ruby2.4 runs in 25

Gheorghe Popescu (Jira) Wed, 18 Aug 2021 06:11:06 -0700

Title: Message Title

So I did a profiling on calling `Facter.value(:kernel)` on Facter 4 and Facter 2 and Facter 4 is noticeably slower(0.13s vs 0.01s). From the call tree seems that the execution API is slower in Facter 4.
Attached are the traces for each call:

facter2.html - trace for calling `Facter.value(:kernel) with facter 2
Thread: 1580, Fiber: 1560 (100.00% ~ 0.010857273999135941)

facter4.html - trace for calling `Facter.value(:kernel) with facter 4
Thread: 2160, Fiber: 2140 (37.49% ~ 0.13846525101689622)
Thread: 2200, Fiber: 2180 (20.90% ~ 0.13846525101689622)
Thread: 2240, Fiber: 2220 (20.83% ~ 0.13846525101689622)
Thread: 2280, Fiber: 2260 (20.78% ~ 0.13846525101689622)

facter4_patched.html - trace for calling `Facter.value(:kernel) with facter 4 without hitting the resolver. Monkey patching the BaseResolver class
Thread: 2160, Fiber: 2140 (100.00% ~ 0.01828752097208053)

if defined? Facter::Resolvers::BaseResolver

  Facter::Resolvers::BaseResolver.class_eval do

    def self.resolve(_fact_name, _options = {}); end

end

end

Having rspec-puppet to monkey patch Facter 4(https://github.com/gimmyxd/rspec-puppet/commit/49a0c2128ff079a97d2322446098f09cdf6185b1), does not seem to improve the full test suite speed when running on gh actions. Locally I managed to get an improvement of around 20% when running a single test file from puppet-nginx module, for a single OS

Add Comment

This message was sent by Atlassian Jira

--
You received this message because you are subscribed to the Google Groups "Puppet Bugs" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
To view this discussion on the web visit https://groups.google.com/d/msgid/puppet-bugs/JIRA.380977.1608048829000.110767.1629292260057%40Atlassian.JIRA.

Jira (FACT-3039) Puppet6/ruby2.5 runs in 42 minutes, where Puppet5/ruby2.4 runs in 25

Reply via email to