Tweet
Tweet

Slide 1

Slide 1 text

@sebamontini MEDALLIA Monitoring is Dead And why you’re (probably) doing it wrong

Slide 2

Slide 2 text

@sebamontini MEDALLIA

Slide 3

Slide 3 text

@sebamontini MEDALLIA Why?

Slide 4

Slide 4 text

@sebamontini MEDALLIA The big 5

Slide 5

Slide 5 text

@sebamontini MEDALLIA The big 5 ✓ CPU → uptime | mailx -s “cpu” root ✓ MEM → free | mailx -s “mem” root ✓ DISK → (df -h; du -sh /home/*) | mailx -s “disk” root ✓ PROC → (ps -ef | grep important) | mailx -s root ✓ SYS → ping -c 4 google.com | mailx -s root

Slide 6

Slide 6 text

@sebamontini MEDALLIA OK: x < something

Slide 7

Slide 7 text

@sebamontini MEDALLIA WARN: something < x < something

Slide 8

Slide 8 text

@sebamontini MEDALLIA CRITICAL: x > something

Slide 9

Slide 9 text

@sebamontini MEDALLIA

Slide 10

Slide 10 text

@sebamontini MEDALLIA Observability

Slide 11

Slide 11 text

@sebamontini MEDALLIA A system is observable if you can determine the behavior of the system based on it’s outputs.

Slide 12

Slide 12 text

@sebamontini MEDALLIA A system is observable if you can determine the behavior of the system based on it’s outputs.

Slide 13

Slide 13 text

@sebamontini MEDALLIA A system is a set of connected components.

Slide 14

Slide 14 text

@sebamontini MEDALLIA A system is observable if you can determine the behavior of the system based on it’s outputs.

Slide 15

Slide 15 text

@sebamontini MEDALLIA The manner in which a system acts is it’s behavior.

Slide 16

Slide 16 text

@sebamontini MEDALLIA A system is observable if you can determine the behavior of the system based on it’s outputs.

Slide 17

Slide 17 text

@sebamontini MEDALLIA The outputs of a system are the concrete results of it’s behaviors.

Slide 18

Slide 18 text

@sebamontini MEDALLIA Monitoring is the action of observing and checking the behavior and outputs of a system and it’s components over time.

Slide 19

Slide 19 text

@sebamontini MEDALLIA The (real) big 5

Slide 20

Slide 20 text

@sebamontini MEDALLIA Instrumentation Collection Storage Alerting Visualization

Slide 21

Slide 21 text

@sebamontini MEDALLIA Instrumentation

Slide 22

Slide 22 text

@sebamontini MEDALLIA Gauges Counters Histogram Timers

Slide 23

Slide 23 text

@sebamontini MEDALLIA Gauges A gauge is an instantaneous measurement of a value. For example, we may want to measure the number of pending jobs in a queue

Slide 24

Slide 24 text

@sebamontini MEDALLIA Counters A counter is just a gauge that you can increment or decrement its value. For example, we may want a more efficient way of measuring the pending job in a queue

Slide 25

Slide 25 text

@sebamontini MEDALLIA Histogram A histogram measures the statistical distribution of values in a stream of data like median or percentiles

Slide 26

Slide 26 text

@sebamontini MEDALLIA Timers A timer measures both the rate that a particular piece of code is called and the distribution of its duration.

Slide 27

Slide 27 text

@sebamontini MEDALLIA Collection

Slide 28

Slide 28 text

@sebamontini MEDALLIA

Slide 29

Slide 29 text

@sebamontini MEDALLIA Storage

Slide 30

Slide 30 text

@sebamontini MEDALLIA Storage

Slide 31

Slide 31 text

@sebamontini MEDALLIA Alerting

Slide 32

Slide 32 text

@sebamontini MEDALLIA Thresholds Dead man Delta Anomaly detection

Slide 33

Slide 33 text

@sebamontini MEDALLIA Visualization

Slide 34

Slide 34 text

@sebamontini MEDALLIA

Slide 35

Slide 35 text

@sebamontini MEDALLIA The big 5 ✓ Instrumentation → gauges, histograms, timers, counters ✓ Collection → pull vs push ✓ Storage → Time Series DB ✓ Alerting → threshold, flatline, delta, anomaly ✓ Visualization → dashboards

Slide 36

Slide 36 text

@sebamontini MEDALLIA

Slide 37

Slide 37 text

@sebamontini MEDALLIA The Four Golden Signals

Slide 38

Slide 38 text

@sebamontini MEDALLIA Latency Traffic Errors Saturation

Slide 39

Slide 39 text

@sebamontini MEDALLIA Thanks