mirror of
https://github.com/Studio-Saelix/sencho.git
synced 2026-07-26 11:49:16 +00:00
6527bc971b
Scan-policy deploy gates can now block on a known-exploited CVE (CISA KEV) and on a fixable Critical/High finding, in addition to an optional severity threshold. New policies default risk-first (KEV and fixable on, severity off); existing policies keep their severity-only behavior. CVSS stays captured for context but is never the sole basis for a block, and a finding whose exploitability cannot be confirmed is treated as risky rather than safe (incomplete scan detail fails closed on KEV/fixable inputs). The decision logic is shared between the pre-deploy gate and the informational post-scan banner via a pure helper, so the two never disagree. Block messages and the block dialog now name the conditions an image matched. Backend and frontend gates move together, the new inputs replicate across the fleet, and a blocking policy with no active input is rejected on both sides.
293 lines
11 KiB
Plaintext
293 lines
11 KiB
Plaintext
---
|
|
title: Security
|
|
description: Automate scan policies, CVE suppressions, and vulnerability scans from CI. Includes the deploy 409 response shape and admin bypass flow.
|
|
---
|
|
|
|
The Security API lets you manage scan policies, CVE suppressions, and trigger vulnerability scans from CI pipelines and automation scripts. Every endpoint in this reference is intended for external automation; internal frontend-only endpoints (finding listings, SARIF downloads) are not documented here.
|
|
|
|
All endpoints require [Bearer token authentication](/api-reference/overview#authentication). Manual scans, secret and misconfiguration results, scan comparison, and CVE suppressions are available on every tier. Scan policies (with `block_on_deploy` enforcement), SBOM, and SARIF require Admiral. See the per-endpoint **License** row for details.
|
|
|
|
## Scan policies
|
|
|
|
Scan policies define risk conditions that Sencho evaluates on every post-deploy and scheduled scan: a known-exploited CVE (CISA KEV), a fixable Critical/High finding, and an optional severity threshold. When a policy has `block_on_deploy=1`, a deploy whose images match any enabled condition is rejected at the pre-flight stage with an HTTP 409 response. See [Deploy Enforcement](/features/deploy-enforcement) for the full flow.
|
|
|
|
Writes are admin-only and rejected on replica nodes (policies are managed on the control instance and replicate fleet-wide).
|
|
|
|
### List policies
|
|
|
|
**`GET /api/security/policies`**
|
|
|
|
**License:** Admiral
|
|
|
|
```bash
|
|
curl -H "Authorization: Bearer YOUR_API_TOKEN" \
|
|
https://your-sencho-instance:1852/api/security/policies
|
|
```
|
|
|
|
**Response:**
|
|
|
|
```json
|
|
[
|
|
{
|
|
"id": 1,
|
|
"name": "prod-high-gate",
|
|
"node_id": null,
|
|
"node_identity": null,
|
|
"stack_pattern": "prod-*",
|
|
"max_severity": "HIGH",
|
|
"block_on_deploy": 1,
|
|
"enabled": 1,
|
|
"block_on_severity": 1,
|
|
"block_on_kev": 0,
|
|
"block_on_fixable": 0,
|
|
"replicated_from_control": 0,
|
|
"created_at": 1745107200000,
|
|
"updated_at": 1745107200000
|
|
}
|
|
]
|
|
```
|
|
|
|
### Create policy
|
|
|
|
**`POST /api/security/policies`**
|
|
|
|
**License:** Admiral · **Role:** Admin
|
|
|
|
| Field | Type | Required | Description |
|
|
|-------|------|:--------:|-------------|
|
|
| `name` | string | yes | Human-readable name shown in the UI. |
|
|
| `stack_pattern` | string or `null` | no | Glob against stack names (e.g. `prod-*`). `null` matches every stack. |
|
|
| `max_severity` | string | yes | One of `CRITICAL`, `HIGH`, `MEDIUM`, `LOW`. The threshold used when `block_on_severity` is on. Always stored for context. |
|
|
| `block_on_deploy` | `0` or `1` | yes | When `1`, pre-flight matches reject deploys with HTTP 409. When `0`, post-deploy and scheduled scans dispatch warning alerts instead. |
|
|
| `enabled` | `0` or `1` | no | Defaults to `1`. Disabled policies are never evaluated. |
|
|
| `block_on_severity` | `0` or `1` | no | Block when an image's highest non-suppressed finding meets or exceeds `max_severity`. Defaults to `0`. |
|
|
| `block_on_kev` | `0` or `1` | no | Block when an image carries a CVE on the CISA known-exploited (KEV) list. Defaults to `1`. |
|
|
| `block_on_fixable` | `0` or `1` | no | Block when an image has a Critical/High finding with a fix available. Defaults to `1`. |
|
|
| `node_id` | number or `null` | no | Scope the policy to one node. `null` applies the policy fleet-wide. |
|
|
|
|
A blocking policy (`block_on_deploy=1`) must enable at least one of `block_on_severity`, `block_on_kev`, or `block_on_fixable`. When the three input fields are omitted, they default risk-first (`block_on_kev=1`, `block_on_fixable=1`, `block_on_severity=0`).
|
|
|
|
```bash
|
|
curl -X POST https://your-sencho-instance:1852/api/security/policies \
|
|
-H "Authorization: Bearer YOUR_API_TOKEN" \
|
|
-H "Content-Type: application/json" \
|
|
-d '{
|
|
"name": "prod-high-gate",
|
|
"stack_pattern": "prod-*",
|
|
"max_severity": "HIGH",
|
|
"block_on_deploy": 1,
|
|
"enabled": 1
|
|
}'
|
|
```
|
|
|
|
**Response:** `201 Created` with the full policy object.
|
|
|
|
**Errors:**
|
|
|
|
- `400` `max_severity must be CRITICAL, HIGH, MEDIUM, or LOW`
|
|
- `400` `Policy name is required`
|
|
- `400` `A blocking policy must enable at least one of: severity threshold, KEV, or fixable.`
|
|
- `403` when called on a replica node
|
|
|
|
### Update policy
|
|
|
|
**`PUT /api/security/policies/{id}`**
|
|
|
|
**License:** Admiral · **Role:** Admin
|
|
|
|
Any of the create fields can be updated individually. Omitted fields are left unchanged.
|
|
|
|
```bash
|
|
curl -X PUT https://your-sencho-instance:1852/api/security/policies/1 \
|
|
-H "Authorization: Bearer YOUR_API_TOKEN" \
|
|
-H "Content-Type: application/json" \
|
|
-d '{ "block_on_deploy": 0 }'
|
|
```
|
|
|
|
**Response:** `200 OK` with the updated policy.
|
|
|
|
**Errors:** `400 Invalid policy id`, `404 Policy not found`, `403` on replicas.
|
|
|
|
### Delete policy
|
|
|
|
**`DELETE /api/security/policies/{id}`**
|
|
|
|
**License:** Admiral · **Role:** Admin
|
|
|
|
```bash
|
|
curl -X DELETE https://your-sencho-instance:1852/api/security/policies/1 \
|
|
-H "Authorization: Bearer YOUR_API_TOKEN"
|
|
```
|
|
|
|
**Response:** `200 OK` `{"success": true}`. The call is idempotent; deleting a non-existent id returns the same shape.
|
|
|
|
## CVE suppressions
|
|
|
|
Suppressions let you mark individual CVEs as acknowledged so scan reads, comparisons, and exports downgrade them to a dimmed state. See [CVE Suppressions](/features/cve-suppressions) for the full flow.
|
|
|
|
### List suppressions
|
|
|
|
**`GET /api/security/suppressions`**
|
|
|
|
**License:** Community
|
|
|
|
Response rows include an `active` boolean computed from the `expires_at` timestamp.
