5 Commits
v0.2.3 ... main

Author SHA1 Message Date
Morten Olsen
6969f13fa7 multiarch build 2025-12-25 21:07:25 +01:00
Morten Olsen
7c014b9898 set defaults in docker instead of scanner 2025-12-13 08:21:40 +01:00
Morten Olsen
707f0dcaad fix: missing nuclei path 2025-12-13 08:19:34 +01:00
Morten Olsen
4c14e2294a fix: get templates before running scan 2025-12-13 07:56:28 +01:00
Morten Olsen
1677d02aa7 fix: run job in operator namespace to avoid permission issues 2025-12-12 21:49:15 +01:00
12 changed files with 712 additions and 326 deletions

View File

@@ -87,7 +87,7 @@ jobs:
uses: docker/build-push-action@v6 uses: docker/build-push-action@v6
with: with:
context: . context: .
# platforms: linux/amd64,linux/arm64 platforms: linux/amd64,linux/arm64
push: ${{ github.event_name != 'pull_request' }} push: ${{ github.event_name != 'pull_request' }}
tags: ${{ steps.meta.outputs.tags }} tags: ${{ steps.meta.outputs.tags }}
labels: ${{ steps.meta.outputs.labels }} labels: ${{ steps.meta.outputs.labels }}

View File

@@ -60,6 +60,7 @@ USER 65532:65532
# Environment variables for nuclei # Environment variables for nuclei
ENV NUCLEI_TEMPLATES_PATH=/nuclei-templates ENV NUCLEI_TEMPLATES_PATH=/nuclei-templates
ENV NUCLEI_BINARY_PATH=/usr/local/bin/nuclei
ENV HOME=/home/nonroot ENV HOME=/home/nonroot
ENTRYPOINT ["/manager"] ENTRYPOINT ["/manager"]

View File

@@ -77,6 +77,9 @@ type JobReference struct {
// Name of the Job // Name of the Job
Name string `json:"name"` Name string `json:"name"`
// Namespace of the Job (may differ from NucleiScan namespace)
Namespace string `json:"namespace"`
// UID of the Job // UID of the Job
UID string `json:"uid"` UID string `json:"uid"`

View File

@@ -5,8 +5,6 @@ metadata:
annotations: annotations:
controller-gen.kubebuilder.io/version: v0.19.0 controller-gen.kubebuilder.io/version: v0.19.0
name: nucleiscans.nuclei.homelab.mortenolsen.pro name: nucleiscans.nuclei.homelab.mortenolsen.pro
labels:
{{- include "nuclei-operator.labels" . | nindent 4 }}
spec: spec:
group: nuclei.homelab.mortenolsen.pro group: nuclei.homelab.mortenolsen.pro
names: names:
@@ -14,310 +12,456 @@ spec:
listKind: NucleiScanList listKind: NucleiScanList
plural: nucleiscans plural: nucleiscans
shortNames: shortNames:
- ns - ns
- nscan - nscan
singular: nucleiscan singular: nucleiscan
scope: Namespaced scope: Namespaced
versions: versions:
- additionalPrinterColumns: - additionalPrinterColumns:
- jsonPath: .status.phase - jsonPath: .status.phase
name: Phase name: Phase
type: string type: string
- jsonPath: .status.summary.totalFindings - jsonPath: .status.summary.totalFindings
name: Findings name: Findings
type: integer type: integer
- jsonPath: .spec.sourceRef.kind - jsonPath: .spec.sourceRef.kind
name: Source name: Source
type: string type: string
- jsonPath: .metadata.creationTimestamp - jsonPath: .metadata.creationTimestamp
name: Age name: Age
type: date type: date
name: v1alpha1 name: v1alpha1
schema: schema:
openAPIV3Schema: openAPIV3Schema:
description: NucleiScan is the Schema for the nucleiscans API description: NucleiScan is the Schema for the nucleiscans API
properties: properties:
apiVersion: apiVersion:
description: |- description: |-
APIVersion defines the versioned schema of this representation of an object. APIVersion defines the versioned schema of this representation of an object.
Servers should convert recognized schemas to the latest internal value, and Servers should convert recognized schemas to the latest internal value, and
may reject unrecognized values. may reject unrecognized values.
More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#resources
type: string type: string
kind: kind:
description: |- description: |-
Kind is a string value representing the REST resource this object represents. Kind is a string value representing the REST resource this object represents.
Servers may infer this from the endpoint the client submits requests to. Servers may infer this from the endpoint the client submits requests to.
Cannot be updated. Cannot be updated.
In CamelCase. In CamelCase.
More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds More info: https://git.k8s.io/community/contributors/devel/sig-architecture/api-conventions.md#types-kinds
type: string type: string
metadata: metadata:
type: object type: object
spec: spec:
description: NucleiScanSpec defines the desired state of NucleiScan description: NucleiScanSpec defines the desired state of NucleiScan
properties: properties:
schedule: scannerConfig:
description: |- description: ScannerConfig allows overriding scanner settings for
Schedule for periodic rescanning in cron format this scan
If empty, scan runs once properties:
type: string image:
severity: description: Image overrides the default scanner image
description: Severity filters scan results by severity level
items:
type: string type: string
type: array nodeSelector:
sourceRef: additionalProperties:
description:
SourceRef references the Ingress or VirtualService being
scanned
properties:
apiVersion:
description: APIVersion of the source resource
type: string type: string
kind: description: NodeSelector for scanner pod scheduling
description: Kind of the source resource - Ingress or VirtualService
enum:
- Ingress
- VirtualService
type: string
name:
description: Name of the source resource
type: string
namespace:
description: Namespace of the source resource
type: string
uid:
description: UID of the source resource for owner reference
type: string
required:
- apiVersion
- kind
- name
- namespace
- uid
type: object
suspend:
description: Suspend prevents scheduled scans from running
type: boolean
targets:
description:
Targets is the list of URLs to scan, extracted from the
source resource
items:
type: string
minItems: 1
type: array
templates:
description: |-
Templates specifies which Nuclei templates to use
If empty, uses default templates
items:
type: string
type: array
required:
- sourceRef
- targets
type: object
status:
description: NucleiScanStatus defines the observed state of NucleiScan
properties:
completionTime:
description: CompletionTime is when the last scan completed
format: date-time
type: string
conditions:
description: Conditions represent the latest available observations
items:
description:
Condition contains details for one aspect of the current
state of this API Resource.
properties:
lastTransitionTime:
description: |-
lastTransitionTime is the last time the condition transitioned from one status to another.
This should be when the underlying condition changed. If that is not known, then using the time when the API field changed is acceptable.
format: date-time
type: string
message:
description: |-
message is a human readable message indicating details about the transition.
This may be an empty string.
maxLength: 32768
type: string
observedGeneration:
description: |-
observedGeneration represents the .metadata.generation that the condition was set based upon.
For instance, if .metadata.generation is currently 12, but the .status.conditions[x].observedGeneration is 9, the condition is out of date
with respect to the current state of the instance.
format: int64
minimum: 0
type: integer
reason:
description: |-
reason contains a programmatic identifier indicating the reason for the condition's last transition.
Producers of specific condition types may define expected values and meanings for this field,
and whether the values are considered a guaranteed API.
The value should be a CamelCase string.
This field may not be empty.
maxLength: 1024
minLength: 1
pattern: ^[A-Za-z]([A-Za-z0-9_,:]*[A-Za-z0-9_])?$
type: string
status:
description: status of the condition, one of True, False, Unknown.
enum:
- "True"
- "False"
- Unknown
type: string
type:
description: type of condition in CamelCase or in foo.example.com/CamelCase.
maxLength: 316
pattern: ^([a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*/)?(([A-Za-z0-9][-A-Za-z0-9_.]*)?[A-Za-z0-9])$
type: string
required:
- lastTransitionTime
- message
- reason
- status
- type
type: object type: object
type: array resources:
x-kubernetes-list-map-keys: description: Resources defines resource requirements for the scanner
- type pod
x-kubernetes-list-type: map
findings:
description: |-
Findings contains the array of scan results from Nuclei JSONL output
Each element is a parsed JSON object from Nuclei output
items:
description: Finding represents a single Nuclei scan finding
properties: properties:
description: claims:
description: Description provides details about the finding description: |-
type: string Claims lists the names of resources, defined in spec.resourceClaims,
extractedResults: that are used by this container.
description:
ExtractedResults contains any data extracted by This field depends on the
the template DynamicResourceAllocation feature gate.
This field is immutable. It can only be set for containers.
items: items:
type: string description: ResourceClaim references one entry in PodSpec.ResourceClaims.
properties:
name:
description: |-
Name must match the name of one entry in pod.spec.resourceClaims of
the Pod where this field is used. It makes that resource available
inside a container.
type: string
request:
description: |-
Request is the name chosen for a request in the referenced claim.
If empty, everything from the claim is made available, otherwise
only the result of this request.
type: string
required:
- name
type: object
type: array type: array
host: x-kubernetes-list-map-keys:
description: Host that was scanned - name
type: string x-kubernetes-list-type: map
matchedAt: limits:
description: additionalProperties:
MatchedAt is the specific URL or endpoint where anyOf:
the issue was found - type: integer
type: string - type: string
metadata: pattern: ^(\+|-)?(([0-9]+(\.[0-9]*)?)|(\.[0-9]+))(([KMGTPE]i)|[numkMGTPE]|([eE](\+|-)?(([0-9]+(\.[0-9]*)?)|(\.[0-9]+))))?$
description: Metadata contains additional template metadata x-kubernetes-int-or-string: true
description: |-
Limits describes the maximum amount of compute resources allowed.
More info: https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/
type: object
requests:
additionalProperties:
anyOf:
- type: integer
- type: string
pattern: ^(\+|-)?(([0-9]+(\.[0-9]*)?)|(\.[0-9]+))(([KMGTPE]i)|[numkMGTPE]|([eE](\+|-)?(([0-9]+(\.[0-9]*)?)|(\.[0-9]+))))?$
x-kubernetes-int-or-string: true
description: |-
Requests describes the minimum amount of compute resources required.
If Requests is omitted for a container, it defaults to Limits if that is explicitly specified,
otherwise to an implementation-defined value. Requests cannot exceed Limits.
More info: https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/
type: object type: object
x-kubernetes-preserve-unknown-fields: true
reference:
description:
Reference contains URLs to additional information
about the finding
items:
type: string
type: array
severity:
description: Severity of the finding
type: string
tags:
description: Tags associated with the finding
items:
type: string
type: array
templateId:
description: TemplateID is the Nuclei template identifier
type: string
templateName:
description: TemplateName is the human-readable template name
type: string
timestamp:
description: Timestamp when the finding was discovered
format: date-time
type: string
type:
description: Type of the finding - http, dns, ssl, etc.
type: string
required:
- host
- severity
- templateId
- timestamp
type: object type: object
type: array templateURLs:
lastError: description: TemplateURLs specifies additional template repositories
description: LastError contains the error message if the scan failed to clone
type: string items:
lastRetryTime: type: string
description: type: array
LastRetryTime is when the last availability check retry timeout:
occurred description: Timeout overrides the default scan timeout
format: date-time type: string
type: string tolerations:
lastScanTime: description: Tolerations for scanner pod scheduling
description: LastScanTime is when the last scan was initiated items:
format: date-time description: |-
type: string The pod this Toleration is attached to tolerates any taint that matches
nextScheduledTime: the triple <key,value,effect> using the matching operator <operator>.
description: properties:
NextScheduledTime is when the next scheduled scan will effect:
run description: |-
format: date-time Effect indicates the taint effect to match. Empty means match all taint effects.
type: string When specified, allowed values are NoSchedule, PreferNoSchedule and NoExecute.
observedGeneration: type: string
description: key:
ObservedGeneration is the generation observed by the description: |-
controller Key is the taint key that the toleration applies to. Empty means match all taint keys.
format: int64 If the key is empty, operator must be Exists; this combination means to match all values and all keys.
type: integer type: string
phase: operator:
description: Phase represents the current scan phase description: |-
enum: Operator represents a key's relationship to the value.
- Pending Valid operators are Exists and Equal. Defaults to Equal.
- Running Exists is equivalent to wildcard for value, so that a pod can
- Completed tolerate all taints of a particular category.
- Failed type: string
type: string tolerationSeconds:
retryCount: description: |-
description: |- TolerationSeconds represents the period of time the toleration (which must be
RetryCount tracks the number of consecutive availability check retries of effect NoExecute, otherwise this field is ignored) tolerates the taint. By default,
Used for exponential backoff when waiting for targets it is not set, which means tolerate the taint forever (do not evict). Zero and
type: integer negative values will be treated as 0 (evict immediately) by the system.
summary: format: int64
description: Summary provides aggregated scan statistics type: integer
properties: value:
durationSeconds: description: |-
description: DurationSeconds is the duration of the scan in seconds Value is the taint value the toleration matches to.
format: int64 If the operator is Exists, the value should be empty, otherwise just a regular string.
type: integer type: string
findingsBySeverity:
additionalProperties:
type: integer
description:
FindingsBySeverity breaks down findings by severity
level
type: object type: object
targetsScanned: type: array
description: type: object
TargetsScanned is the number of targets that were schedule:
scanned description: |-
type: integer Schedule for periodic rescanning in cron format
totalFindings: If empty, scan runs once
description: TotalFindings is the total number of findings type: string
severity:
description: Severity filters scan results by severity level
enum:
- info
- low
- medium
- high
- critical
items:
type: string
type: array
sourceRef:
description: SourceRef references the Ingress or VirtualService being
scanned
properties:
apiVersion:
description: APIVersion of the source resource
type: string
kind:
