Name	Name	Last commit message	Last commit date
parent directory ..
README.md	README.md

#########################################################################################

ADDENDA 0: Commands good to know

#########################################################################################

Here, I will bring you a few useful commands that you can use sometimes.

A. BASH

When you get more & more familiar with Kubernetes, you start wondering how to be more efficient in typing commands... The first step would to put a bunch of alias in the .bashrc file

Some examples:

$ cp ~/.bashrc ~/.bashrc.bak
$ cat <<EOT >> ~/.bashrc
alias kc='kubectl create'
alias kg='kubectl get'
alias kdel='kubectl delete'
alias kx='kubectl exec -it'
alias kdesc='kubectl describe'
alias kedit='kubectl edit'
alias trident='tridentctl -n trident'
EOT

Don't forget to type in bash in order to take the modifications into account

B. How can I easily list all the PVC of an application

If you are dealing with applications with lots of PODs & volumes, you may want to get a matrix which shows all the PVC per POD: The following example was used in the Scenario12, which presents StatefulSets.

$ kubectl get pods -n mysql -o=jsonpath='{range .items[*]}{"\n"}{.metadata.name}{":\t"}{range .spec.volumes[*].persistentVolumeClaim}{.claimName}{" "}{end}{end}{"\n"}'
mysql-0:        data-mysql-0
mysql-1:        data-mysql-1
mysql-2:        data-mysql-2

C. How can I easily list all the containers in a POD

As there is no option to do so, you need to extract this information from the POD definition. Here are some examples, in several Kubernetes versions

# 1.18 with Trident Operator (v20.07.1)
$ kubectl get pods -n trident -o=jsonpath='{range .items[*]}{"\n"}{.metadata.name}{":\t"}{range .spec.containers[*]}{.image}{", "}{end}{end}' |sort
trident-csi-68d979fb85-dsrmn:   netapp/trident:20.07.1, netapp/trident-autosupport:20.07.0, quay.io/k8scsi/csi-provisioner:v2.0.1, quay.io/k8scsi/csi-attacher:v2.2.0, quay.io/k8scsi/csi-resizer:v0.5.0, quay.io/k8scsi/csi-snapshotter:v2.1.1,
trident-csi-8jfhf:      netapp/trident:20.07.1, quay.io/k8scsi/csi-node-driver-registrar:v1.3.0,
trident-csi-jtnjz:      netapp/trident:20.07.1, quay.io/k8scsi/csi-node-driver-registrar:v1.3.0,
trident-csi-lcxvh:      netapp/trident:20.07.1, quay.io/k8scsi/csi-node-driver-registrar:v1.3.0,
trident-operator-7569744d4f-hwgnr:      netapp/trident-operator:20.07.1,

# 1.18 with Trident Operator (v21.01.0)
$ kubectl get pods -n trident -o=jsonpath='{range .items[*]}{"\n"}{.metadata.name}{":\t"}{range .spec.containers[*]}{.image}{", "}{end}{end}' |sort
trident-csi-58d557dcdf-zw699:   netapp/trident:21.01.0, netapp/trident-autosupport:21.01, k8s.gcr.io/sig-storage/csi-provisioner:v2.1.0, k8s.gcr.io/sig-storage/csi-attacher:v3.1.0, k8s.gcr.io/sig-storage/csi-resizer:v1.1.0, k8s.gcr.io/sig-storage/csi-snapshotter:v3.0.3,
trident-csi-ps7dq:      netapp/trident:21.01.0, k8s.gcr.io/sig-storage/csi-node-driver-registrar:v2.1.0,
trident-csi-t4kpn:      netapp/trident:21.01.0, k8s.gcr.io/sig-storage/csi-node-driver-registrar:v2.1.0,
trident-csi-x4l55:      netapp/trident:21.01.0, k8s.gcr.io/sig-storage/csi-node-driver-registrar:v2.1.0,
trident-operator-747c5ff65c-86x5s:      netapp/trident-operator:21.01.0,

What is interesting to notice is that when upgrading Kubernetes, new sidecars are added to CSI Trident:

Kubernetes 1.16: Volume Expansion (CSI Resizer) was promoted to Beta status (cf https://kubernetes-csi.github.io/docs/volume-expansion.html)
Kubernetes 1.17: Snapshot & Restore (CSI Snapshotter) was promoted to Beta status (cf https://kubernetes-csi.github.io/docs/snapshot-restore-feature.html)

D. How can I easily delete all PVC based on a filter

If you want to delete a large amount of PVC with one command, the easiest would be to put them all in one specific namespace.
Once you delete this namespace, all its PVC will also go away.

However, if you want to delete a few PVC based on a filter, you could use the following example, where I want to delete only the PVC containing the word thick:

$ kubectl get pvc
NAME          STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
thick-5gb-1   Bound    pvc-58f21eea-a9a1-4007-ab95-411bcbc6755c   5Gi        RWX            sc-eco-thick   20s
thick-5gb-2   Bound    pvc-9dbd3f76-7979-488f-bdf2-de587499974d   5Gi        RWX            sc-eco-thick   20s
thick-5gb-3   Bound    pvc-f456b9b6-90b9-40bc-bd0e-2fed36d77056   5Gi        RWX            sc-eco-thick   20s
thin-5gb-1    Bound    pvc-1d2fe395-4e50-4cc7-979c-597e5c608a21   5Gi        RWX            sc-eco-thin    2m31s
thin-5gb-2    Bound    pvc-1b44405d-177f-4b12-8f7c-1816242f9a8e   5Gi        RWX            sc-eco-thin    2m31s
thin-5gb-3    Bound    pvc-fda63953-ef16-45b4-986b-99544abbbd09   5Gi        RWX            sc-eco-thin    2m31s

