DATA_HOME/rho/scan_log by default, for debugging. This location can be changed with the --logfile flag.

Common Flags

All rho commands accept the -v flag, which increases the verbosity of rho's output. It comes in four varieties: -v, -vv, -vvv, and -vvvv, with more v's indicating more verbose output. The verbose output can be useful in debugging.

Output

The important part about a scan is the results report. By default, this contains a large amount of information about the operating system, hardware, and platform.

The output can then be configured to contain any combination of these fields by using the --facts option. The following is the format rho understands for all the facts. Some or all of these facts can be requested by either as a CLI list i.e. --facts <fact_1> <fact_2> etc or by passing in a file with a new fact on every line in the format as follows. A value of 'default' will get all the information listed above.

For further details of the command usage view the following example.

Scan User Permissions

Some of the output facts will report an error if the user used to perform the scan does not have the appropriate permissions to execute the command used to gather the targeted facts. The following set of facts require admin/root permissions to collect the facts:

  • cpu.socket_count
  • date.anaconda_log
  • date.yum_history
  • dmi.x
  • subman.x
  • virt.virt
  • virt.type
  • virt-what.x

The scan user can successful collect these values if the user is root or has the ability to perform a sudo. The following provides the necessary content for the /etc/sudoers file where scanuser represents the username used for the scan.

Cmnd_Alias SCAN = /sbin/subscription-manager, /usr/sbin/dmidecode, /usr/sbin/virt-what
scanuser ALL=NOPASSWD: SCAN

If the scan user uses a password to sudo, one can be given with the --sudo-password option to the auth add and auth edit commands. The sudo-with-password fundtionality can be tested by using the 'askpass' box in the Vagrantfile.

JBoss Lightweight and Heavyweight Scans

The JBoss facts come in two kinds. Some facts attempt to detect JBoss by looking at the running process table, or at a small list of common installation paths. These facts are run in rho by default, and can be selected using --facts jboss. They are called the "lightweight scan".

The other set of facts uses find to search the entire filesystem on the machine being scanned. These are more thorough, because they can find an EAP installation even if it is in an unusual location and not running, but they also require much more I/O and computation on the scanned machine. These are called the "heavyweight scan". They are not run by default, but can be selected with --facts all. There is a danger that the heavyweight scan could interfere with a user application running on the scanned machine, especially if that application uses a lot of CPU or does a lot of I/O.

The distinction is meant to allow a tradeoff between performance and completeness. A reasonable approach would be to run the lightweight scan first, see if the results make sense, and decide whether to run the heavyweight scan on a host-by-host basis.

Programs on Remote Machines

Besides standard Unix utilities, some rho fact collectors depend on specific programs being installed on the machines being scanned. The complete list is at remote programs.