Track parser's recoverable errors at warning level

[Neurone]

Description:

Track recoverable errors of the parser at warning level.

Considering that other parts of the code seem to consider this as a warning, and considering that the user has the ability to set whether to consider this type of error as a real error to worry about with the use of the HEDERA_MIRROR_IMPORTER_PARSER_HALTONERROR env parameter, it would be useful to set these messages to warning level so that we can exclude them from the logs, while retaining the ability to track all other parser's errors.

Related issue(s):

N/A

Notes for reviewer:

N/A

Checklist

• Documented (Code comments, README, etc.)
• Tested (unit, integration, etc.)

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.32%. Comparing base (d1b41f1) to head (ac95439).

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #9634      +/-   ##
============================================
+ Coverage     92.22%   92.32%   +0.09%     
+ Complexity     7609     7505     -104     
============================================
  Files           932      914      -18     
  Lines         31940    31684     -256     
  Branches       4029     4007      -22     
============================================
- Hits          29456    29251     -205     
+ Misses         1535     1489      -46     
+ Partials        949      944       -5     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[steven-sheehy]

We currently have alerts that trigger based upon recoverable errors that this would break. For full mirror nodes, these are errors not warnings that have helped us catch serious issues with the record stream. Plus, setting them to warn doesn't help since by default we log warnings.

[Neurone]

I see. I care about setting this as a warning because in my case, for example, I only want to log errors that require human intervention, so I set the logs that way, overriding the default logging levels via application.yml.

My understanding of a recoverable error is that this is something that the mirror node will take care of and I don't actually need to do anything on the DB.

Is this the case, or do I need to do something if I see these errors in the logs?