You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@geode.apache.org by "Blake Bender (Jira)" <ji...@apache.org> on 2019/10/18 22:20:00 UTC

[jira] [Created] (GEODE-7324) FIx invalid encoding in log files

Blake Bender created GEODE-7324:
-----------------------------------

             Summary: FIx invalid encoding in log files
                 Key: GEODE-7324
                 URL: https://issues.apache.org/jira/browse/GEODE-7324
             Project: Geode
          Issue Type: Bug
          Components: native client
            Reporter: Blake Bender


As a developer, I wish to be able to write log parsing utilities for the native client if need be.  To do this, I need to be able to read the text in the log files via standard languages like Python, C#, etc.  Unfortunately, recent changes to some log statements in the native client code will cause invalid utf-8 bytes to be written to the log in some circumstances, making reading the log very difficult.

 

repro steps:

i. Enable debug-level logging in NC integration tests, and set log-file to a known filename

ii. Run a test case, to generate the log file

iii.  Attempt to parse the file as utf-8 in Python

 

Expected result:
 * File parses correctly

Actual result:
 * Python throws an exception, saying it has encountered an invalid start byte

 

This is known to happen when using the '%zu' format specifier to log a value of type std::chrono::Rep on MacOS.  Other compilers/OSes/types may or may not display this behavior.

 



--
This message was sent by Atlassian Jira
(v8.3.4#803005)