TO: cdns/capacity incorrect format on log messages when TM errors

[shamrickus]

I'm submitting a ...

• bug report

Traffic Control components affected ...

• Traffic Ops

Current behavior:

Currently, TO will query {tmHost}/publish/CacheStatsNew (added in 5.0) and then {tmHost}/publish/CacheStats if that fails.

Both of these log errors that are formatted bad:
getCapacity failed to get CacheStatsNew from cdn '{cdn}' monitor '{monitor}', trying CacheStats{err}
Where err is getting cache stats for CDN '{cdn}' monitor '{monitor}': {tmError}
resulting in:
getCapacity failed to get CacheStatsNew from cdn '{cdn}' monitor '{monitor}', trying CacheStatsgetting cache stats for CDN '{cdn}' monitor '{monitor}': {tmError}

CacheStats has roughly the same issues with some wording changed.

Expected behavior:

Actually output something readable

Minimal reproduction of the problem with instructions:

Run TO 5.0+ with a TM <=4.1 and call {to}/api/v3/cdns/capacity
Or, make TM unreachable from TO right before calling {to}/api/v3/cdns/capacity

[rawlinp]

Additionally, it would be helpful for TO to also include the HttpProxy value in the error message (if one is being used) and the full request URL that failed.