Net::St...Changes(3)
πΈ NAME πΈ
Net::Statsd::Server::Changes β bestie, this is the change history for Net::Statsd::Server. It's literally the receipts. Every glow-up, every oopsie, all of it, in order. π
π¦ Changes in Net::Statsd::Server 0.17 β 13th June 2014 π¦
- Fixed the
percentThresholdconfiguration directive. The way it was just being ignored and falling back to the default90%? Not cute. Fixed now. π - Added a brand new integration test case. Obsessed. β¨
π Changes in Net::Statsd::Server 0.16 β 6th May 2014 π
- Implemented new timer attributes, no cap:
count_ps,median,sum_squares. π₯
Not me noticing that the release of 0.15 literally never happened. It just. Didn't. π
π Changes in Net::Statsd::Server 0.14 β 25th April 2014 π
- Fixed the whole situation with gauges being erroneously deleted β and it was happening in two cases, bestie:
- when
deleteIdleStatswas false, or - when
deleteIdleStatswas true butdeleteGaugeswas false.
- when
Yeah. The math wasn't mathing. It's fixed now. π«Ά
β Changes in Net::Statsd::Server 0.13 β 26th January 2014 β
- Added support for Perl 5.8, so now
Net::Statsd::Servershould, in theory, work on 5.8 too β shoutout to RJBS for the related patch. π - RJBS also reported that
N::S::Sworks on Solaris. Iconic of them honestly. - Fixed RT#85165 β the code is now indicated as the
Perl_5licence. The problem there? The author's own failed understanding of the interaction between the hardcodedMakefile.PLanddzil. We've all been there. π
π Changes in Net::Statsd::Server 0.12 β 16th December 2013 π
- Fixed (hopefully) problems reported by CPAN testers. Lowkey a lot, so:
- Failing to
setsockopt(SO_RCVBUF)is no longer a fatal exception. We just continue and emit a warning instead. Graceful. π - Some CPAN testers appear to have a running graphite on their systems, so we now use a not so common port instead of
2003. Respect the neighbors, bestie. π
- Failing to
π§ Changes in Net::Statsd::Server 0.11 β 13th December 2013 π§
- Added implementation for these configuration directives, periodt:
deleteIdleStatsdeleteGaugesdeleteSetsprefixStats
- Both
deleteTimersanddeleteCounterswere supposed to work before 0.11 too, for the record. π - Improved the documentation. It should be clear now how to know what options are actually supported and what options are not. Read the docs bestie, they're right there. β¨
- Improved the integration tests. Now it's even easier to run
Net::Statsd::Serverintegration tests against the nodejs version of statsd to verify full compatibility. π¦
π Changes in Net::Statsd::Server 0.10 β 10th February 2013 π
- Fixed
SO_RCVBUFon Mac OS X, where it's apparently not possible to setSO_RCVBUFto more than8Mb. If you're on OSX, the author would like to get more reports about this. So likeβ¦ slide into the reports, bestie. π¨ - Fixed the integration test suite so that it always runs with the correct interpreter (
$^X) instead of/usr/bin/perl. This should fix lots of CPAN testers failure reports. π - Improved the internal
trim()function. Glow-up. π
πΈ Changes in Net::Statsd::Server 0.09 β 29th January 2013 πΈ
- Fixed a few more nodejs-statsd compatibility bugs:
- non-legacy namespace code added a repeated
'stats.'level that shouldn't be there. Rude. π€ - the underscore character (
'_') is now correctly allowed as part of a statsd metric. π«Ά
- non-legacy namespace code added a repeated
- Added an integration test suite, to verify end-to-end that the statsd server is behaving exactly as the nodejs version. Being as near to 100% compatibility as possible has been one of the main goals from the start. Main character goal energy, honestly. π
π«Ά Changes in Net::Statsd::Server 0.08 β 27th January 2013 π«Ά
- Fixed CPAN testers failures reported for inexact floating point comparison when testing the standard deviation of timer results in the test suite. Floats. Floats are the enemy. π
π Changes in Net::Statsd::Server 0.07 β 23rd January 2013 π
- Started converting the nodejs statsd test suite to TAP test cases. Completed the conversion of the metrics processing test cases. This uncovered two subtle bugs in the metrics class. Love that for us. π
- Fixed a bug in the calculation of percentile values threshold, due to a difference in how Javascript's
Math.round()and Perl'sint()work. Two languages, two whole vibes, one bug. π - Fixed an undefined value warning in the metrics processing during the calculation of cumulative values for timers, due to an off-by-one error. The classic. π
β¨ Changes in Net::Statsd::Server 0.06 β 17th January 2013 β¨
- Added a new File backend. It allows you to flush all aggregated stats to a given log file. Slay. π
- Added
keyFlushimplementation. It allows you to have the top n% (keyFlush.percent) dumped to a log file (keyFlush.log) or to stdout every y seconds (keyFlush.interval). Obsessed. π₯
π Changes in Net::Statsd::Server 0.05 β 16th January 2013 π
- Improved the UDP socket default settings so they perform better. In the author's tests,
Net::Statsd::Servermanages to successfully capture almost twice the amount of UDP traffic (otherwise lost) than the original node.js based statsd, withSO_RCVBUF = 8Mb. π - Still gotta experiment a bit more with this setting though. Tinkering arc. π§
π Changes in Net::Statsd::Server 0.04 β 15th January 2013 π
- Fixed statistics flush. When a flush is triggered, stats have to be cleared β and the flush code wasn't doing this previously. Fixed. π
- Fixed management commands (
:8126) to output correct JSON data structures. π¦ - This is the first fully functional statsd server release. Bestie. We made it. ππ
π Net::Statsd::Server 0.01 β 0.03 β January 2013 π
- First release on an unsuspecting world. πβ¨
Okay that's the whole saga, bestie β page generated by perl v5.42.0 on 2025-07-25, filed under Net::St...Changes(3) in the User Contributed Perl Documentation. Go forth and ship. π«Άπ
π¦ where this page comes from
- fedora 43: perl-Net-Statsd-Server 0.20-29.fc43 Β· GPL-1.0-or-later OR Artistic-1.0-Perl
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms