x/scalpels
Code Issues Proposed changes
Digging and Tracing system for OpenStack
109 Commits 1 Branch 0 Tags 1 MiB
Python 76.2%
Shell 19.6%
Mako 4.2%
d0ed1dbd4f
Go to file
Kun Huang d0ed1dbd4f test multiple result for one tracer 
Change-Id: Ibb619a267b2a3ba28a05b850673fb1b73aac8e4d
2015-11-19 22:41:51 +08:00
devstack test multiple result for one tracer 2015-11-19 22:41:51 +08:00
doc add tracer-worker relationship image 2015-11-15 23:27:36 +08:00
scalpels append results to tracer 2015-11-19 17:03:54 +08:00
scripts mv scripts/agent.py to cmd/ 2015-11-11 13:46:37 +08:00
tests test multiple result for one tracer 2015-11-19 22:41:51 +08:00
tools add is_running status to tracers 2015-11-18 11:43:11 +08:00
.gitignore add git ignore 2015-10-20 17:33:28 +08:00
.gitreview add gitreview 2015-10-23 19:49:46 +08:00
.testr.conf add tox-related stuff 2015-10-23 23:06:30 +08:00
LICENSE Initial commit 2015-09-16 14:20:51 +08:00
README.rst add mission 2015-11-15 23:55:49 +08:00
requirements.txt add oslo.messaging requirements 2015-11-09 01:42:39 +08:00
setup.cfg redesign endpoints 2015-11-11 13:18:43 +08:00
setup.py add simple python framework 2015-10-20 15:39:06 +08:00
test-requirements.txt add tox-related stuff 2015-10-23 23:06:30 +08:00
tox.ini fix unit test job and doc job 2015-10-25 09:23:31 +08:00

README.rst

Scalpels

Scalpels is distributed tracing or digging system for OpenStack.

Background

OpenStack is made of multiple Python-based projects. Each project has similiar but different architecture. Scalpels gathers useful scripts or 3rd tools to help operator find what happen in your cloud.

Contribute

This project is prototype now and under development. If you have interests in this work, please contact @kun_huang, at #openstack-chinese channel.

Mission

Scalpels is kind of "debugfs" for OpenStack. It gathers data from number of tracers like proc or systemtap which could be used to quantize preformance under workloads.

Single Node Architecture

This type of deployment is used as POC in OpenStack community CI.

All-in-One deployment

Multiple Node Architecture

This is under Designing:

Multiple deployment

Agent-Tracer-Worker

The relationship is:

agent-tracer-worker.png

Ideas

Each project will have scripts working:

  • on python calls
  • on sql queries
  • on filesystem I/O
  • on RPC calls if need
  • on necessary system calls
  • on common system statistics