Do not raise on missing status blob; reduce amount of logging for vms…

[narrieta]

The status blob url should not be a required property in vmSettings. Instead of raising an error, not it is initialized to None (to preserve the behavior established by extensionsConfig).

Also, since some errors are more common than expected, reduced the max number of errors from 5/hour to 3/hour (telemetry) and 1/hour (local log), and reduced the amount of details in parsing errors.

[codecov]

Codecov Report

Merging #2492 (21960ae) into release-2.7.0.0 (a21d9dd) will increase coverage by 0.01%.
The diff coverage is 80.95%.

@@                 Coverage Diff                 @@
##           release-2.7.0.0    #2492      +/-   ##
===================================================
+ Coverage            71.92%   71.94%   +0.01%     
===================================================
  Files                  101      101              
  Lines                15028    15030       +2     
  Branches              2386     2387       +1     
===================================================
+ Hits                 10809    10813       +4     
+ Misses                3735     3734       -1     
+ Partials               484      483       -1     

Impacted Files	Coverage Δ
...protocol/extensions_goal_state_from_vm_settings.py	81.67% <69.23%> (+0.87%)	⬆️
...inuxagent/common/protocol/extensions_goal_state.py	88.46% <100.00%> (-0.08%)	⬇️
azurelinuxagent/common/protocol/wire.py	81.61% <100.00%> (+0.04%)	⬆️
azurelinuxagent/common/version.py	70.58% <100.00%> (ø)
azurelinuxagent/ga/exthandlers.py	85.75% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b4fa41b...21960ae. Read the comment docs.

[nagworld9]

where is this file used?

[narrieta]

removed

[nagworld9]

duplicate check, I think it should be self._status_upload_blob_type

[nagworld9]

can you add failure scenario if you could?

[narrieta]

thanks; fixed typo. the test already exists: test_create_from_vm_settings_should_assume_block_when_blob_type_is_not_valid

[nagworld9]

Actually, this typo carried from develop and mentioned test case not covering None case that's why we didn't catch it before.

[nagworld9]

Approving it unless you want to add blob_type is None case, thanks

[narrieta]

checking for invalid is good enough. it is unrealistic to check for every possible invalid value: missing, empty string, mixed case, garbage, etc.

the existing test captures the desired behavior: an invalid value defaults to Block

[nagworld9]

Not agreed 100%, At least we should have a test case that was handled it in the code. My intention is we would have caught this typo if we have test case where status_upload_blob_type is None (not some invalid value) and assert BlockBlob type.

[narrieta]

Yes, we would have captured this typo, but we did not know we were going to make this typo. As another example: should we add a test for mixed case that could catch a possibly defective piece of code that doesn't handle that case?

[nagworld9]

I agree but my only point in general, we were expected certain behavior when it is None. That behavior should have been tested.

Beyond this, we fixed the typo I'm good with that :)