You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
The only way to get diagnostics information relevant to issues is by inspecting a long string created by the ToString() method of the CosmosDiagnostics class. Having to log this string can be problematic to as we have come across issues whereby log buffers can overflow resulting in broken/incomplete logs.
We are therefore forced to log the only publicly exposed properties e.g, (query metrics, client elapsed time etc.), which have proven to be insufficient for debugging when some issues arise.
Describe the solution you'd like
Expose relevant information for debugging as public properties that can be included in structured logging and supplied as and when they are needed. For instance, information on the exception thrown, replica endpoints that are hit, command type, etc. can be exposed as properties that can be logged in a structured manner.
Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.
Logging exposed properties on CosmosDiagnostics class. Eg., Index hit ratio and various latency values. These have proven to be insufficient in the event where certain potential service issues are encountered. Recently reported issues have required us to present full diagnostics strings to effectuate investigations, and logging these have been problematic in the past..
Truncating diagnostics strings to avoid logging diagnostics data that could be potentially unbounded. However, some data could be lost
Additional context
Add any other context or screenshots about the feature request here.
The text was updated successfully, but these errors were encountered:
Is your feature request related to a problem? Please describe.
The only way to get diagnostics information relevant to issues is by inspecting a long string created by the
ToString()
method of theCosmosDiagnostics
class. Having to log this string can be problematic to as we have come across issues whereby log buffers can overflow resulting in broken/incomplete logs.We are therefore forced to log the only publicly exposed properties e.g, (query metrics, client elapsed time etc.), which have proven to be insufficient for debugging when some issues arise.
Describe the solution you'd like
Expose relevant information for debugging as public properties that can be included in structured logging and supplied as and when they are needed. For instance, information on the exception thrown, replica endpoints that are hit, command type, etc. can be exposed as properties that can be logged in a structured manner.
Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.
Additional context
Add any other context or screenshots about the feature request here.
The text was updated successfully, but these errors were encountered: