Hi,
We recommend setting up a small environment but scalable. For example:
- Server 1: Wazuh manager master
- Server 2: Wazuh manager worker
- Server 3: Wazuh dashboard
- Server 4: Wazuh indexer
In this way, you can monitor your deployment and scale it if necessary. For example, checking these variables:
- /var/ossec/var/run/wazuh-analysisd.state: events_dropped
- /var/ossec/var/run/wazuh-remoted.state: discarded_count
These two variables should be zero if the environment is working properly. If it is not the case, additional nodes can be added to your deployment (wazuh managers).
Regarding the hardware:
- Wazuh usually needs more CPU than RAM. It is recommended 8 cores and 4 GB
- The indexer usually needs more RAM than CPU: It is recommended 8 cores and 16 GB
- The storage will depend on the size of your events and if you enable archives or only alerts. Keep in mind that the average event size is 1KB and the ratio event:alert is 10:1.
I hope it helps.