Minimalist Error collection Service compatible with Rollbar clients. Sentry or Rollbar alternative.

Last update: Nov 11, 2022

Overview

Minimalist Error collection Service

Features

Compatible with any Rollbar client(see https://docs.rollbar.com/docs). Just change the endpoint URL to your errorpush URL.
Inserts all error logs into a single PostgreSQL table.

Why use this over Sentry/Rollbar?

Instant setup
Free
No rate limiting
When you don't need all the bells and whistles: just want to log all errors into a database.
Flexibility - use whatever you want to query the PostgresQL table for errors

Running

errorpush requires docker

docker run -p 5000:5000 -e ACCESS_TOKEN=<your_access_token_of_choice> -e POSTGRES_URI=postgres://username:[email protected]/yourdb hauxir/errorpush:latest

That's it, just set up a reverse proxy and point your rollbar client to your server.

Example Query

SELECT error_id, Max(( ( BODY ->> 'trace' ) :: jsonb ->> 'exception' ) :: jsonb ->> 'class') AS EXCEPTION, Max(( ( BODY ->> 'message' ) :: jsonb ->> 'body' )) AS message, Count(*), Max(timestamp) AS last_seen FROM   errors GROUP  BY error_id ORDER  BY last_seen DESC;

             error_id             | exception |   message    | count |         last_seen          
----------------------------------+-----------+--------------+-------+----------------------------
 8cca0a18f56269b5a5243f7cc2906f79 | NameError |              |     4 | 2021-09-08 18:34:05.751548
 b6012c1be2bef37f570077f2ce2e908b |           |              |     2 | 2021-09-08 18:15:09.944348
 5acf76ad5f327d811ca9282b5d5a933a |           | Hello world! |     3 | 2021-09-08 18:15:09.944308
 794ef3b916db810d9162ad54aff32a14 |           | HEY          |     1 | 2021-09-08 18:12:19.705926
(4 rows)

Metabase

You can use metabase to visualize the data.

PostgreSQL view for the above image:

create view error_report as 
select 
  error_id, 
  max(
    concat(
      coalesce(
        (
          (body ->> 'trace'):: jsonb ->> 'exception'
        ):: jsonb ->> 'class', 
        'Error'
      ), 
      ': ', 
      (
        (body ->> 'trace'):: jsonb ->> 'exception'
      ):: jsonb ->> 'message', 
      (body ->> 'message'):: jsonb ->> 'body'
    )
  ) as error, 
  count(*) as occurences, 
  max(timestamp) as last_seen, 
  max(
    array_to_string(
      array(
        select 
          concat(
            el ->> 'filename', ':', el ->> 'lineno'
          ) 
        from 
          jsonb_array_elements(
            (
              (body ->> 'trace'):: jsonb ->> 'frames'
            ):: jsonb
          ) as el 
        limit 
          4
      ), 
      ', '
    )
  ) as trace, 
  max(environment) as environment, 
  max(custom ->> 'revision') as revision 
from 
  errors 
group by 
  error_id 
order by 
  last_seen desc;

Minimalist Error collection Service compatible with Rollbar clients. Sentry or Rollbar alternative.

Related tags

Overview

Features

Why use this over Sentry/Rollbar?

Running

Example Query

Metabase

Owner

Haukur Rósinkranz

AlgoVision - A Framework for Differentiable Algorithms and Algorithmic Supervision

The codebase for Data-driven general-purpose voice activity detection.

Neuralnetwork - Basic Multilayer Perceptron Neural Network for deep learning

Clustering with variational Bayes and population Monte Carlo

Employs neural networks to classify images into four categories: ship, automobile, dog or frog

A package for music online and offline rhythmic information analysis including music Beat, downbeat, tempo and meter tracking.

Intelligent Video Analytics toolkit based on different inference backends.

Exploration-Exploitation Dilemma Solving Methods

EASY - Ensemble Augmented-Shot Y-shaped Learning: State-Of-The-Art Few-Shot Classification with Simple Ingredients.

PyTorch implementation of the Flow Gaussian Mixture Model (FlowGMM) model from our paper

This is an official implementation of the CVPR2022 paper "Blind2Unblind: Self-Supervised Image Denoising with Visible Blind Spots".

Image Super-Resolution by Neural Texture Transfer

Cluttered MNIST Dataset

Picasso: a methods for embedding points in 2D in a way that respects distances while fitting a user-specified shape.

Compares various time-series feature sets on computational performance, within-set structure, and between-set relationships.

TLDR: Twin Learning for Dimensionality Reduction

A Pytorch implementation of "Splitter: Learning Node Representations that Capture Multiple Social Contexts" (WWW 2019).

GazeScroller - Using Facial Movements to perform Hands-free Gesture on the system

HAT: Hierarchical Aggregation Transformers for Person Re-identification

Balancing Principle for Unsupervised Domain Adaptation