|
|
|
|
```json
|
|
[
|
|
{
|
|
"id": 3,
|
|
"cve_id": "CVE-2024-12345",
|
|
"pkg_name": "openssl",
|
|
"image_pattern": "alpine:*",
|
|
"reason": "Awaiting upstream patch. Reviewed by security on 2026-04-10.",
|
|
"created_by": "ops@example.com",
|
|
"created_at": 1744000000000,
|
|
"expires_at": 1746592000000,
|
|
"replicated_from_control": 0,
|
|
"active": true
|
|
}
|
|
]
|
|
```
|
|
|
|
### Create suppression
|
|
|
|
**`POST /api/security/suppressions`**
|
|
|
|
**License:** Community · **Role:** Admin
|
|
|
|
| Field | Type | Required | Description |
|
|
|-------|------|:--------:|-------------|
|
|
| `cve_id` | string | yes | `CVE-YYYY-NNNN` or `GHSA-xxxx-xxxx-xxxx` format. |
|
|
| `pkg_name` | string or `null` | no | Package name to narrow the scope (e.g. `openssl`). `null` suppresses every package match. |
|
|
| `image_pattern` | string or `null` | no | Glob against image references. `null` applies fleet-wide. |
|
|
| `reason` | string | yes | Required justification, up to 2000 characters. Surfaced on every dimmed row. |
|
|
| `expires_at` | number or `null` | no | Unix timestamp in milliseconds. `null` for an indefinite suppression. |
|
|
|
|
```bash
|
|
curl -X POST https://your-sencho-instance:1852/api/security/suppressions \
|
|
-H "Authorization: Bearer YOUR_API_TOKEN" \
|
|
-H "Content-Type: application/json" \
|
|
-d '{
|
|
"cve_id": "CVE-2024-12345",
|
|
"pkg_name": "openssl",
|
|
"image_pattern": "alpine:*",
|
|
"reason": "Awaiting upstream patch.",
|
|
"expires_at": 1746592000000
|
|
}'
|
|
```
|
|
|
|
**Response:** `201 Created` with the full suppression.
|
|
|
|
**Errors:** `400` on invalid shape (missing `reason`, malformed `cve_id`, over-length fields), `409` when a suppression already exists for the same `(cve_id, pkg_name, image_pattern)` triple, `403` on replicas.
|
|
|
|
### Delete suppression
|
|
|
|
**`DELETE /api/security/suppressions/{id}`**
|
|
|
|
**License:** Community · **Role:** Admin
|
|
|
|
```bash
|
|
curl -X DELETE https://your-sencho-instance:1852/api/security/suppressions/3 \
|
|
-H "Authorization: Bearer YOUR_API_TOKEN"
|
|
```
|
|
|
|
## Vulnerability scans
|
|
|
|
### Trigger a scan
|
|
|
|
**`POST /api/security/scan`**
|
|
|
|
**License:** Community · **Role:** Admin
|
|
|
|
Accepts an image reference and starts an asynchronous scan. The response returns immediately with a `scanId` that you can poll.
|
|
|
|
| Field | Type | Required | Description |
|
|
|-------|------|:--------:|-------------|
|
|
| `imageRef` | string | yes | Image reference Trivy will scan (must match Sencho's validator; `/`, `:`, `@`, alphanumerics, `-`, `_`, `.`). |
|
|
| `stackName` | string | no | Associates the scan with a stack for display purposes. |
|
|
| `force` | boolean | no | Default `false`. When `true`, ignores the 24-hour digest cache and runs Trivy again. |
|
|
| `scanners` | `["vuln"]` or `["vuln","secret"]` | no | Omit for vuln-only. Pass `["vuln","secret"]` to include secret detection. |
|
|
|
|
```bash
|
|
curl -X POST https://your-sencho-instance:1852/api/security/scan \
|
|
-H "Authorization: Bearer YOUR_API_TOKEN" \
|
|
-H "Content-Type: application/json" \
|
|
-d '{ "imageRef": "nginx:1.27" }'
|
|
```
|
|
|
|
**Response:** `202 Accepted` `{ "scanId": 142 }`. The scan continues in the background.
|
|
|
|
**Errors:** `400 imageRef is required`, `400 Invalid imageRef format`, `409 Already scanning this image`, `503 Trivy is not available on this host`.
