Log-Analysis and Alerting for OPNFV-VSPERF
Internship Projects/Mentors
Title | Log-Analysis and Alerting for OPNFV-VSPERF |
Status | completed |
Difficulty | Low |
Description
OPNFV VSPERF already includes comprehensive Monitoring solution. However, it lacks the Log-Analysis and Alert-Management solution. This project aims to include log and alert management solutions based on opensources - ELK Stack. This will enhance the ability of VSPERF to:
Perform post-test correlation/causation analysis.
Adapt the Test-Traffic patterns based on the Alerts.
The first aim is to create custom kibana dashboard analyzing and synthesizing VSPERF logs/events. For every test-run, VSPERF generates following logs:
vsperf-overall
vswitchd
trafficgen
host-cmds
guest-cmds
These logs have useful information that can help user to understand the test behavior, and analyze the performance results.
VSPERF will have Alerting from the Monitoring component, and lacks alerting from the logs. As the next step, this project aims to use the same information - The logs - and create alerts than will be notified to VSPERF. The Trafficgenerators of the VSPERF will be the main consumers of these alerts.
Additional Information
Some Links:
https://www.elastic.co/guide/en/kibana/current/dashboard-create-new-dashboard.html
https://landing.google.com/sre/sre-book/chapters/practical-alerting/
https://docs.google.com/document/d/199PqyG3UsyXlwieHaqbGiWVa8eMWi8zzAn0YfcApr8Q/preview
Tentative Process:
1. Learn about VSPERF
2. Install and run the OSS tools successfully.
3. Create sample Kibana Dashboard
4. Create sample Alert Rules.
5. Inspect the logs the tests have created
6. Build the Kibana dashboard to VSPERF's requirements & specifications
7. Create VSPERF specific alerts and send notifications.
Learning Objectives
Experience in real-world logging/monitoring/alerting domain.
Experience in creating custom dashboards in Kibana.
Experience in Authoring custom Alert-Rules.
Expected Outcomes
Custom Kibana Dashboards.
Custom Alert Notifications.
Relation to LF Networking
OPNFV-VSPERF.
Education Level
Undergraduate
Skills
1. Python.
2. Yaml
3. Dashboard creation - handling different controls
Future plans
This project is related to larger "Performance Management" and "Analytics in Networking" . Hence, the gained experience will be very useful in the student is planning to pursue a career in these domains.
Preferred Hours and Length of Internship
Full-time (40 hours a week for 12 weeks during the summer)
Mentor(s) Names and Contact Info
Sridhar K. N. Rao sridhar.rao@spirent.com
Mentee(s) Names and Contact Info
Adarsh Yadav, adarsh.yadav@ramanujan.du.ac.in
Aditya Srivastava, adityasrivastava301199@gmail.com