SLING-11748 - Improve logging output of HTTP retries in testing clients

[marcpf]

https://issues.apache.org/jira/browse/SLING-11748

Proposal for improving the HTTP retry log output in the testing clients.

Current Output (from unit test):

Request retry needed due to service unavailable response
Response headers contained:
Header Date:Wed, 11 Jan 2023 08:36:43 GMT
Header Server:TEST/1.1
Header Content-Length:8
Header Content-Type:text/plain; charset=ISO-8859-1
Header Connection:Keep-Alive
Response content: TEST_NOK

Proposed output with this PR (from unit test):

Request retry condition met: [count=1/4], [expected-codes=[200]], [retry-codes=[500, 503]]
Request: GET /test/internalerror/resource HTTP/1.1 [Host: 127.0.0.1:32953, Connection: Keep-Alive, User-Agent: Java, Accept-Encoding: gzip,deflate, Authorization: Basic dXNlcjpwYXNz]
Response: HTTP/1.1 500 Internal Server Error [Date: Wed, 11 Jan 2023 08:39:59 GMT, Server: TEST/1.1, Content-Length: 8, Content-Type: text/plain; charset=ISO-8859-1, Connection: Keep-Alive, ]
Response Body: TEST_NOK

[joerghoh]

Why do write out this information in 3 distinct log messages? I would prefer a single log message, which is then much easier to capture than 3 subsequent messages.

[marcpf]

Thanks @joerghoh. I think 3 log lines is a good middle way. We had a log line per request/response header before. I as well think its easier to deal with in Splunk queries.

[sonarcloud]

Kudos, SonarCloud Quality Gate passed!   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
1 Code Smell

91.7% Coverage
0.0% Duplication

[marcpf]

@dulvac, @joerghoh, quickly wanted to check with you what the status is with my PR. Do you consider it to merge?

[dulvac]

I assume silent consensus so merging :)