|
|
|
|
### List scans
|
|
|
|
**`GET /api/security/scans`**
|
|
|
|
**License:** Community
|
|
|
|
Supports filters: `imageRef`, `imageRefLike`, `status`, `limit`, `offset`. Response is paginated with `total` and `scans` fields.
|
|
|
|
```bash
|
|
curl -H "Authorization: Bearer YOUR_API_TOKEN" \
|
|
"https://your-sencho-instance:1852/api/security/scans?imageRefLike=nginx&limit=20"
|
|
```
|
|
|
|
## Stack deploy with policy
|
|
|
|
Every stack deploy endpoint evaluates the scan policy gate before `docker compose up` runs. When a policy blocks the deploy, the endpoint returns HTTP 409.
|
|
|
|
**Affected endpoints:** `POST /api/stacks/{stackName}/deploy`, `POST /api/stacks/{stackName}/update`, `POST /api/stacks/{stackName}/recreate`, `POST /api/stacks/from-git`, `POST /api/stacks/git-source/{sourceId}/apply`, `POST /api/stacks/deploy-template`.
|
|
|
|
### Blocked-deploy response
|
|
|
|
HTTP 409 Conflict. The response body is parseable JSON so CI pipelines can branch on it:
|
|
|
|
```json
|
|
{
|
|
"error": "Policy \"prod-high-gate\" blocked deploy: 2 image(s) exceed HIGH",
|
|
"policy": {
|
|
"id": 3,
|
|
"name": "prod-high-gate",
|
|
"maxSeverity": "HIGH"
|
|
},
|
|
"violations": [
|
|
{
|
|
"imageRef": "myapp:v1",
|
|
"severity": "CRITICAL",
|
|
"criticalCount": 2,
|
|
"highCount": 5,
|
|
"scanId": 142
|
|
},
|
|
{
|
|
"imageRef": "nginx:1.14",
|
|
"severity": "HIGH",
|
|
"criticalCount": 0,
|
|
"highCount": 3,
|
|
"scanId": 143
|
|
}
|
|
]
|
|
}
|
|
```
|
|
|
|
### Bypassing a block
|
|
|
|
Admins can bypass the gate on a per-deploy basis by passing `?ignorePolicy=true` on the deploy URL. The flag is only honored when the calling session resolves to an admin user; API tokens with read-only or deploy-only scope cannot bypass a policy.
|
|
|
|
```bash
|
|
curl -X POST "https://your-sencho-instance:1852/api/stacks/myapp/deploy?ignorePolicy=true" \
|
|
-H "Authorization: Bearer ADMIN_API_TOKEN"
|
|
```
|
|
|
|
Every bypass is recorded in the [Audit Log](/features/audit-log) with the originating path and method, the actor's username, the policy name, and the list of bypassed images.
|
|
|
|
**Response when bypass is honored:** `200 OK` with the normal deploy success payload and `bypassedPolicy: true`.
|
|
|
|
**Response when bypass is denied:** the original `409 Conflict` payload is returned (the caller was not admin, or the token scope disallowed bypass).
|
|
|
|
### Fail-open when Trivy is missing
|
|
|
|
If Trivy is not installed on the target node, the gate fails open: the deploy proceeds with an HTTP 200 response and Sencho dispatches a warning notification `Pre-deploy scan for "<stack>" skipped: Trivy not installed on this node`. This keeps teams from being locked out when the scanner binary is unavailable.
|
|
|
|
Install Trivy on the affected node to enforce the policy; see [Installing Trivy](/operations/trivy-setup).
|