undev / errbit Goto Github PK

If I mark an error resolved and then another exception is submitted related to that error, the "Errors" count in application list is not updated. If I open the specific app details in errbit, I see that new error is submitted

To reproduce:

1. create an app in errbit
2. submit an exception to errbit with rake airbrake:test
3. open "Apps" tab in errbit. In the list of all the applications there will be 1 error visible for the app.
4. mark the error as resolved
5. submit another exception with rake airbrake:test
6. open "Apps" tab. There are no errors visible for the app in applications list
7. open the app details in errbit and there will be an unresolved error with count 1

Expected behavior: Application list should also show 1 unresolved error for that app.

The last commit with correct behavior is: c139988

As per this comment where the owner of Errbit requested that this project be renamed:

errbit#623 (comment)

It would be helpful if this project was called something like errbit-postgres to distinguish it.

I'm trying to setup errbit on PG. Nad run into this error

Processing by AppsController#index as HTML
Completed 401 Unauthorized in 18ms
Started GET "/users/sign_in" for 127.0.0.1 at 2013-12-10 12:33:09 -0800
Processing by Devise::SessionsController#new as HTML
  Rendered devise/sessions/new.html.haml within layouts/application (8.3ms)
Completed 500 Internal Server Error in 61ms

NoMethodError (undefined method `find_or_initialize_by' for #<Class:0x9639558>):
  lib/overrides/hoptoad_notifier/hoptoad_notifier.rb:11:in `send_notice'

Ruby version: 2.0.0-p353

What is the policy for bugfixes that are submitted to the original repo https://github.com/errbit/errbit?
Wouldn't it make sense to cherry pick / merge them all into this repo as well (as long as they are not Mongo specific)?

Thank you for your effords.
Is there any chance that you have a yet unpushed branch that contains newer patches from the official repo?

Have a wonderful day and thank you in advance for the response

I'm trying to migrate existing MongoDB data into Postgres using rake errbit:db:migrate_from_mongo.

However I cannot finish it, because the copying of backtraces leaks memory like a sieve. Memory usage of ruby 2.1.0 increases by about 10 MB per imported backtrace. So I would need ~4GB of free RAM to import the ~400 backtraces.

Please have a look at what's causing the leak. The code create so many dynamic objects and methods that it's hard to understand where the problem lies.