A new version of the count-mvs tool is located at IBM Fix Central.
This is a script that is meant to be ran on a Qradar Console and give a count of the MVS (Managed Virtual Servers) for the deployment.
To run the script you need to copy it to the Console, make it executable
chmod +x countMVS.py, and then run the script with ./countMVS.py. If your
deployment is a multi-tenant environment (i.e. it receives events from multiple
domains) then you will be prompted for authentication. You can choose between
running the script as the Admin user, or using an Authorized Service with the
Admin Security Profile. After you choose the script will prompt you for either the
Admin user's password or the Authorized Service's security token before executing.
The script works by querying the database to compile a list of log sources that have sent events to Qradar within the last 24 hours. From that list of log sources we build a list of the machines those sources are running on. If there are any log sources that are not explicitly running on a machine we consider an MVS, for example a SaaS service, then we remove those log sources from the list. In this version of the script we have a list of log sources we consider as not running on an MVS, we plan on improving our algorithm for determining log sources are/are not running on an MVS. After we filter out those log sources we ensure that no machine in the list is referenced more than once. Finally we output the list, a map of MVS IPs to a list of each of their log sources, to a file and return the final count of MVS the deployment has received events from in the last 24 hours
There are additional considerations for multi-tenant environments, as the script needs to consider the case where a given IP could refer to multiple different machines across multiple domains, which would affect our MVS count. If the script detects that a log source has received events from multiple domains then it will try to compile a list of those domains by issuing calls to the Ariel API in order to inspect the events that were sent to that log source in the last 24 hours. These API calls are why the script requires authentication in this case. If we find a log source has been receiving events from the same IP from different domains then that means the IP represents multiple machines, and this will be reflected in the MVS count. We include the list of domains for each log source in the csv output file. In addition to the MVS count for the entire deployment, in this case we will also provide a separate MVS count for each domain.
The script generates the MVScount.csv file. The header of the file is a list of the
IP of each MVS, and the columns are the list of log sources for each MVS.