Interview questions
From Federal Burro of Information
For the Candidate to ask the potential employer
for IT professional job
People
- how big is the team now.
- how big do you need it to be now, and in 6 month , 2 years , what is the vision, what does the pipeline look like. How many current open head count?
- what is the composition of the team level 1 2 3 , technology spread : OS / platform / infrastructure vs app vs OSS
- what does the management structure look like.
- how often are people trained.
- what is the individual training budget quarter. Do you have to sign anything that says you will stay if the company spends money on training on you?
- what does the PMO office look like.
- where did the people in the team come from? What type of back ground? IT? Oil Industry? Software Dev? Banks? ISP? How do they think? Are they engineers? Scientists?
The farm:
- number of clients
- number of environments
- number of service
- number of devices
- number of data centres
- number of cabinets
- agregate storage.
- virtual to physical spread
- OS spread
- Private versus public spread.
Infrastructure:
- What do you use for version control?
- inventory management: uCMDB? IPs Hardware software license contract
- do you use a tool to manage all that?
- centralized authentication? Redundancy?
- what sort of DNS / registration services do you have?
- storage infrastrcuture?
- NFS
- FC
- iSCSI
- Netapp / other? Off site storage? replication?
ITSM:
- is incident management separate from problem management?
- what does incident management look like?
- dedicate incident managers?
- is there a traige group?
- automated escalation?
- what tool do you use?
- how much dev resources do you allocate to the incident management systems?
- how do you ensure that ticket are properly populated? what is your coverage?
- what does the reporting look like for incident and problem management? by severity? by client? by staff memeber ? by technology?
- change control
- service desk?
Security:
- number and types of firewalls
- do you use IPS / IDS?
- What is the patching policy and how close do you get to the policy ? How do you measure variance?
- do you used centralized identify management: AAA , Clients? Staff?
- what does your incident reposne process look like? Can you show me a log if incidents?
Process:
- what does the deal development process look like? what does Vetting and sign off look like?
- what does the on call rotation look like now? number of people , how long are the cycles, what are the escalation levels.
Developemnt:
- Is there a project manager for each product? Or for a suite? Is there a technical product manager?
- what is the product developement process? Agile? DevOps? Does it follow any software developement models?
- How are products tested before release? ALpha program? Beta Program? Show me some history? Use cases? Feature lists? In Scope / Out of scope documents?
- How are new products decided on?
Pay:
- what is base salary?
- what does the Over time and on call schedule look like? minimum call , week day versus weekend / holiday?
- what does the bonus plan look like?
- can I get a copy of the benifits package?
- how much vacation and how do you get more vacation as you stay longer?
- vesting options ? Stock?
Excellence:
- what is is to you?
- is that important?
- how do you do that ?
- how do you deal with scope creep?
- where does scope creep come from?
Interpersonal:
- Tell me about a time you disagreed with someone.
- Tell me about a time you failed or made a mistake.
- Tell me about a time you showed leadership.
Dissent
- How do you deal with it?
- Challenges to
- Dealing with conflict.
- Examples
Answers
For potential employer to ask a candidate
Internet basics
- what is the name of a document where protocols are described.
Linux
- what is the difference between iptables and ipchains.
linux knowledge
what distro do you use and why?
netstat ( ss 0 versus lsof for network ) :
$ lsof -i -P -n | awk '{print $10}' | sort | uniq -c ; echo ; netstat -avpn -4 -6 --tcp | awk '{print $6}' | sort | uniq -c
135
2 (CLOSE_WAIT)
1093 (ESTABLISHED)
164 (LISTEN)
2 CLOSE_WAIT
1 established)
1143 ESTABLISHED
1 FIN_WAIT1
3 FIN_WAIT2
1 Foreign
177 LISTEN
6380 TIME_WAIT
why are these different?
Kubernetes Questions
Q: When and why use a daemonset versus deployment versus replica set.
DS: one per node. More nodes, more pods, less nodes? less pods. filebeat pods are DS.
Typically if you want a persistent set of pods up, you create a deployment that in-turn manages a replica set.
Q: Liveliness versus readiness.
OLD pre 1.20 k8s:
- ReadinessProbe should requests be routed to this pod? i.e. should it's endpoint be added to the service?
- LivelinessProbe is about checking to make sure the pod is _still_ alive. if thisfailed , restart the pod.
NEW 1.20 k8s and greater
- ReadinessProbe - should this pod be in the service?
- LivenessProbe - should I restart the pod?
- StartupProbe - Is a pod coming up finally ready to get traffic?
Q: You get a crash backoffloop how do you troubleshoot this?
A:looks at the pod log:
kubectl logs <pod>
Look at the system events
kubectl --namespace kube-system get events --sort-by='{.lastTimestamp}'
kubectl --namespace <application pod's namespace> get events --sort-by='{.lastTimestamp}'
Look at the description of the pod:
kubectl describe pod
Q: I want to get the performance of a pod or node
A: kubectl top pod / node
