network reporters generate WARNING in local stage

This bug was originally filed in Launchpad as [LP: #1742494](https://bugs.launchpad.net/cloud-init/+bug/1742494)

Launchpad details

affected_projects = []

assignee = None

assignee_name = None

date_closed = None

date_created = 2018-01-10T17:18:28.147844+00:00

date_fix_committed = None

date_fix_released = None

id = 1742494

importance = medium

is_complete = False

lp_url = https://bugs.launchpad.net/cloud-init/+bug/1742494

milestone = None

owner = smoser

owner_name = Scott Moser

private = False

status = confirmed

submitter = smoser

submitter_name = Scott Moser

tags = []

duplicates = []

_Launchpad user **Scott Moser(smoser)** wrote on 2018-01-10T17:18:28.147844+00:00_

Any of the network based reporters (MAAS) that are configured in system config (/etc/cloud/cloud.cfg*) will attempt to be used during the local phase of cloud-init (cloud-init-local.service).

Unsurprisingly, these will fail.

Unfortunately, they log WARNING to the logs like:
2018-01-10 15:41:20,842 - handlers.py[WARNING]: failed posting event: start: init-local/check-cache: attempting to read from cache [check]

This is clearly not a WARNING that should be paid attention to.

It'd be good to only log debug or in some other way handle this better.

The end result is just false positives for WARNING in /var/log/cloud-init.log on nodes deployed by maas.

Start by tracing the network-based reporters used during the cloud-init-local.service local phase, with handlers.py[WARNING] as the log entry point. Check how failures are reported when networking is unavailable; done means expected local-phase failures no longer create WARNING entries in /var/log/cloud-init.log while genuine reporter problems remain visible.

Written by the indexing model from the issue text.