+ Configuration details:
Are they still using
secondary tags (WATO)?
Did they use service groups? do they still?
Is this stuff assigned based on rules?
+ Configuration management:
Ideas about re-use of standardized config - having one config that only loads hostname lists and is 100% same on all boxes?
Are they backing the config with a SCM? automated checkin/checkout using hooks?
Is it driven from any central management tool (cfengine, puppet, bcfg, chef...ksh stacks)
Are they using WATO / Distributed WATO?
+ Config dynamics: Example:
we can set the variable monitoring_host in our config to influence what the parent scan considers the root node.
monitoring_host = "192.168.19.24"
But we can also do something dynamic using sys.uname or os.system.
import sys, os
monitoring_host = os.system("hostname")
+ "Data pump":
Does the config load any dynamic data?
+ Tinkering and Fudging of data:
Did they ever consider stuff like service_descriptions.update()
Did they have to do much modifications over what a general OMD setup has?
+ Advanced features:
How commonly using averaging, trending?
Do they have different trending sets?
+ ALI3Ns constant questions about --scan-network