logreduce - extract anomaly from log files

Based on success logs, logreduce highlights useful text in failed logs. The goal is to save time in finding a failure’s root cause.

On average, learning run at 2000 lines per second, and testing run at 1300 lines per seconds.

How it works

logreduce uses a model to learn successful logs and detect novelties in failed logs:

Random words are manually removed using regular expression
Then lines are converted to a matrix of token occurrences (using HashingVectorizer),
An unsupervised learner implements neighbor searches (using NearestNeighbors).

Caveats

This method doesn’t work when debug content is only included in failed logs. To successfully detect anomalies, failed and success logs needs to be similar, otherwise the extra informations in failed logs will be considered anomalous.

For example this happens with testr where success logs only contains ‘SUCCESS’.

Install

Fedora:

sudo dnf install -y python3-scikit-learn
git clone https://softwarefactory-project.io/r/logreduce
pushd logreduce
python3 setup.py develop --user
popd

openSUSE:

sudo zypper install python3-scikit-learn
git clone https://softwarefactory-project.io/r/logreduce
pushd logreduce
python3 setup.py develop --user
popd

Pip:

pip install --user logreduce

Command Line Interface Usage

Logreduce needs a baseline for success log training, and a target for the log to reduce.

Logreduce prints anomalies on the console, the log files are not modified:

"%(distance)f | %(log_path)s:%(line_number)d: %(log_line)s"

Local file usage

Compare two files or directories without building a model:

$ logreduce diff testr-nodepool-01/output.good testr-nodepool-01/output.fail
0.232 | testr-nodepool-01/output.fail:0677:  File "voluptuous/schema_builder.py", line 370, in validate_mapping
0.462 | testr-nodepool-01/output.fail:0678:    raise er.MultipleInvalid(errors)
0.650 | testr-nodepool-01/output.fail:0679:  voluptuous.error.MultipleInvalid: required key not provided @ data['providers'][2]['cloud']

Compare two files or directories:

$ logreduce dir preprod-logs/ /var/log/

Or build a model first and run it separately:

$ logreduce dir-train sosreport.clf old-sosreport/ good-sosreport/
$ logreduce dir-run sosreport.clf new-sosreport/

Zuul job usage

Logreduce can query Zuul build database to train a model.

Extract novelty from a job logs:

$ logreduce job http://logs.openstack.org/...

# Reduce comparaison to a single project (e.g. for tox jobs)
$ logreduce job --project openstack/nova http://logs.openstack.org/...

# Compare using many baselines
$ logreduce job --count 10 http://logs.openstack.org/...

# Include job artifacts
$ logreduce job --include-path logs/ http:/logs.openstack.org/...

Or build a model first and run it separately:

$ logreduce job-train --job job_name job_name.clf
$ logreduce job-run job_name.clf http://logs.openstack.org/.../

Journald usage

Logreduce can look for anomaly in journald, comparing the last day/week/month to the previous one:

Extract novelty from last day journal:

$ logreduce journal --range day

Build a model using journal of last month and look for novelty in last week:

$ logreduce journal-train --range month good-journal.clf
$ logreduce journal-run --range week good-journal.clf

Filters configuration

Some content yields false positives that can be ignored through filters. Using the –config command line attribute, filters can be set for exclude_files, exclude_paths and exclude_lines. Here is an example filters configuration file:

filters:
  exclude_files:
    - "deployment-hieradata.j2.yaml"
    - "tempest.html"
  exclude_paths:
    - "group_vars/Compute"
    - "group_vars/Controller"
    - "group_vars/Undercloud"
  exclude_lines:
    # neutron dhcp interface
    - "^tap[^ ]*$"
    # IPA cookies
    - "^.*[Cc]ookie.*ipa_session="

Python Module API

Logreduce can be used as a python module for custom use-case.