$ kubectl get pvc -o name | grep thick | xargs kubectl delete
persistentvolumeclaim "thick-5gb-1" deleted
persistentvolumeclaim "thick-5gb-2" deleted
persistentvolumeclaim "thick-5gb-3" deleted

E. How can you get the state of all containers of a POD in one command

If for some reason you need to quickly access the state of each container of a POD (debug purpose), you could use the following command:

$ kubectl -n trident get pod trident-csi-676dd87bc6-69gc9 -o=jsonpath='{range .status.containerStatuses[*]}{"\n"}{.image}{":\tSTARTED "}{.started}{"\tSTATE:"}{.state}{end}{"\n"}'
quay.io/k8scsi/csi-attacher:v3.0.1:     STARTED true    STATE:map[running:map[startedAt:2020-11-12T12:46:51Z]]
quay.io/k8scsi/csi-provisioner:v2.0.3:  STARTED true    STATE:map[running:map[startedAt:2020-11-12T12:46:49Z]]
quay.io/k8scsi/csi-resizer:v1.0.1:      STARTED true    STATE:map[running:map[startedAt:2020-11-12T12:46:53Z]]
quay.io/k8scsi/csi-snapshotter:v3.0.1:  STARTED true    STATE:map[running:map[startedAt:2020-11-12T12:46:55Z]]
netapp/trident-autosupport:20.10.0:     STARTED true    STATE:map[running:map[startedAt:2020-11-12T12:46:47Z]]
netapp/trident:20.10.0:                 STARTED true    STATE:map[running:map[startedAt:2020-11-12T12:46:44Z]]

F. How to easily see what backends can be used by a storage class

I am using as an example a Storage Class called sc-topology, and would like to find with one single command what Trident Backends will be used.
This requires the unix command jq to be installed which is a JSON parser. (command to install it: yum install -y jq).

$ tridentctl -n trident get backend
+-----------------+-------------------+--------------------------------------+--------+---------+
|      NAME       |  STORAGE DRIVER   |                 UUID                 | STATE  | VOLUMES |
+-----------------+-------------------+--------------------------------------+--------+---------+
| nas-east        | ontap-nas         | 8f821381-cb00-4020-8b5e-64c3b3a0e5d7 | online |       0 |
| nas-west        | ontap-nas         | 34fd0cf8-bcbb-4fad-af3c-579e25efe8c0 | online |       0 |
| NAS_Vol-default | ontap-nas         | 8fab01ed-70b2-4e81-8e63-5ea05f5d312e | online |       1 |
| NAS_ECO-default | ontap-nas-economy | 8d8b25fe-d2ab-4416-b8e3-3d4a304b0aae | online |       0 |
+-----------------+-------------------+--------------------------------------+--------+---------+

$ tridentctl -n trident get storageclass sc-topology -o json | jq  '[.items[] | {storageClass: .Config.name, backends: [.storage]|unique}]'
[
  {
    "storageClass": "sc-topology",
    "backends": [
      {
        "nas-east": [
          "aggr1"
        ],
        "nas-west": [
          "aggr1"
        ]
...

In this very case, when provisioning a volume against the sc-topology storage class, Trident will choose between 2 different available backends.

G. How can I enable debug mode with the Trident operator

Starting with Trident 21.01, debug mode is disabled by default.
If you would like to enable it, you can simply patch the operator, step that will trigger restarting new pods.

$ kubectl describe torc | grep Spec: -A 1
Spec:
  Debug:      false

$ kubectl patch torc trident -n trident --type=merge -p '{"spec":{"debug":true}}'
tridentorchestrator.trident.netapp.io/trident patched

$ kubectl describe torc | grep Spec: -A 1
Spec:
  Debug:      true

H. How do I quickly retrieve CPU & Memory usage for my Kubernetes nodes

kubectl describe nodes | grep -A 3 -e "^\\s*Resource"
  Resource           Requests    Limits
  --------           --------    ------
  cpu                350m (17%)  0 (0%)
  memory             70Mi (4%)   170Mi (9%)
--
  Resource           Requests    Limits
  --------           --------    ------
  cpu                350m (17%)  0 (0%)
  memory             70Mi (4%)   170Mi (9%)
--
  Resource           Requests    Limits
  --------           --------    ------
  cpu                800m (40%)  0 (0%)
  memory             0 (0%)      0 (0%)

I. What's next

Back to the frontpage?

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Addenda00

Addenda00

README.md

ADDENDA 0: Commands good to know

A. BASH

B. How can I easily list all the PVC of an application

C. How can I easily list all the containers in a POD

D. How can I easily delete all PVC based on a filter

E. How can you get the state of all containers of a POD in one command

F. How to easily see what backends can be used by a storage class

G. How can I enable debug mode with the Trident operator

H. How do I quickly retrieve CPU & Memory usage for my Kubernetes nodes

I. What's next

Files

Addenda00

Directory actions

More options

Directory actions

More options

Latest commit

History

Addenda00

Folders and files

parent directory

README.md

ADDENDA 0: Commands good to know

A. BASH

B. How can I easily list all the PVC of an application

C. How can I easily list all the containers in a POD

D. How can I easily delete all PVC based on a filter

E. How can you get the state of all containers of a POD in one command

F. How to easily see what backends can be used by a storage class

G. How can I enable debug mode with the Trident operator

H. How do I quickly retrieve CPU & Memory usage for my Kubernetes nodes

I. What's next