Health check endpoint ignores fee escalation:

[ximinez]

High Level Overview of Change

Changed the health check endpoint (https://[host]:[peerport]/health) to only consider and consume the portion of the load_factor that is limited to the server. This ignores any changes related to fee escalation. Fee escalation describes how busy the ledger is, not the server.

Also refactored that function to use jss labels instead of strings for JSON fields.

Context of Change

I was looking at the health of my server and was annoyed when it went into warning status solely with a load factor in the 500's because fee escalation kicked in.

It is my opinion that this is not correct or desired behavior. I may be wrong about that, so that discussion would be appropriate here.

Type of Change

• [X ] Bug fix (non-breaking change which fixes an issue)
• [X ] Refactor (non-breaking change that only restructures code)

Before / After

Before: If the server's load_factor goes over 100 because of fee escalation, the /health endpoint will return a 503 result code and include the load_factor in the result JSON.

After: If the server's load_factor goes over 100 because of fee escalation, but the server's internal load is under 100, the /health endpoint will return a 200, and not include load_factor in the JSON. If the internal load goes over 100 because the server itself is busy, it will continue to be flagged by /health as before.

[manojsdoshi]

LGTM