Free NVIDIA NCP-AIO Actual Exam Questions - Question 4 Discussion

C, because --replicas is the correct flag for scaling resources in kubectl.

C sounds right since kubectl scale uses --replicas to set pod count. But I’m curious if the job’s parallelism field also needs adjusting to actually run 4 pods simultaneously?

C imo, the --replicas flag is the right way to scale a Job’s pods directly.

B tbh doesn’t fit because autoscale is for deployments, not jobs. A is definitely made-up. Between C and D, the flag -r isn’t actually valid in kubectl scale, so that rules out D. C uses the correct syntax and makes sense if you want to run 4 parallel pods for the Job. So even if scaling a Job affects completions or parallelism, C is the only legit command here.

C. Jobs aren’t like deployments where autoscaling applies, so B is out. And A uses a non-existent “stretch” command, so that’s a no. Between C and D, the official kubectl docs use --replicas, not -r, so C feels right. Plus, scaling a job to multiple replicas is just about setting the replicas count, which C does cleanly.

C/D but I’d toss D because -r isn’t a recognized flag for kubectl scale. The job resource can be scaled with the scale command, and --replicas=4 is definitely the standard syntax. Also, A’s “stretch” is not a real kubectl command, and B is about autoscaling deployments, not jobs. So C really makes the most sense here.

C imo, since scale is the right command and --replicas is the correct flag.

C/D? I’d drop D since -r isn’t a valid flag for scaling. A is definitely wrong because there's no "stretch" command in kubectl. B talks about autoscaling a deployment, which doesn’t fit a Job resource. So C seems like the only legit choice, especially since you can scale jobs with kubectl scale and the --replicas flag is the standard way to specify the count.

C/D? D’s shorthand flag looks off, so C’s full --replicas=4 seems more correct. The scale command is definitely the right approach here.

C vs D but D’s flag looks wrong; C uses the proper --replicas syntax.

C imo. Jobs can be scaled using kubectl scale, and the --replicas flag is the standard way. Also, the other options either use nonexistent commands like stretch or autoscale, which doesn’t apply to jobs directly. D’s -r flag definitely isn’t valid. So C fits both syntax and function here.

C. Jobs can be scaled using kubectl scale, and the --replicas flag is the proper syntax. Option D’s -r flag doesn’t exist, so it can’t be right.

Makes sense to go with C here. The scale command is the usual way to change replica counts, and --replicas=4 is the correct flag. Options A and B are clearly off since autoscale doesn’t apply to jobs and “stretch” isn’t a real command. D uses a short flag (-r) that doesn’t exist, so that’s out too. So C fits best with how Kubernetes scaling works for jobs.

Option C looks right because kubectl scale is the command to change replicas, and --replicas=4 matches the syntax. Options A and B mention commands or flags that don’t exist or don’t apply to jobs specifically. Option D uses -r which isn't a valid flag, so it can be ruled out pretty quickly. Jobs aren’t usually autoscaled like deployments, so B is off too.

C imo. Jobs aren’t designed to be scaled like deployments with autoscaling or stretch commands. The standard way is to use kubectl scale with the correct --replicas flag, so option C fits. Option D's -r flag isn’t valid for scaling. Also, A and B don’t seem right because "stretch" isn’t a kubectl command and autoscale applies to deployments or replicasets, not jobs.

This one seems kinda too simple. Isn't it just kubectl scale job -replicas=4 (option C)? But not sure if the syntax is exactly right for jobs, since they're different from deployments. Anyone confirm?