First you need to create a classifier object:

from logreduce import Classifier, Tokenizer, render_html

clf = Classifier(
    # A function to normalize filename, for example to remove dates or id
    filename_to_modelname=lambda fn: fn,
    # A function to ignore some file, for example configuration file
    keep_file=lambda _: True,
    # A function to process line
    process_line=Tokenizer.process
)

Then you train the object on baseline:

clf.train(["./success-logs/"])

And you test target and create a report:

result = clf.process(["./failed-logs/"])
with open("report.html", "w") as of:
    of.write(render_html(result))

logreduce-tests

This package contains tests data for different type of log such as testr or syslog. Each tests includes a pre-computed list of the anomalies in log failures.

This package also includes a command line utility to run logreduce against all tests data and print a summary of its performance.

Test format

Each tests case is composed of:

A .good file (or directory) that holds the baseline
A .fail file (or directory)
A info.yaml file that describe expected output:

threshold: float # set the distance threshold for the test
anomalies:
  - optional: bool  # to define minor anomalies not considered false positive
    lines: |        # the expected lines to be highlighted
      Traceback...
      RuntimeError...

Evaluate

To run the evaluation, first install logreduce-tests:

git clone https://softwarefactory-project.io/r/logreduce-tests
pushd logreduce-tests
python3 setup.py develop --user

logreduce-tests expect tests directories as argument:

$ logreduce-tests tests/testr-zuul-[0-9]*
[testr-zuul-01]: 100.00% accuracy,  5.00% false-positive
[testr-zuul-02]:  80.00% accuracy,  0.00% false-positive
...
Summary:  90.00% accuracy,  2.50% false-positive

Add –debug to display false positive and missing chunks.

TODOs

Add terminal colors output
Add progress bar
Better differentiate training debug from testing debug
Add a starting log line and report written
Add tarball traversal in utils.files_iterator
Add logstash filter module
Improve tokenization tests

Roadmap

Discard files that are 100% anomalous
Report mean diviation instead of absolute distances
Investigate second stage model

Contribute

Contribution are most welcome, use git-review to propose a change. Setup your ssh keys after sign in https://softwarefactory-project.io/auth/login

Code style is managed with black, run black logreduce before commit to format the source file.

Release files for logreduce 0.6.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for logreduce 0.6.1
File	Size	Uploaded
logreduce-0.6.1.tar.gz	56.5 kB	Jan 19, 2021	Details

Built distribution (wheel)

Table of built distributions (wheels) for logreduce 0.6.1
File	Interpreter	ABI	Platform	Reset
logreduce-0.6.1-py2.py3-none-any.whl 49.8 kB Jan 19, 2021	Python 3, Python 2	none	any	Details

Total release size: 106.3 kB

Release files / logreduce-0.6.1.tar.gz

Download URL	logreduce-0.6.1.tar.gz
Size	56.5 kB
Tags	Source
SHA-256 checksum How to use checksums	`599464cfb79db2c6070601f04606f16ddab294e576479d3feef10c53fbb6d2b3`
BLAKE2b-256 checksum How to use checksums	`912806d9fdb1a1763bfa7a88e09645315ac0d599af14ac5e7b079136440d560d`
Upload date	Jan 19, 2021
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/2.0.0 pkginfo/1.5.0.1 requests/2.22.0 setuptools/41.6.0 requests-toolbelt/0.9.1 tqdm/4.47.0 CPython/3.8.5`

Release files / logreduce-0.6.1-py2.py3-none-any.whl

Download URL	logreduce-0.6.1-py2.py3-none-any.whl
Size	49.8 kB
Tags	Python 2 Python 3
SHA-256 checksum How to use checksums	`a7f9aec81a46fe16cd53d0c52b0bbde990f589575f81227d7e56ab18762a0847`
BLAKE2b-256 checksum How to use checksums	`b4213e277447be054c8a9a11aa469acbf1c7a10020d113643392cb6f2febb772`
Upload date	Jan 19, 2021
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/2.0.0 pkginfo/1.5.0.1 requests/2.22.0 setuptools/41.6.0 requests-toolbelt/0.9.1 tqdm/4.47.0 CPython/3.8.5`