Pre-Con Ed: Triage and Diagnose Apps through Improved Monitoring Coverage

World®
’16
Pre-Con Ed: Triage And Diagnose
Apps through Improved Monitoring
Coverage
Andreas Reiss, Sr. Principal Engineering Services Architect, CA Technologies
Guenter Grossberger, Sr. Engineering Services Architect, CA Technologies
DO4X51E
DEVOPS – AGILE OPERATIONS

2 © 2016 CA. ALL RIGHTS RESERVED.@CAWORLD #CAWORLD
© 2016 CA. All rights reserved. All trademarks referenced herein belong to their respective companies.
The content provided in this CA World 2016 presentation is intended for informational purposes only and does not form any type of
warranty. The information provided by a CA partner and/or CA customer has not been reviewed for accuracy by CA.
For Informational Purposes Only
Terms of this Presentation

Abstract
Over the past year CA Technologies has made numerous improvements that are designed to
dramatically reduce the time required to diagnose, triage and remediate app performance
problems. Join this lab-based session to receive a detailed tour of APM Team Center with deep-
dives into Differential Analysis, Timeline Views and Assisted Triage features that proactively
suggest fixes so you can optimize app performance with minimal effort. We will also
demonstrate how you can use CA Cross-Enterprise Application Performance Management to
monitor business transactions across common middleware and mainframe subsystems with a
single and intuitive interface. Pre-registration strongly encouraged as space is limited; pre-
registered attendees will be admitted first.
Andreas Reiss
CA Technologies,
Sr. Principal
Engineering Services
Architect
Guenter
Grossberger
CA Technologies, Sr.
Engineering Services
Architect

Agenda
WELCOME TO NOWHERE BANK
APPLICATION TRIAGE REDEFINED
STATE OF THE ART TRIAGE
FROM TRANSACTION TO MAP COMPONENTS
NEXT GENERATION TRIAGE
1
2
3
4
5

Welcome to Nowhere Bank

Triage And Diagnose Apps
Introduction to Nowhere Bank
Nowhere Bank Server
Portal
Mediator
Engine
HTTP Communication
Request Queues
Reply QueuesQueue
Server
§ Three tier application
§ HTTP communication between
portal and mediator
§ Mediator and engine
communicate with request / reply
queues

Getting Nowhere Bank Started
§ Locate Banking-Control.cmd at
folder C:NowhereBank and
start it.
§ Start in sequence
– Messaging server
– Engine
– Mediator
– Portal
§ Log into CA APM web view
– http://localhost:8080
1 2 3 4

Getting Nowhere Bank Started
§ Start traffic
§ Validate transaction traces
in investigator
1

Application Triage Redefined

§ Problems in the history of CA APM have changed significantly
§ The time when the slowest component is no longer the root cause
Evolution of Problems
Coding
Exceptions
Threading
and
Concurrency
SOA
Integration
Resource/
Containers
Business
Modelling vs.
Programming
Loose
Coupled
Services
Time
Complexity/ Abstraction

§ CA APM evolution in sync with problem evaluation
§ CA APM designs and develops to face the problems of tomorrow
Evolution of Problems Requires an Evolution of CA APM
Time
Evolution
Alerts
Custom
Dashboards
Custom
Perspectives
Assisted
Triage

State of the Art Triage

Legacy Triage
Alert Flooding Bridge Call
Crisis Meeting Restart PowerPoint

Next Generation Triage
Awareness Understand
IdentifyCommunicate
Isolate
Damage of the
outbreak

From Transaction to Map Components

Vertex attributes:
• Execution context
• System variables
App A
Servlet
WS
Call
App B
SQLWS
Transactions
discovered by
the monitoring
agent
Discovery
Servlet
Vertex
Vertex attributes
Automatic
attributes
Custom
attributes
• Analysis of transactions
• Creation of vertexes for
significant components

Dashboard
Homepage
Servlet Vertex
Map
MyServlet
Perspectives define the columns of the dashboard; one vertex,
dependent on attributes appears on multiple columns
Perspectives define the nesting and grouping of Vertexes.
Use Perspectives to introduce semantic layers of an
environment; a vertex can only appear once
Homepage visualizes most left Vertexes as “Experiences”; an
experience can only appear once

Warm Up
§ Define a perspective
“Nowhere Bank Architecture”

Awareness

Awareness
Alert
Alert
Alert
Alert
Alert Alert
Alert
Alert
Alert
Alert
Alert
Alert
Alert

Awareness – Using our Dashboard (“Status Console”)
Perspectives
define the
columns

Trigger Scenario
§ Trigger our error scenario
by clicking on the ”Start
Scenario” button

Introducing the New Home Screen
All status
information
on
experiences
Overall health
information
across all
experiences

