Trouble searching the node

[kalapakim]

Here is the search I'm applying:

sql_ip = -> { search(:node, "id:primarysql*") }
​

And I try to put:

puts "IP Address: #{sql_ip.call[0]['ipaddress']}"
​
​

This is the node that I am searching against

{
  "id": "primarysql-Windows",
  "chef_environment": "_default",
  "automatic": {
    "ipaddress": "10.12.235.71",
    "platform": "windows",
    "fqdn": "MTPCTSCID012.consilio.com",
    "recipes": [
      "relativity-scaled-automation::setup_sql"
    ]
  },
  "normal": {
  },
  "run_list": [
    "recipe[relativity-scaled-automation::setup_sql]"
  ]
}

​

What is returned is

IP Address: ["fe80::a816:9ab7:f164:108e%12", "10.0.2.15"]
​

which is wrong all together...wthat IP address is not correct?

so I try a different search...

sqlID = []
search(:node, "id:primarysql*").each do |server|
  sqlID.push(server['ipaddress'])
end
puts "FQDN: #{sqlID}"

and this is what is returned...

FQDN: [["fe80::a816:9ab7:f164:108e%12", "10.0.2.15"], "10.12.235.71"]

the second item in the list is correct, but what is the first item? where is that coming from?

I searched my project and there is no such information

[mwrock]

could you post your .kitchen.yml or a gist of it. One thing to check is on the node that is issuing the chef search, look in its %temp%\kitchen\nodes directory. That holds the json data that chef_zero is using to conduct the search.

If the node that is performing the search is looking for itself. In other words, if you are performing the search on the primarysql-* box, my guess is that you are using Virtualbox and this is reporting back the ohai data for its default network interfaces. This is usually not compatible with nodes actually wishing to use the IPAddress to connect to the box. You can fix this by eiting the ohai data and also ensuring yo uare using a private_network. I have blogged on this exact topic at http://www.hurryupandwait.io/blog/multi-node-test-kitchen-tests-and-working-with-vagrant-nat-addressing-with-virtualbox.

Now the second search result you mention above is surprising. It sounds like its finding 2 nodes for the id id:primarysql* I'd be curious to know if it in fact has 2 node json files for the same node.

[mwrock]

closing due to inactivity