-
Notifications
You must be signed in to change notification settings - Fork 0
/
cloud_checklist.txt
47 lines (46 loc) · 2.54 KB
/
cloud_checklist.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
Courtesy of Arun
Architecture
How are your services laid out today?
How do they need to be defined in Kubernetes?
What remains in the datacenter, and how does it communicate to services based in cloud?
For kubernetes based services:
How should you scale the application?
Where are the processing bottlenecks and how do you scale those?
Implementation
What is your v1? What is your vShipReady?
What is the roadmap from current state to v1 to vShipReady? What do you need to do to get from current state to shipping?
Security
How are you securing the application and all services?
Describe authentication, authorization, and encryption
How are you securing communication in a hybrid scenario?
How are you securing communication with the database?
Are you encrypting data at rest?
What are you auditing and where does your audit log live?
Deployment
(assuming a pipeline) is deployment to production going to be manual?
What deployment strategy are you going to use, e.g. Blue/green? Canary? Etc?
What liveness checks do you use for kubernetes to validate successful instance deployment?
How do you trigger a rollback?
How are you going to ensure your schema can roll back as well?
Testing and Validation
How are you running unit tests and integration tests?
When are containers built and used in the testing process
What about other NFR tests?
How do you flag an entire application consisting of separate containers as ‘ready to deploy’?
What is your versioning scheme?
Monitoring and Alerting
What are the failure points in your application and how do you mitigate those failures?
What are the set of processes to remediate those failures?
What are other system indicators of pending failure that you can monitor?
What is not a failure but can indicate eventual failure?
Describe your monitoring architecture:
What are you going to monitor (see failure points and system indicators above)
What are the thresholds per monitor that you will alert on?
Remediation: for each failure please describe the remediation scenario:
What does the person on call need to do to remediate the failure, in simple to follow steps.
Game Day
How are you simulating failures and recoveries?
What are you validating when you simulate a failure?
Are your alerts alerting?
Do your logs indicate the failure cause and any other failure data?
Do your run books help restore service?