ChenSammi commented on code in PR #11053: URL: https://github.com/apache/ozone/pull/11053#discussion_r4266918042
########## hadoop-hdds/docs/content/design/s3-object-lock.md: ########## @@ -0,0 +1,612 @@ +--- +title: S3 Object Lock +summary: Design to support S3 object lock. +date: 2026-08-18 +jira: HDDS-15945 +status: draft +author: Chung En Lee +--- +<!-- + Licensed under the Apache License, Version 2.0 (the "License"); + you may not use this file except in compliance with the License. + You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. See accompanying LICENSE file. +--> + +# S3 Object Lock Design Doc + +## Summary + +This design document aims to plan and implement the Object Lock mechanism for OBS buckets integrated with Ranger. +The primary objective is to provide data immutability and tamper-proof protection through the object locking feature. + +## Problem statement + +With growing demands for data security and compliance, ensuring that critical data stored in OBS (Object Storage) is +protected from accidental or malicious deletion and overwriting has become an essential system protection requirement. +To establish a more rigorous data protection mechanism, we plan to introduce the Object Lock feature. + +Considering the current system architecture and access control strategies, +this design integrates with existing Apache Ranger to manage Object Lock permissions on OBS buckets. +Meanwhile, to accelerate core feature delivery, we have decided to exclude complex multi-version locking (Versioning Lock), +Legacy buckets, and FSO buckets from this initial release. In addition, support for Native ACLs is excluded; Native ACLs typically grant permissions +at the granular bucket or object level, whereas Object Lock permission management favors broad, role-based authorization, +creating a conflict in design philosophies. Narrowing the scope allows us to focus on the core functionality and ensure a rapid, +stable rollout of baseline tamper-proof protection. + +**Goal:** + +* Implement the Object Lock feature on standard OBS buckets, fully integrated with Ranger for permission and access control. +* Support single-version objects only. + +## Non-Goal + +* Versioning Lock: Support for locking across multiple object versions is deferred (multi-version core features are currently under development). +* Legacy and FSO Buckets: Object Lock support for legacy and FSO buckets is excluded. The following considerations are deferred to a future phase: + * Lock Granularity: FSO separates files and directories across FileTable and DirectoryTable, requiring explicit policy decisions on whether lock semantics apply at the file level, directory level, or both. + * Rename & Move Semantics: Renaming or moving paths alters namespace hierarchy without mutating object content; rules must be defined to either block operations on locked paths or preserve lock metadata across the same objectID +* Native ACL Support: Native ACLs will not be used for access control or advanced configuration such as Retention Mode (Governance); access management is centralized exclusively via Ranger. + +## Differences from AWS + +* Single-Version Model: Locks apply to the current key rather than individual object versions. +* Direct Overwrite Blocking: Unlike AWS S3 (which writes a new version), Ozone strictly rejects Put, Copy, and MPU requests on locked keys with 403 Access Denied (WORMProtectionException). +* Ranger-Dependent Auth: Lock governance relies exclusively on Apache Ranger and STS action matching (enableActionMatcherInPoliciesCondition). Object Lock is not supported on clusters without Ranger: OM rejects all Object Lock configuration requests unless the Ranger authorizer is in use. + +## Technical Description + +### Terminology + +**Legal Hold** + +* **Definition**: Applies an indefinite lock status to an object. The object remains protected until an administrator explicitly removes the lock (Remove Legal Hold). +* **Restricted Operations**: + * Put Object + * Delete Object + * Multipart Initial / Complete +* **Allowed Operations**: + * Get Object + * Get Legal Hold + * Put Legal Hold (Depends on permission) + +**Retention** + +* Definition: Configures a retention policy for an object to prevent deletion or modification. Retention can be duration-based (configured in days or years, establishing a fixed `RetainUntilDate`) or event-driven (**Event Hold / Event-based Retention**, where an object remains protected indefinitely until an external business or legal event triggers the final retention countdown). +* Retention Modes: + * Compliance Mode: The strictest protection tier. Once applied, no user (including root/admin) can remove the lock, shorten the duration, or overwrite the object before the retention period expires. + * Governance Mode: A flexible protection tier. Standard users are restricted by locking rules, but users with `BypassGovernanceRetention` permissions can bypass restrictions to modify or delete the object. +* Event Hold / Event-based Retention: + * Used for records management where the retention lifecycle depends on external events (e.g., contract termination, employee departure, loan closure). + * Keeps the object immutable while awaiting event notification; once the event occurs, the definitive expiration date (`RetainUntilDate`) is calculated and applied. + * While the event has not occurred, `retentionDate` is unset and `eventHold.enabled = true` keeps the key locked indefinitely. `eventHold.rule` is