Update csi-attacher to v4.2.0

[sunnylovestiramisu]

What type of PR is this?

/kind cleanup

What this PR does / why we need it:

Update csi-attacher to https://github.com/kubernetes-csi/external-attacher/releases/tag/v4.2.0

Which issue(s) this PR fixes:

Fixes #

Special notes for your reviewer:

• The rbac file hasn't changed for 3 years.

• Changelog from 3.5.0 to 4.2.0 shows that the added flag is: --max-grpc-log-length. In external-attacher it is set to -1. But in the driver there is a default 10000. We need to set this value at the csi-attacher side.

• The breaking change from 3.x.x to 4.x.x is --default-fstype, and we need to set that value here. For windows, we've had an explicit storageclass that sets the fstype to ntfs, the default-fstype flag will not do anything for windows.

Does this PR introduce a user-facing change?:

Update csi-attacher to v4.2.0

[k8s-ci-robot]

Skipping CI for Draft Pull Request.
If you want CI signal for your change, please convert it to an actual PR.
You can still manually trigger a test run with /test all

[sunnylovestiramisu]

/ok-to-test

[sunnylovestiramisu]

/assign @leiyiz
/assign @saikat-royc

[msau42]

csi-attacher is the 2nd container: https://github.com/kubernetes-sigs/gcp-compute-persistent-disk-csi-driver/blob/cf727bbf6c79ee91aa77c27220f2b5babc00d1c1/deploy/kubernetes/base/controller/controller.yaml

[sunnylovestiramisu]

Updated, and the local build is:

  - args:
        - --v=5
        - --csi-address=/csi/csi.sock
        - --http-endpoint=:22012
        - --leader-election
        - --leader-election-namespace=$(PDCSI_NAMESPACE)
        - --timeout=250s
        - --max-grpc-log-length=10000
        env:
        - name: PDCSI_NAMESPACE
          valueFrom:
            fieldRef:
              fieldPath: metadata.namespace
        image: k8s.gcr.io/sig-storage/csi-attacher:v4.2.0
        livenessProbe:
          failureThreshold: 1
          httpGet:
            path: /healthz/leader-election
            port: http-endpoint
          initialDelaySeconds: 10
          periodSeconds: 20
          timeoutSeconds: 10
        name: csi-attacher
        ports:
        - containerPort: 22012
          name: http-endpoint
          protocol: TCP
        volumeMounts:
        - mountPath: /csi
          name: socket-di

[msau42]

The breaking change from 3.x.x to 4.x.x is --default-fstype, and we set that value here.

That is pointing to csi-provisioner arg, but this is for csi-attacher.

[sunnylovestiramisu]

Add the default-fstype and build the yaml again bin/kustomize build deploy/kubernetes/overlays/prow-stable-sidecar-rc-master:

  - args:
      - --v=5
      - --csi-address=/csi/csi.sock
      - --http-endpoint=:22012
      - --leader-election
      - --leader-election-namespace=$(PDCSI_NAMESPACE)
      - --timeout=250s
      - --max-grpc-log-length=10000
      - --default-fstype=ext4
      env:
      - name: PDCSI_NAMESPACE
        valueFrom:
          fieldRef:
            fieldPath: metadata.namespace
      image: k8s.gcr.io/sig-storage/csi-attacher:v4.2.0
      livenessProbe:
        failureThreshold: 1
        httpGet:
          path: /healthz/leader-election
          port: http-endpoint
        initialDelaySeconds: 10
        periodSeconds: 20
        timeoutSeconds: 10
      name: csi-attacher
      ports:
      - containerPort: 22012
        name: http-endpoint
        protocol: TCP
      volumeMounts:
      - mountPath: /csi
        name: socket-dir

[sunnylovestiramisu]

From the log:

W0216 17:18:26.167] $ gcloud auth login
E0216 17:18:26.167] Command failed

/retest

[sunnylovestiramisu]

e2e tests constantly failing with:

W0216 19:48:20.799] ERROR: (gcloud.auth.activate-service-account) There was a problem refreshing your current auth tokens: ('invalid_grant: Invalid JWT Signature.', {'error': 'invalid_grant', 'error_description': 'Invalid JWT Signature.'})

There is ongoing issue with the prow test: https://kubernetes.slack.com/archives/C09QZ4DQB/p1676561497236049

[sunnylovestiramisu]

The issue still exists: kubernetes/test-infra#27157 (comment)

It happens in a once per year base in the past: kubernetes/test-infra#27157

https://storage.googleapis.com/k8s-triage/index.html?pr=1&text=error%20during%20gcloud%20auth%20activate-service-account

[sunnylovestiramisu]

Key has been rotated.

/retest

[msau42]

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: msau42, sunnylovestiramisu

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [msau42]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment