You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
- openshift pod security mutations (added separate test for this)
- sandbox use of limit ranges for cpu/mem requests/limits
- deferral for now on requiring liveness / readiness probes
- pedantic host network checks (added separate test for this)
- argocd has to manipluate cluster scoped validatingwebhookconfigurations
- deferral on network policies for blocking ingress/egress
- chains/pac need to access secrets in arbitrary namespaces
add openshift scc / pod security check to tests
add openshift pod host network checks to tests
Copy file name to clipboardExpand all lines: ci/images/static-checks/content/config/checkov.yaml
+41-1Lines changed: 41 additions & 1 deletion
Original file line number
Diff line number
Diff line change
@@ -13,7 +13,47 @@ quiet: true
13
13
skip-check:
14
14
# skip Healthcheck instruction error for Docker Images
15
15
- CKV_DOCKER_2
16
+
# not enforcing liveness/readiness probes at this time; minimally, chains, results, metric exporter do not have
17
+
# we have opened https://github.com/tektoncd/results/issues/280 upstream
18
+
- CKV_K8S_8
19
+
- CKV_K8S_9
20
+
# RHTAP utilizes LimitRanges for cpu/mem requests and limits settings (handles 10-13)
21
+
- CKV_K8S_10
22
+
- CKV_K8S_11
23
+
- CKV_K8S_12
24
+
- CKV_K8S_13
25
+
# image ref related
26
+
- CKV_K8S_43 # deployments referenced by checkov are either items like chains which will be replace by openshift-pipelines 1.10 or a fooled by our use of kustomize for image setting
27
+
- CKV_K8S_14 # deployments referenced by checkov are either items like chains which will be replace by openshift-pipelines 1.10 or a fooled by our use of kustomize for image setting
28
+
- CKV_K8S_15 # with sha specific image refs setting pull policy to always is redundant and negates us of openshift node cache
29
+
# need to reivew chains/pac needs to read secrets in a couple of namespaces, not a clusterrolebinding, create webhooks
30
+
- CKV2_K8S_5
31
+
# there is no use of hostPID, hostIPC, hostNetwork in repo, but scan complains about not setting explicitly to false
32
+
# will check in live tests
33
+
- CKV_K8S_17
34
+
- CKV_K8S_18
35
+
- CKV_K8S_19
36
+
# openshift scc / security addresses these check by mutating pod under the covers
37
+
# with pods getting assigned the restricted scc unless explicitly allowed otherwise
38
+
- CKV_K8S_20 # no allowPrivilegeEscalation
39
+
- CKV_K8S_22 # read only FS
40
+
- CKV_K8S_23 # admission of root containers
41
+
- CKV_K8S_25 # we are not adding capabilities, running under restricted-scc
42
+
- CKV_K8S_28 # admission of NET RAW capability
43
+
- CKV_K8S_29 # apply security context to pod and containers
44
+
- CKV_K8S_30 # apply security context to containers
45
+
- CKV_K8S_31 # runtime/default seccomp profile
46
+
- CKV_K8S_33 # also, no kubernetes-dashboard on openshift
47
+
- CKV_K8S_37 # any capabilities
48
+
- CKV_K8S_38 # our pods almost always a) need to access api svr, b) do not have privileged SA
49
+
- CKV_K8S_40 # high UID number
50
+
- CKV_K8S_35 # opened https://github.com/tektoncd/results/issues/432 for secrets via env var
51
+
# need to allow argocd to create/delete the validatingadmissionwebhooks for tekton (core part of knative)
52
+
- CKV_K8S_155
53
+
- CKV_K8S_157
54
+
- CKV2_K8S_6 # use NetworkPolicy like what registration-service and integration-service employ are untenable for tekton controllers
0 commit comments