CloudKitty bootstrap fails when using internal TLS

[Pierre Riteau]

When InfluxDB is behind HAProxy's internal TLS, CloudKitty fails to bootstrap its InfluxDB database with the following error:

TASK [cloudkitty : Creating Cloudkitty influxdb database] ***************************************************************************************************************************
fatal: [controller01 -> controller01]: FAILED! => changed=false
  action: influxdb_database
  msg: ('Connection aborted.', RemoteDisconnected('Remote end closed connection without response'))

When Elasticsearch is used, a similar problem happens during TASK [cloudkitty : Running Cloudkitty bootstrap container]. cloudkitty-api.log shows:

requests.exceptions.SSLError: HTTPSConnectionPool(host='<INTERNAL_VIP>', port=9200): Max retries exceeded with url: /cloudkitty (Caused by SSLError(SSLCertVerificationError(1, '[SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get local issuer certificate (_ssl.c:1131)')))

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.opendev.org/c/openstack/kolla-ansible/+/866598

[joek-office (joek-office)]

Hello priteau,
what have to be done to release/merge the bug fix.
I'm new in such cases but in my opinion, the bug fix is suitable. There's only thing, could we make it possibly configurable over a new configuration parameter in the ansible/roles/cloudkitty/defaults/main.yml file?
Can't we use the value of cloudkitty_influxdb_insecure_connections also on deployment by standard?