Skip to content

Respect existing GC::Profiler.enabled? state#4

Closed
jasonrclark wants to merge 1 commit intopeek:masterfrom
jasonrclark:kinder_enabling
Closed

Respect existing GC::Profiler.enabled? state#4
jasonrclark wants to merge 1 commit intopeek:masterfrom
jasonrclark:kinder_enabling

Conversation

@jasonrclark
Copy link

Currently peek-gc will enable and disable GC::Profiler before every request.
This has unfortunate side-effects if anyone else has set
GC::Profiler.enable and assumes the setting won't be changed under them.

newrelic_rpm was making this assumption (which we're fixing in the upcoming
release), but I thought others might benefit from being a little more cautious
here.

Note that there's still an issue here related to different conflicting code
calling GC::Profiler.clear, but I don't have a good solution to that. At the
very least preserving the existing setting on this seemed less likely to cause
suprises.

Currently peek-gc will enable and disable GC::Profiler before every
request. This has unfortunate side-effects if anyone else has set
`GC::Profiler.enable` and assumes the setting won't be changed under
them.

newrelic_rpm was making this assumption (which we're fixing
in the upcoming release), but I thought others might benefit from being a
little more cautious here.

Note that there's still an issue here related to different conflicting
code calling `GC::Profiler.clear`, but I don't have a good solution to
that. At the very least preserving the existing setting on this seemed
less likely to cause suprises.
@jasonrclark
Copy link
Author

D'oh, should have looked at the repo. Looks like this dups #3.

@jasonrclark jasonrclark deleted the kinder_enabling branch June 17, 2014 17:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant