nayakar – Best DevOps

Post test answers (nayakar)

nayakar — Sat, 18 Jun 2022 10:15:58 +0000

What is SRE? – 3-4 Lines
Ans. SRE are engineers who work on system monitoring, reliability, design, toil reduction, incident management, deployment and automation in infrastructure task. SRE also work closes with Developers in achieving their tasks. Their focus mostly lies within the client SLOs and automation tools and principles in engineering disciplines in serving those SLOs.
5 Action items to perform to transform from Ops to SRE
Ans.
– Collect data which are causing toil reduction and breaching SLAs.
– Analyze the risk and advantage in automating those tasks
– Involve with dev on infra availability and observibility
– Prioritize the tasks which can be easily automated and promote automation
– Bring in tools which can be used for better monitoring and observability to system.
Top 10 SRE tools to be implemented
Ans.
1. Datadog
2. Kubernetes
3. Terraform
4. RunDeck
5. Gitlab CI/CD
6. Kubernetes autoscaler
7. Ansible
8. Docker
9. Prometheus
10. Jira

About SRE – nayakar

nayakar — Thu, 16 Jun 2022 07:44:05 +0000

Define SRE in your own Words- 2 Lines
Ans. SRE is an approach to reduce manual work by toil reduction, automate the task which are repetitive. Also involved in managing monitoring stack, incident management and defining SLA/SLOs for applications.
Define 3-4 bullet points on migrating process of Ops to SRE?
Ans.

Understanding and list down the tasks which can be automated
Grouping and calculating manual efforts put by members in a month
Classifying the time saved and cost factors(if any) involved in migration or if adopting new tools
Prioritizing the tasks which can be automated with less effort first.

3. 3 Charactertics to identify Toil in SRE
Ans. Toil work will have below characteristics

Manual, Repetitive and can be automated
Following a defined set of actions for that work
No scope for learning new things

1 line explanation of SLI – SLO and SLA
Ans. SLI are indicators which we get from application and how they perform with a trend of data. SLO are objectives set based on SLI data and which can be achieved with defined SLIs. SLA and agreement set with client which we can achieve. SLOs are supposed to be lower than SLAs.

PSRE – Pretest (nayakar)

nayakar — Thu, 16 Jun 2022 04:07:30 +0000

Q1 – What are the differences between SRE and DevOps?
Ans. SRE and DevOps are closely related fields. SRE perspective is understanding and managing service availability and monitoring them by making use of tools or automation. Devops is focused more on release management and service automation tasks along through tools or scripts. SREs focus is on client SLA/SLOs, DevOps need not consider client SLA/SLOs.

Q2 – What SRE team is responsible for?
Ans. SRE team is responsible for application availability, make sure SLAs are met for client, monitoring stack for organization and tools and automation to achieve these aspects.

Q3 – What is an error budget?
Ans. An error budget is getting details of error rates for any particular application and calculating if it meets client expectations or if we breaching SLA.

Q4 – What are MTTF (mean time to failure) and MTTR (mean time to repair)? What these metrics help us to evaluate?
Ans. These metrics help us in evaluating and understanding incident metrics. Not quite sure of their specifications.

Q5 – What is the role of monitoring in SRE?
Ans. To monitor host resources like CPU,memory,disk and also application url monitoring along with slowness using APMs and opensource monitoring tools.

Q6 – How do you differentiate between process and thread?
Ans. Threads are child processes of a process. The PPID of a thread has process PID.

Q7 – What activity means Reducing Toil?
Ans. I don’t know

Q8 – Have you ever heard of SLO? If yes then explain.
Ans. SLOs are Service level objectives, these are objectives defined within application which help in meeting client SLAs.

Q9 – Enlist all the Linux signals you are aware of
Ans. kill and terminate

Q10 – What is obserbability
Ans. Observability is ability to get a complete view and understanding of systems available in an application stack and able to monitor and identify any failures.