[Azure] DataSourceAzure: send failure signal on Azure datasource failure

[johnsonshi]

Proposed Commit Message

DataSourceAzure: send failure signal on Azure datasource failure

On systems where the Azure datasource
is a viable platform for crawling metadata,
cloud-init occasionally encounters fatal
irrecoverable errors during the crawling
of the Azure datasource.

When this happens, cloud-init crashes,
and Azure VM provisioning would fail.
However, instead of failing immediately,
the user will continue seeing provisioning
for a long time until it times out with
"OS Provisioning Timed Out" message.

In these situations, cloud-init should
report failure to the Azure datasource
endpoint indicating provisioning failure.
The user will immediately see provisioning
terminate, giving them a much better
failure experience instead of pointlessly
waiting for OS provisioning timeout.

Additional Context

Additional context in the proposed commit message.

Test Steps

Test: No regression for pre-existing report ready workflow

• Create an Azure VM from the latest Ubuntu 18.04 LTS image.
• Apply this PR's patch.
• Capture the Azure VM into a custom image.
• Successful deployment from this captured custom image.

Test: Report failure to Azure upon Azure crawl metadata fatal exception

• Create an Azure VM from the latest Ubuntu 18.04 LTS image.
• Apply this PR's patch.
• Modify DataSourceAzure's crawl_metadata and add a raise Exception before the return statement. Tthis ensures a fatal crawl_metadata implementation, which mimics a fatal irrecoverable error during cloud-init's crawling of Azure metadata.
• Capture the Azure VM into a custom image.
• Deploy this custom image.
• The custom image should terminate early within 2-3 minutes with the error code OSProvisioningInternalError. It should look like the following:
  

Checklist:

• My code follows the process laid out in the documentation
• I have updated or added any unit tests accordingly
• I have updated or added any documentation accordingly

[OddBloke]

Thanks for the update! One minor comment on the code itself, and I've also reviewed the tests: no major comments, just some minor requests (and a few nits; feel free to ignore those).

[johnsonshi]

Thanks @OddBloke for the comments and NITs. I've addressed all of them in the series of commits above. To make it easier for you to review, for each of your comments above, I've linked the relevant commit that addresses the issue.

I've also built a deb with the latest changes and performed Azure deployment tests, which conform to the expected behavior (no regressions on successful deployments + report failure on intentionally buggy deployments).

[OddBloke]

Thanks for all your work here, Johnson, this looks good to me now. 👍