-
Notifications
You must be signed in to change notification settings - Fork 999
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Avoid same error if other line than first change #555
Avoid same error if other line than first change #555
Conversation
Only the first line of backtrace was use to mark error like same. Avoid this behavior because some language ( like python ) change only last line of backtrace
Thanks, I'll run this PR to see the difference in the upcoming days. /cc @jone |
@shingara sorry for the delay, I've just deployed the branch and it does indeed split the errors up where previously we would have gotten a single group. As this change could lead to many more groups I'd like to monitor it for some time running with the patch and see how the grouping turns out. Is that alright for you? I'll provide feedback in the upcoming week. |
Ok, thanks. I waiting your feedback to merge this patch and release the 0.2.1 errbit version. |
@shingara the patch works good and the grouping is much more useful than before. Thank you. |
…cktrace_line Avoid same error if other line than first change
How should I produce custom errors if I want to use this? I think the behaviour of "Fingerprint" needs some documentation |
Same here. The exceptions all seemed to have occured in delayed_job 3.0.5 in combination with delayed-plugins-airbrake-1.0.0 |
Ok, I really need release a errbit 0.2.1 version with this fix so. |
Yes - actually we patched the code locally to restore old behaviour which is of course not ideal :( I suppose a lot of people , as we do, use CustomExceptions to catch all exceptions in a certain piece of code. There should be the possibility to see these exceptions grouped. It would be great if errbit continued supporting this behaviour either via a global config option or a per application behaviour. Typical example :) begin
response = HTTParty.get(self.url)
rescue Exception => e
raise CustomWebRequestSomehowFailedException.new(e)
end would have been grouped nicely before (while still providing the different error classes on the overview as percentage) |
Only the first line of backtrace was use to mark error like same. Avoid
this behavior because some language ( like python ) change only last
line of backtrace
See mailing https://groups.google.com/d/msg/errbit/JuIGnN289ps/FWWF6_6qDR8J