description: Kind of the source resource - Ingress or VirtualService
enum:
- Ingress
- VirtualService
type: string
name:
description: Name of the source resource
type: string
namespace:
description: Namespace of the source resource
type: string
uid:
description: UID of the source resource for owner reference
type: string
required:
- apiVersion
- kind
- name
- namespace
- uid
type: object
suspend:
description: Suspend prevents scheduled scans from running
type: boolean
targets:
description: Targets is the list of URLs to scan, extracted from the
source resource
items:
type: string
minItems: 1
type: array
templates:
description: |-
Templates specifies which Nuclei templates to use
If empty, uses default templates
items:
type: string
type: array
required:
- sourceRef
- targets
type: object
status:
description: NucleiScanStatus defines the observed state of NucleiScan
properties:
completionTime:
description: CompletionTime is when the last scan completed
format: date-time
type: string
conditions:
description: Conditions represent the latest available observations
items:
description: Condition contains details for one aspect of the current
state of this API Resource.
properties:
lastTransitionTime:
description: |-
lastTransitionTime is the last time the condition transitioned from one status to another.
This should be when the underlying condition changed. If that is not known, then using the time when the API field changed is acceptable.
format: date-time
type: string
message:
description: |-
message is a human readable message indicating details about the transition.
This may be an empty string.
maxLength: 32768
type: string
observedGeneration:
description: |-
observedGeneration represents the .metadata.generation that the condition was set based upon.
For instance, if .metadata.generation is currently 12, but the .status.conditions[x].observedGeneration is 9, the condition is out of date
with respect to the current state of the instance.
format: int64
minimum: 0
type: integer type: integer
reason:
description: |-
reason contains a programmatic identifier indicating the reason for the condition's last transition.
Producers of specific condition types may define expected values and meanings for this field,
and whether the values are considered a guaranteed API.
The value should be a CamelCase string.
This field may not be empty.
maxLength: 1024
minLength: 1
pattern: ^[A-Za-z]([A-Za-z0-9_,:]*[A-Za-z0-9_])?$
type: string
status:
description: status of the condition, one of True, False, Unknown.
enum:
- "True"
- "False"
- Unknown
type: string
type:
description: type of condition in CamelCase or in foo.example.com/CamelCase.
maxLength: 316
pattern: ^([a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*/)?(([A-Za-z0-9][-A-Za-z0-9_.]*)?[A-Za-z0-9])$
type: string
required: required:
- targetsScanned - lastTransitionTime
- totalFindings - message
- reason
- status
- type
type: object type: object
type: object type: array
type: object x-kubernetes-list-map-keys:
served: true - type
storage: true x-kubernetes-list-type: map
subresources: findings:
status: {} description: |-
Findings contains the array of scan results from Nuclei JSONL output
Each element is a parsed JSON object from Nuclei output
items:
description: Finding represents a single Nuclei scan finding
properties:
description:
description: Description provides details about the finding
type: string
extractedResults:
description: ExtractedResults contains any data extracted by
the template
items:
type: string
type: array
host:
description: Host that was scanned
type: string
matchedAt:
description: MatchedAt is the specific URL or endpoint where
the issue was found
type: string
metadata:
description: Metadata contains additional template metadata
type: object
x-kubernetes-preserve-unknown-fields: true
reference:
description: Reference contains URLs to additional information
about the finding
items:
type: string
type: array
severity:
description: Severity of the finding
type: string
tags:
description: Tags associated with the finding
items:
type: string
type: array
templateId:
description: TemplateID is the Nuclei template identifier
type: string
templateName:
description: TemplateName is the human-readable template name
type: string
timestamp:
description: Timestamp when the finding was discovered
format: date-time
type: string
type:
description: Type of the finding - http, dns, ssl, etc.
type: string
required:
- host
- severity
- templateId
- timestamp
type: object
type: array
jobRef:
description: JobRef references the current or last scanner job
properties:
name:
description: Name of the Job
type: string
namespace:
description: Namespace of the Job (may differ from NucleiScan
namespace)
type: string
podName:
description: PodName is the name of the scanner pod (for log retrieval)
type: string
startTime:
description: StartTime when the job was created
format: date-time
type: string
uid:
description: UID of the Job
type: string
required:
- name
- namespace
- uid
type: object
lastError:
description: LastError contains the error message if the scan failed
type: string
lastRetryTime:
description: LastRetryTime is when the last availability check retry
occurred
format: date-time
type: string
lastScanTime:
description: LastScanTime is when the last scan was initiated
format: date-time
type: string
nextScheduledTime:
description: NextScheduledTime is when the next scheduled scan will
run
format: date-time
type: string
observedGeneration:
description: ObservedGeneration is the generation observed by the
controller
format: int64
type: integer
phase:
description: Phase represents the current scan phase
enum:
- Pending
- Running
- Completed
- Failed
type: string
retryCount:
description: |-
RetryCount tracks the number of consecutive availability check retries
Used for exponential backoff when waiting for targets
type: integer
scanStartTime:
description: ScanStartTime is when the scanner pod actually started
scanning
format: date-time
type: string
summary:
description: Summary provides aggregated scan statistics
properties:
durationSeconds:
description: DurationSeconds is the duration of the scan in seconds
format: int64
type: integer
findingsBySeverity:
additionalProperties:
type: integer
description: FindingsBySeverity breaks down findings by severity
level
type: object
targetsScanned:
description: TargetsScanned is the number of targets that were
scanned
type: integer
totalFindings:
description: TotalFindings is the total number of findings
type: integer
required:
- targetsScanned
- totalFindings
type: object
type: object
type: object
served: true
storage: true
subresources:
status: {}

View File

@@ -70,6 +70,8 @@ spec:
value: {{ .Values.scanner.ttlAfterFinished | quote }} value: {{ .Values.scanner.ttlAfterFinished | quote }}
- name: SCANNER_SERVICE_ACCOUNT - name: SCANNER_SERVICE_ACCOUNT
value: {{ include "nuclei-operator.fullname" . }}-scanner value: {{ include "nuclei-operator.fullname" . }}-scanner
- name: OPERATOR_NAMESPACE
value: {{ .Release.Namespace | quote }}
{{- if .Values.scanner.defaultTemplates }} {{- if .Values.scanner.defaultTemplates }}
- name: DEFAULT_TEMPLATES - name: DEFAULT_TEMPLATES
value: {{ join "," .Values.scanner.defaultTemplates | quote }} value: {{ join "," .Values.scanner.defaultTemplates | quote }}

View File

@@ -241,6 +241,16 @@ func main() {
scannerServiceAccount = "nuclei-scanner" scannerServiceAccount = "nuclei-scanner"
} }
operatorNamespace := os.Getenv("OPERATOR_NAMESPACE")
if operatorNamespace == "" {
// Try to read from the downward API file
if data, err := os.ReadFile("/var/run/secrets/kubernetes.io/serviceaccount/namespace"); err == nil {
operatorNamespace = string(data)
} else {
operatorNamespace = "nuclei-operator-system"
}
}
defaultTemplates := []string{} defaultTemplates := []string{}
if v := os.Getenv("DEFAULT_TEMPLATES"); v != "" { if v := os.Getenv("DEFAULT_TEMPLATES"); v != "" {
defaultTemplates = strings.Split(v, ",") defaultTemplates = strings.Split(v, ",")
@@ -259,6 +269,7 @@ func main() {
BackoffLimit: 2, BackoffLimit: 2,
MaxConcurrent: maxConcurrentScans, MaxConcurrent: maxConcurrentScans,
ServiceAccountName: scannerServiceAccount, ServiceAccountName: scannerServiceAccount,
OperatorNamespace: operatorNamespace,
DefaultResources: jobmanager.DefaultConfig().DefaultResources, DefaultResources: jobmanager.DefaultConfig().DefaultResources,
DefaultTemplates: defaultTemplates, DefaultTemplates: defaultTemplates,
DefaultSeverity: defaultSeverity, DefaultSeverity: defaultSeverity,

View File

@@ -376,6 +376,10 @@ spec:
name: name:
description: Name of the Job description: Name of the Job
type: string type: string
namespace:
description: Namespace of the Job (may differ from NucleiScan
namespace)
type: string
podName: podName:
description: PodName is the name of the scanner pod (for log retrieval) description: PodName is the name of the scanner pod (for log retrieval)
type: string type: string
@@ -388,6 +392,7 @@ spec:
type: string type: string
required: required:
- name - name
- namespace
- uid - uid
type: object type: object
lastError: lastError:

View File

@@ -257,8 +257,13 @@ func (r *NucleiScanReconciler) handleDeletion(ctx context.Context, nucleiScan *n
// Clean up any running scanner job // Clean up any running scanner job
if nucleiScan.Status.JobRef != nil { if nucleiScan.Status.JobRef != nil {
log.Info("Deleting scanner job", "job", nucleiScan.Status.JobRef.Name) jobNamespace := nucleiScan.Status.JobRef.Namespace
if err := r.JobManager.DeleteJob(ctx, nucleiScan.Status.JobRef.Name, nucleiScan.Namespace); err != nil { if jobNamespace == "" {
// Fallback for backwards compatibility
jobNamespace = nucleiScan.Namespace
}
log.Info("Deleting scanner job", "job", nucleiScan.Status.JobRef.Name, "namespace", jobNamespace)
if err := r.JobManager.DeleteJob(ctx, nucleiScan.Status.JobRef.Name, jobNamespace); err != nil {
if !apierrors.IsNotFound(err) { if !apierrors.IsNotFound(err) {
log.Error(err, "Failed to delete scanner job", "job", nucleiScan.Status.JobRef.Name) log.Error(err, "Failed to delete scanner job", "job", nucleiScan.Status.JobRef.Name)
} }
@@ -324,6 +329,7 @@ func (r *NucleiScanReconciler) handlePendingPhase(ctx context.Context, nucleiSca
nucleiScan.Status.Phase = nucleiv1alpha1.ScanPhaseRunning nucleiScan.Status.Phase = nucleiv1alpha1.ScanPhaseRunning
nucleiScan.Status.JobRef = &nucleiv1alpha1.JobReference{ nucleiScan.Status.JobRef = &nucleiv1alpha1.JobReference{
Name: job.Name, Name: job.Name,
Namespace: job.Namespace,
UID: string(job.UID), UID: string(job.UID),
StartTime: &now, StartTime: &now,
} }
@@ -401,8 +407,13 @@ func (r *NucleiScanReconciler) handleRunningPhase(ctx context.Context, nucleiSca
return ctrl.Result{Requeue: true}, nil return ctrl.Result{Requeue: true}, nil
} }
// Get the job // Get the job - use namespace from JobRef (may be different from scan namespace)
job, err := r.JobManager.GetJob(ctx, nucleiScan.Status.JobRef.Name, nucleiScan.Namespace) jobNamespace := nucleiScan.Status.JobRef.Namespace
if jobNamespace == "" {
// Fallback for backwards compatibility
jobNamespace = nucleiScan.Namespace
}
job, err := r.JobManager.GetJob(ctx, nucleiScan.Status.JobRef.Name, jobNamespace)
if err != nil { if err != nil {
if apierrors.IsNotFound(err) { if apierrors.IsNotFound(err) {
logger.Info("Scanner job not found, resetting to Pending") logger.Info("Scanner job not found, resetting to Pending")

View File

@@ -82,6 +82,9 @@ type Config struct {
// ServiceAccountName is the service account to use for scanner pods // ServiceAccountName is the service account to use for scanner pods
ServiceAccountName string ServiceAccountName string
// OperatorNamespace is the namespace where the operator runs and where scanner jobs will be created
OperatorNamespace string
// DefaultResources are the default resource requirements for scanner pods // DefaultResources are the default resource requirements for scanner pods
DefaultResources corev1.ResourceRequirements DefaultResources corev1.ResourceRequirements
@@ -101,6 +104,7 @@ func DefaultConfig() Config {
BackoffLimit: DefaultBackoffLimit, BackoffLimit: DefaultBackoffLimit,
MaxConcurrent: 5, MaxConcurrent: 5,
ServiceAccountName: "nuclei-scanner", ServiceAccountName: "nuclei-scanner",
OperatorNamespace: "nuclei-operator-system",
DefaultResources: corev1.ResourceRequirements{ DefaultResources: corev1.ResourceRequirements{
Requests: corev1.ResourceList{ Requests: corev1.ResourceList{
corev1.ResourceCPU: resource.MustParse("100m"), corev1.ResourceCPU: resource.MustParse("100m"),
@@ -136,14 +140,22 @@ func (m *JobManager) CreateScanJob(ctx context.Context, scan *nucleiv1alpha1.Nuc
job := m.buildJob(scan) job := m.buildJob(scan)
// Set owner reference so the job is garbage collected when the scan is deleted // Only set owner reference if the job is in the same namespace as the scan
if err := controllerutil.SetControllerReference(scan, job, m.Scheme); err != nil { // Cross-namespace owner references are not allowed in Kubernetes
return nil, fmt.Errorf("failed to set controller reference: %w", err) if job.Namespace == scan.Namespace {
if err := controllerutil.SetControllerReference(scan, job, m.Scheme); err != nil {
return nil, fmt.Errorf("failed to set controller reference: %w", err)
}
} }
// When job is in a different namespace (operator namespace), we rely on:
// 1. TTLSecondsAfterFinished for automatic cleanup of completed jobs
// 2. Labels (LabelScanName, LabelScanNamespace) to track which scan the job belongs to
// 3. CleanupOrphanedJobs to clean up jobs whose scans no longer exist
logger.Info("Creating scanner job", logger.Info("Creating scanner job",
"job", job.Name, "job", job.Name,
"namespace", job.Namespace, "jobNamespace", job.Namespace,
"scanNamespace", scan.Namespace,
"image", job.Spec.Template.Spec.Containers[0].Image, "image", job.Spec.Template.Spec.Containers[0].Image,
"targets", len(scan.Spec.Targets)) "targets", len(scan.Spec.Targets))
@@ -277,14 +289,41 @@ func (m *JobManager) CleanupOrphanedJobs(ctx context.Context) error {
} }
for _, job := range jobList.Items { for _, job := range jobList.Items {
// Check if owner reference exists and the owner still exists // Check if the associated NucleiScan still exists using labels
ownerRef := metav1.GetControllerOf(&job) scanName := job.Labels[LabelScanName]
if ownerRef == nil { scanNamespace := job.Labels[LabelScanNamespace]
logger.Info("Deleting orphaned job without owner", "job", job.Name, "namespace", job.Namespace)
if err := m.DeleteJob(ctx, job.Name, job.Namespace); err != nil && !apierrors.IsNotFound(err) { if scanName != "" && scanNamespace != "" {
logger.Error(err, "Failed to delete orphaned job", "job", job.Name) // Try to get the associated NucleiScan
scan := &nucleiv1alpha1.NucleiScan{}
err := m.Get(ctx, types.NamespacedName{Name: scanName, Namespace: scanNamespace}, scan)
if err != nil {
if apierrors.IsNotFound(err) {
// The scan no longer exists - delete the job
logger.Info("Deleting orphaned job (scan not found)",
"job", job.Name,
"namespace", job.Namespace,
"scanName", scanName,
"scanNamespace", scanNamespace)
if err := m.DeleteJob(ctx, job.Name, job.Namespace); err != nil && !apierrors.IsNotFound(err) {
logger.Error(err, "Failed to delete orphaned job", "job", job.Name)
}
continue
}
// Other error - log and continue
logger.Error(err, "Failed to check if scan exists", "scanName", scanName, "scanNamespace", scanNamespace)
continue
}
} else {
// Job doesn't have proper labels - check owner reference as fallback
ownerRef := metav1.GetControllerOf(&job)
if ownerRef == nil {
logger.Info("Deleting orphaned job without owner or labels", "job", job.Name, "namespace", job.Namespace)
if err := m.DeleteJob(ctx, job.Name, job.Namespace); err != nil && !apierrors.IsNotFound(err) {
logger.Error(err, "Failed to delete orphaned job", "job", job.Name)
}
continue
} }
continue
} }
// Check if the job is stuck (running longer than 2x the timeout) // Check if the job is stuck (running longer than 2x the timeout)
@@ -305,12 +344,18 @@ func (m *JobManager) CleanupOrphanedJobs(ctx context.Context) error {
// buildJob creates a Job specification for the given NucleiScan // buildJob creates a Job specification for the given NucleiScan
func (m *JobManager) buildJob(scan *nucleiv1alpha1.NucleiScan) *batchv1.Job { func (m *JobManager) buildJob(scan *nucleiv1alpha1.NucleiScan) *batchv1.Job {
// Generate a unique job name // Generate a unique job name that includes the scan namespace to avoid collisions
jobName := fmt.Sprintf("nucleiscan-%s-%d", scan.Name, time.Now().Unix()) jobName := fmt.Sprintf("nucleiscan-%s-%s-%d", scan.Namespace, scan.Name, time.Now().Unix())
if len(jobName) > 63 { if len(jobName) > 63 {
jobName = jobName[:63] jobName = jobName[:63]
} }
// Determine the namespace for the job - use operator namespace if configured
jobNamespace := m.Config.OperatorNamespace
if jobNamespace == "" {
jobNamespace = scan.Namespace
}
// Determine the scanner image // Determine the scanner image
image := m.Config.ScannerImage image := m.Config.ScannerImage
if scan.Spec.ScannerConfig != nil && scan.Spec.ScannerConfig.Image != "" { if scan.Spec.ScannerConfig != nil && scan.Spec.ScannerConfig.Image != "" {
@@ -360,7 +405,7 @@ func (m *JobManager) buildJob(scan *nucleiv1alpha1.NucleiScan) *batchv1.Job {
job := &batchv1.Job{ job := &batchv1.Job{
ObjectMeta: metav1.ObjectMeta{ ObjectMeta: metav1.ObjectMeta{
Name: jobName, Name: jobName,
Namespace: scan.Namespace, Namespace: jobNamespace,
Labels: labels, Labels: labels,
}, },
Spec: batchv1.JobSpec{ Spec: batchv1.JobSpec{

View File

@@ -43,14 +43,22 @@ func TestBuildJob(t *testing.T) {
job := manager.buildJob(scan) job := manager.buildJob(scan)
// Verify job name prefix // Verify job name prefix - should include scan namespace to avoid collisions
if len(job.Name) == 0 { if len(job.Name) == 0 {
t.Error("Job name should not be empty") t.Error("Job name should not be empty")
} }
// Verify namespace // Verify namespace - job should be created in operator namespace
if job.Namespace != "default" { if job.Namespace != config.OperatorNamespace {
t.Errorf("Expected namespace 'default', got '%s'", job.Namespace) t.Errorf("Expected namespace '%s', got '%s'", config.OperatorNamespace, job.Namespace)
}
// Verify scan labels are set correctly for cross-namespace tracking
if job.Labels[LabelScanName] != scan.Name {
t.Errorf("Expected scan name label '%s', got '%s'", scan.Name, job.Labels[LabelScanName])
}
if job.Labels[LabelScanNamespace] != scan.Namespace {
t.Errorf("Expected scan namespace label '%s', got '%s'", scan.Namespace, job.Labels[LabelScanNamespace])
} }
// Verify labels // Verify labels
@@ -115,3 +123,29 @@ func TestBuildJobWithCustomConfig(t *testing.T) {
t.Errorf("Expected deadline %d, got %d", expectedDeadline, *job.Spec.ActiveDeadlineSeconds) t.Errorf("Expected deadline %d, got %d", expectedDeadline, *job.Spec.ActiveDeadlineSeconds)
} }
} }
func TestBuildJobInSameNamespace(t *testing.T) {
config := DefaultConfig()
// Clear operator namespace to test same-namespace behavior
config.OperatorNamespace = ""
manager := &JobManager{
Config: config,
}
scan := &nucleiv1alpha1.NucleiScan{
ObjectMeta: metav1.ObjectMeta{
Name: "test-scan",
Namespace: "my-namespace",
},
Spec: nucleiv1alpha1.NucleiScanSpec{
Targets: []string{"https://example.com"},
},
}
job := manager.buildJob(scan)
// Verify namespace - when operator namespace is empty, job should be in scan's namespace
if job.Namespace != scan.Namespace {
t.Errorf("Expected namespace '%s', got '%s'", scan.Namespace, job.Namespace)
}
}

