Fix secondary (overcloud) subnode indexes

Historically, as of Iaf43023db700d875d3569ed811069454d863cba8 we're
used to build the multinode inventory starting with subnode-2.
That's because nodepool originally was assuming subnode-1 is for
primary/undercloud. Later on there had been added the corresponding
primary/secondary groups with the pass-throw numeration starting from
0.

The old numeration scheme now breaks multi-node jobs deployed with the
new CI reproducer. Let's switch the indexing back to + 1.

Change-Id: I16e6e33179d5df9ddc93bae9046ea76ec337e0bd
Signed-off-by: Bogdan Dobrelya <bdobreli@redhat.com>
This commit is contained in:
Bogdan Dobrelya 2019-09-16 13:41:28 +02:00 committed by Bogdan Dobrelya (bogdando)
parent 68873be3ee
commit 3424850612
1 changed files with 1 additions and 1 deletions

View File

@ -47,7 +47,7 @@
- name: Add subnode to ansible inventory
with_indexed_items: '{{ nodes.stdout_lines | default([]) }}'
add_host:
name: 'subnode-{{ item.0 + 2 }}'
name: 'subnode-{{ item.0 + 1 }}'
groups: "overcloud"
ansible_host: '{{ item.1 }}'
inventory_ip: '{{ item.1 }}'