Introducing the New Home Screen
Vertex:
§ Name: Charge
Account
§ Application:
Portal
§ JVM A
Charge Account
Team Center Perspective
Portal Application
Charge Account
JVMA
JVMB
JVMC
Homepage
JVMA JVMB JVMC
Charge
Account
Portal
Application

Experience
xyz
xyz
Experience vs. Problem vs. Anomaly
§ Experiences
– Most left
components
– Transaction
start
components
xyz xyz xyz
xyz
xyz xyz
xyz

Home Screen Set Up
Define a home screen experience “Nowhere Bank Health” with variations of
– Application
– Business transaction
– …

Damage Of
The Outbreak

Damage of the Outbreak

Damage of the Outbreak
Quickly identify
call hierarchies
and
dependencies
Identify
immediately
the size of the
damage

Observe the Awareness Factor
§ Move to the
dashboard view to
get aware of
problem indicators

Observe the Outbreak
§ Move to the map view to observe the
problem

Impact of the Outbreak
Health score /
quality of
transactions
Quickly identify
the health of an
experience
Are problems
developing? How
many problems to
take care of?
Differentiate
between slow
and failing
transactions
What’s the time
distribution among
requests? Faster than
one second? Two
seconds? Or slower?

Assisted Triage Problem vs. Anomaly
§ Many road blocks, traffic jams,
construction sites
§ Will it affect me?
§ How does it affect me on my trip?
§ Does it even make sense to step
into the car?

Next Generation Problem Analysis
We expect
some trouble
on our way.
“affected
transactions on
my route”
We expect 7
minutes delay on
our trip compared
to usual

Impact of the Outbreak
§ The home screen present real time
“traffic information” for every
experience
– Problems (Do I have severe problems
where I maybe should not even step
into the car)
– Anomalies (Maybe some minor
construction sites which are not a
problem, yet)

xyz xyz
§ Two
components
with problems
identified
§ No experience
affected
Experience
xyz
xyz
xyz xyz
xyz
The system detected
one anomaly with two
affected components

§ Once an
experience
shows errors,
its identified
as problem
xyz xyz
Experience
xyz
xyz
xyz xyz
xyz
The system detected
one problem with
three affected
components

Observe the Outbreak
§ Observe how
different transactions
are impacted and
where

Isolate

Isolate ‒ Focus on the Problem Route
§ Isolate problem transaction
§ Focus on problem TRANSACTIONS
§ How does one problem indicating
component affect end-to-end
transactions

Notebook – Isolating Data
§ The notebook to isolate transactions end to end
§ Focus on problem components and resolve related transactions
§ Compare transactions, performance metrics

Notebook – Isolating Data
§ The isolation of problems and transactions is
now only one click away
Link to analyze
the experience in
the notebook
Link to open
a problem in
the
notebook

Notebook – Isolating data
Timeline with
indicators
Transaction
health
information
Isolated
transaction
components
Problems and
anomalies
section

Isolate the Problem Transactions
§ Isolate the timeframe of
the problem in the
homepage

Understand

Understand
§ Assisted Triage is the latest patent-pending innovation to
more efficiently find problems in software
§ Blame the right source of the problem
§ Deal with alert flooding

Understand
1. Analyze the transaction path end-to-end
2. Record types of problem triggers
3. Put problems into categories and sort them
chronological
18:22
18:35
18:42
18:48

Understand
§ The notebook as a
combined solution
to provide all
information related
to a problem
Business transaction
/ experience related
to a problem
Time stamp of first
and last appearance
of evidences
All detailed
evidences related to
a problem

Isolate the Problem Transactions
§ Analyze the evidences
and story of the problem
§ Navigate to the notebook
§ Identify the problem
transaction

Identify

Identify
§ Explain the path of a problem
§ Differentiate root cause and impact of a problem
§ Compare problem states with historical performance data

Assisted Triage
xyz xyz xyz xyz
18:25 – Differential
Analysis trigger
18:22 -
timeouts
18:20 -
stalls
Evidence types Error component Component relationships Problem sequences
Component relationships
Suspect(s)

Identify
§ Validate the proposed root cause
(suspect) of the problem
Suspect

Communicate

Communicate
§ Communicate the problem
§ Answer with the right information
§ Share the information

Communicate ‒ Share
• Sharable link to the isolated
problem description (notebook)
• All details and evidences at one
place

Answer with the Right Information
How are
transactions
affected?
Which
components in
the system are
affected?
Which BTs are
affected
What was the
problem?
When did it
happen?
Give me the
evidences?

Thank you.
Stay connected at communities.ca.com

Pre-Con Ed: Triage and Diagnose Apps through Improved Monitoring Coverage

More Related Content

What's hot (20)

Similar to Pre-Con Ed: Triage and Diagnose Apps through Improved Monitoring Coverage (20)

More from CA Technologies (20)

Recently uploaded (20)

Pre-Con Ed: Triage and Diagnose Apps through Improved Monitoring Coverage