View File

@@ -118,8 +118,11 @@ func (r *Runner) Run(ctx context.Context) error {
logger.Info("Starting scan", logger.Info("Starting scan",
"targets", len(scan.Spec.Targets), "targets", len(scan.Spec.Targets),
"targetList", scan.Spec.Targets,
"templates", scan.Spec.Templates, "templates", scan.Spec.Templates,
"severity", scan.Spec.Severity) "templatesCount", len(scan.Spec.Templates),
"severity", scan.Spec.Severity,
"severityCount", len(scan.Spec.Severity))
// Update status to indicate scan has started // Update status to indicate scan has started
startTime := metav1.Now() startTime := metav1.Now()

View File

@@ -26,6 +26,8 @@ import (
"strings" "strings"
"time" "time"
"sigs.k8s.io/controller-runtime/pkg/log"
nucleiv1alpha1 "github.com/mortenolsen/nuclei-operator/api/v1alpha1" nucleiv1alpha1 "github.com/mortenolsen/nuclei-operator/api/v1alpha1"
) )
@@ -95,6 +97,8 @@ func NewNucleiScannerWithDefaults() *NucleiScanner {
// Scan executes a Nuclei scan against the given targets // Scan executes a Nuclei scan against the given targets
func (s *NucleiScanner) Scan(ctx context.Context, targets []string, options ScanOptions) (*ScanResult, error) { func (s *NucleiScanner) Scan(ctx context.Context, targets []string, options ScanOptions) (*ScanResult, error) {
logger := log.FromContext(ctx).WithName("nuclei-scanner")
if len(targets) == 0 { if len(targets) == 0 {
return nil, fmt.Errorf("no targets provided for scan") return nil, fmt.Errorf("no targets provided for scan")
} }
@@ -110,10 +114,66 @@ func (s *NucleiScanner) Scan(ctx context.Context, targets []string, options Scan
// Write targets to a file // Write targets to a file
targetsFile := filepath.Join(tmpDir, "targets.txt") targetsFile := filepath.Join(tmpDir, "targets.txt")
if err := os.WriteFile(targetsFile, []byte(strings.Join(targets, "\n")), 0600); err != nil { targetsContent := strings.Join(targets, "\n")
if err := os.WriteFile(targetsFile, []byte(targetsContent), 0600); err != nil {
return nil, fmt.Errorf("failed to write targets file: %w", err) return nil, fmt.Errorf("failed to write targets file: %w", err)
} }
logger.Info("Targets file created", "targetsFile", targetsFile, "targetCount", len(targets))
// Check if nuclei binary exists and is executable
// First try the exact path
if _, err := os.Stat(s.nucleiBinaryPath); os.IsNotExist(err) {
// If not found at exact path, try to find it in PATH
if path, err := exec.LookPath(s.nucleiBinaryPath); err == nil {
logger.Info("Found nuclei binary in PATH", "path", path, "originalPath", s.nucleiBinaryPath)
s.nucleiBinaryPath = path
} else {
return nil, fmt.Errorf("nuclei binary not found at %s and not in PATH: %w", s.nucleiBinaryPath, err)
}
}
// Verify nuclei is executable by running version command
if err := exec.Command(s.nucleiBinaryPath, "-version").Run(); err != nil {
logger.Error(err, "Failed to execute nuclei -version, nuclei may not be properly installed", "path", s.nucleiBinaryPath)
// Don't fail here, just log - the actual scan will fail if nuclei is truly broken
} else {
logger.Info("Nuclei binary verified", "path", s.nucleiBinaryPath)
}
// Check templates availability if templates path is set
templatesAvailable := false
if s.templatesPath != "" {
if info, err := os.Stat(s.templatesPath); err != nil || !info.IsDir() {
logger.Info("Templates path does not exist or is not a directory, nuclei will use default templates",
"templatesPath", s.templatesPath,
"error", err)
} else {
// Count template files
entries, err := os.ReadDir(s.templatesPath)
if err == nil {
templateCount := 0
for _, entry := range entries {
if !entry.IsDir() && (strings.HasSuffix(entry.Name(), ".yaml") || strings.HasSuffix(entry.Name(), ".yml")) {
templateCount++
}
}
templatesAvailable = templateCount > 0
logger.Info("Templates directory found", "templatesPath", s.templatesPath, "templateCount", templateCount)
if templateCount == 0 {
logger.Info("Templates directory is empty, nuclei will download templates on first run or use default location")
}
}
}
} else {
logger.Info("No templates path configured, nuclei will use default template location (~/.nuclei/templates)")
}
// If no specific templates are provided and templates path is empty, warn
if len(options.Templates) == 0 && !templatesAvailable && s.templatesPath != "" {
logger.Info("Warning: No templates specified and templates directory appears empty. Nuclei may not run any scans.")
}
// Build the nuclei command arguments // Build the nuclei command arguments
args := s.buildArgs(targetsFile, options) args := s.buildArgs(targetsFile, options)
@@ -123,6 +183,16 @@ func (s *NucleiScanner) Scan(ctx context.Context, targets []string, options Scan
timeout = 30 * time.Minute timeout = 30 * time.Minute
} }
// Log the command being executed
fullCommand := fmt.Sprintf("%s %s", s.nucleiBinaryPath, strings.Join(args, " "))
logger.Info("Executing nuclei scan",
"command", fullCommand,
"timeout", timeout,
"templates", len(options.Templates),
"templatesList", options.Templates,
"severity", options.Severity,
"templatesPath", s.templatesPath)
// Create context with timeout // Create context with timeout
scanCtx, cancel := context.WithTimeout(ctx, timeout) scanCtx, cancel := context.WithTimeout(ctx, timeout)
defer cancel() defer cancel()
@@ -134,14 +204,31 @@ func (s *NucleiScanner) Scan(ctx context.Context, targets []string, options Scan
cmd.Stdout = &stdout cmd.Stdout = &stdout
cmd.Stderr = &stderr cmd.Stderr = &stderr
logger.Info("Starting nuclei execution")
err = cmd.Run() err = cmd.Run()
duration := time.Since(startTime) duration := time.Since(startTime)
// Log stderr output (nuclei often outputs warnings/info to stderr)
stderrStr := stderr.String()
if stderrStr != "" {
logger.Info("Nuclei stderr output", "stderr", stderrStr)
}
// Log stdout size for debugging
stdoutSize := len(stdout.Bytes())
logger.Info("Nuclei execution completed",
"duration", duration,
"exitCode", cmd.ProcessState.ExitCode(),
"stdoutSize", stdoutSize,
"stderrSize", len(stderrStr))
// Check for context cancellation // Check for context cancellation
if scanCtx.Err() == context.DeadlineExceeded { if scanCtx.Err() == context.DeadlineExceeded {
logger.Error(nil, "Scan timed out", "timeout", timeout, "stderr", stderrStr)
return nil, fmt.Errorf("scan timed out after %v", timeout) return nil, fmt.Errorf("scan timed out after %v", timeout)
} }
if scanCtx.Err() == context.Canceled { if scanCtx.Err() == context.Canceled {
logger.Error(nil, "Scan was cancelled", "stderr", stderrStr)
return nil, fmt.Errorf("scan was cancelled") return nil, fmt.Errorf("scan was cancelled")
} }
@@ -151,22 +238,40 @@ func (s *NucleiScanner) Scan(ctx context.Context, targets []string, options Scan
if exitErr, ok := err.(*exec.ExitError); ok { if exitErr, ok := err.(*exec.ExitError); ok {
// Exit code 1 can mean "no results found" which is not an error // Exit code 1 can mean "no results found" which is not an error
if exitErr.ExitCode() != 1 { if exitErr.ExitCode() != 1 {
return nil, fmt.Errorf("nuclei execution failed: %w, stderr: %s", err, stderr.String()) logger.Error(err, "Nuclei execution failed",
"exitCode", exitErr.ExitCode(),
"stderr", stderrStr,
"stdout", stdout.String())
return nil, fmt.Errorf("nuclei execution failed: %w, stderr: %s", err, stderrStr)
} }
logger.Info("Nuclei exited with code 1 (no results found)", "stderr", stderrStr)
} else { } else {
logger.Error(err, "Failed to execute nuclei", "stderr", stderrStr)
return nil, fmt.Errorf("failed to execute nuclei: %w", err) return nil, fmt.Errorf("failed to execute nuclei: %w", err)
} }
} }
// Parse the JSONL output // Parse the JSONL output
findings, err := ParseJSONLOutput(stdout.Bytes()) stdoutBytes := stdout.Bytes()
logger.Info("Parsing nuclei output", "outputSize", len(stdoutBytes))
findings, err := ParseJSONLOutput(stdoutBytes)
if err != nil { if err != nil {
logger.Error(err, "Failed to parse nuclei output",
"stdout", string(stdoutBytes),
"stderr", stderrStr)
return nil, fmt.Errorf("failed to parse nuclei output: %w", err) return nil, fmt.Errorf("failed to parse nuclei output: %w", err)
} }
logger.Info("Parsed findings", "count", len(findings))
// Calculate summary // Calculate summary
summary := calculateSummary(findings, len(targets), duration) summary := calculateSummary(findings, len(targets), duration)
logger.Info("Scan completed",
"findings", len(findings),
"duration", duration,
"targetsScanned", len(targets))
return &ScanResult{ return &ScanResult{
Findings: findings, Findings: findings,
Summary: summary, Summary: summary,
@@ -181,11 +286,8 @@ func (s *NucleiScanner) buildArgs(targetsFile string, options ScanOptions) []str
"-jsonl", "-jsonl",
"-silent", "-silent",
"-no-color", "-no-color",
} "-rate-limit", "150", // Limit rate to avoid overwhelming targets
"-bulk-size", "25", // Process targets in bulk
// Add templates path if configured
if s.templatesPath != "" {
args = append(args, "-t", s.templatesPath)
} }
// Add specific templates if provided // Add specific templates if provided
@@ -193,6 +295,31 @@ func (s *NucleiScanner) buildArgs(targetsFile string, options ScanOptions) []str
for _, t := range options.Templates { for _, t := range options.Templates {
args = append(args, "-t", t) args = append(args, "-t", t)
} }
} else {
// When no templates are specified, nuclei should use all available templates
// Only add templates path if it's configured AND contains templates
// Otherwise, let nuclei use its default template location (~/.nuclei/templates)
if s.templatesPath != "" {
// Check if templates directory exists and has content
if info, err := os.Stat(s.templatesPath); err == nil && info.IsDir() {
entries, err := os.ReadDir(s.templatesPath)
if err == nil {
hasTemplates := false
for _, entry := range entries {
if !entry.IsDir() && (strings.HasSuffix(entry.Name(), ".yaml") || strings.HasSuffix(entry.Name(), ".yml")) {
hasTemplates = true
break
}
}
if hasTemplates {
args = append(args, "-t", s.templatesPath)
}
// If no templates found, don't add -t flag, let nuclei use default location
}
}
}
// If no templates path or it's empty, nuclei will use default location
// which it will download templates to on first run if needed
} }
// Add severity filter if provided // Add severity filter if provided