3 ms·
That double-grew is awful, certainly if the log files are large and contain hundreds or even thousands of such calls. Also, I don't see how this would work whe
by Someone 10y ago
That double-grew is awful, certainly if the log files are large and contain hundreds or even thousands of such calls.
Also, I don't see how this would work when using tail ("one of my coworkers brought up the fact that it's quite annoying to be tailing the log and constantly seeing requests for /lbhealth (or similar) 'polluting' the log.")
I would log this to a separate file or add a /lbhealth marker that can be easily and robustly grepped for to the log lines (might be possible by configuration of your logging software)
I would also consider normally only logging errors from these calls normally, hiding the 'normal' lines in debug logging (the web server log has a record of the calls being made)