redacted.redislabs.com could not be resolved (2: Server failure)

We recently had an issue with our Redis provider and this error came up which caused our application to throw 500s. I'm trying to understand what could've caused this and why it raised this error. I was unable to recreate this specific error but causing similiar errors by intentionally breaking the hostname are handled fine.

2021/08/18 10:20:33 [error] 28600#28600: *11030103 lua entry thread aborted: runtime error: /usr/local/share/lua/5.1/resty/http_connect.lua:208: bad request
stack traceback:
coroutine 0:
        [C]: in function 'connect'
        /usr/local/share/lua/5.1/resty/http_connect.lua:208: in function 'connect'
        /usr/local/share/lua/5.1/resty/http.lua:913: in function 'request_uri'
        /etc/nginx/lua/bugsnag.lua:331: in function 'send_report'
        /etc/nginx/lua/bugsnag.lua:405: in function 'call'
        /etc/nginx/lua/redacted.lua:233: in main chunk, client: redacted, server: _, request: "GET[redacted] HTTP/1.1", host: "redacted"```

Start at /usr/local/share/lua/5.1/resty/http_connect.lua:208 and trace the request through resty/http.lua:913, comparing the reported DNS failure with the intentionally broken-hostname case. Review the referenced connection and error paths to determine why this failure becomes a 500; done means a reproducible explanation and a clearly scoped handling change.