Describe the bug
/find appears to return HTTP 400 Bad Request when a metric name starts with %.
The metrics /render okay in Grafana after the HTTP 400, but I'm hesitant to roll this out to end-users since it shows a red error pop-up "Bad Request".
Example queries:
E4_Environment.*.Processor._Total.%_User_Time.wsp
Environment.SERVER123.Processor._Total.%_User_Time.wsp
/find will succeed for each part of the metric until %_User_Time.wsp
Describe the bug /find appears to return HTTP 400 Bad Request when a metric name starts with %.
The metrics /render okay in Grafana after the HTTP 400, but I'm hesitant to roll this out to end-users since it shows a red error pop-up "Bad Request".
Example queries: E4_Environment.*.Processor._Total.%_User_Time.wsp Environment.SERVER123.Processor._Total.%_User_Time.wsp
/find will succeed for each part of the metric until
%_User_Time.wsp
Could not reproduce issue after copying metric to the same path but without the %_ i.e.
User_Time
CarbonAPI Version Tried
v0.16.1-1-g3eb7
andv0.16.1
CarbonAPI Configuration:
Error occurred with and without:
Backend response (if possible) Here is the logs from both carbonapi + the go-carbon 0.17.1 backend:
Please let me know if I can provide any additional information. Thank you!