copied from the bucket default when the key is created, so later changes to the bucket default do not affect existing keys. + * While an event hold is active, a `GOVERNANCE` key can be deleted or overwritten with `bypassGovernanceRetention`; a `COMPLIANCE` key cannot. +* Restricted Operations: + * Put Object / Copy Object (Overwrites) + * Delete Object + * Multipart Upload (Initial / Complete) + * Shorten Retention Period (in Compliance Mode) +* Allowed Operations: + * Get Object + * Extend Retention Period + +#### Event Hold Trigger Workflow +For enterprise governance workflows, the retention lifecycle often depends on specific business triggers rather than a static configuration at creation. The Event Hold mechanism handles this dynamic lifecycle through the following state transitions: + +```text +object created + | +event hold applied (indefinite WORM protection) + | +business event occurs + | +hold released + | +retention timer starts (RetainUntilDate is calculated) + | +WORM until timer expires +``` + +The "business event occurs" step is triggered with `PutObjectRetention`, so no new API or Ranger action is needed +and S3 clients can release the hold through S3 Gateway (native clients use `OzoneBucket#setRetention`). +When OM handles `PutObjectRetention` on a key with an active event hold, it clears `eventHold` under the key write lock +and sets `retentionDate` to the later of the requested `RetainUntilDate` and the leader-stamped request time plus the event hold `rule`. +The retention mode cannot be changed by this request, so the usual rules for shortening retention still apply. + +> _**Note**: +> * Background & Root Cause: A prerequisite for enabling WORM (Write Once, Read Many) in AWS S3 is that Object Versioning must be enabled. Under S3 architecture, executing a Put on a locked object generates a new version without affecting the protected prior version; thus, S3 Object Lock primarily restricts Delete Object. +> * Ozone Implementation Status: Because Ozone's versioning feature is still under development, to guarantee absolute immutability during the lock period, Ozone will directly block and reject all overwrite operations (such as any form of Put or overwrite) on locked objects. + +### Table Changes + +**Bucket Table** + +Two new fields: objectLockEnabled & defaultRetention. + +```protobuf + message BucketInfo { + // ... existing fields + optional bool objectLockEnabled = 24 [default = false]; + optional RetentionConfig defaultRetention = 25; + } + + message RetentionConfig { + required RetentionMode retentionMode = 1; + optional Rule rule = 2; + optional EventHold eventHold = 3; + } + + message EventHold { + required bool enabled = 1; + required Rule rule = 2; + } + + message Rule { + required TimeUnit timeUnit = 1; + required uint64 duration = 2; + } + + enum TimeUnit { + DAYS = 1; + YEARS = 2; + } + + enum RetentionMode { + GOVERNANCE = 1; + COMPLIANCE = 2; + } +``` + + + +**Key Table** + +Two new fields: retention & legalHold. + +```protobuf +message KeyInfo { + // ... existing fields + optional Retention retention = 23; + optional bool legalHold = 24 [default = false]; +} + +message Retention { + required RetentionMode retentionMode = 1; + optional string retentionDate = 2; + optional EventHold eventHold = 3; +} +``` + + + +### Ranger Access Control + +Ozone Object Lock enforces a dual-gate protection mechanism combining **Ranger policy-based authorization** and **underlying OM WORM state validation**: + +* **Standard Data Operations (Put / Delete)**: Even if a principal has valid Ranger `CREATE`, `WRITE` or `DELETE` access, any attempt to mutate, overwrite, or delete an object under an active Legal Hold or an unexpired Retention period is immediately rejected with `403 Access Denied` (`WORMProtectionException`). +* **Fine-Grained S3 Action Matching**: Object Lock management integrates with the action-matching authorization framework introduced via STS. S3 Object Lock actions map 1:1 to AWS S3 action names. See [AWS STS Design for Ozone S3](ozone-sts.md) for details. + +--- + +#### 1. Configuration & Component Updates + +To enable action-level authorization and onboard Object Lock actions into the Ozone & Ranger ecosystem, the following updates are required: + +**Prerequisite: Feature Flag Activation**: +* Enable the Ranger action matcher condition in the Ozone configuration: + ```properties + ranger.servicedef.ozone.enableActionMatcherInPoliciesCondition=true + ``` +* When enabled, Ozone S3 Gateway (S3G) intercepts the high-level S3 API operation and populates the S3 action name into the `RangerAccessRequest`, allowing Ranger's `RangerActionMatcher` to evaluate policy conditions against incoming requests. + +1. **UI & Metadata Registration**: + * Update `security-admin/src/main/webapp/react-webapp/src/utils/actionRequirements/ozone.json` in Apache Ranger: + Register the 7 new Object Lock actions along with their prerequisite primitive permissions (e.g., `READ`, `WRITE`) mapped by resource level (Bucket vs. Key). This enforces design-time validation in the Ranger React UI, preventing administrators from creating invalid policy items. + +2. **S3 Gateway & STS Mapping**: + * Update **`S3GActionIamMapper`**: Map incoming HTTP request context and sub-resources (e.g., `?legal-hold`, `?retention`, `?object-lock`) to the corresponding action string. + * Update **`IamSessionPolicyResolver`**: Ensure session-scoped policies (STS AssumeRole / federation tokens) recognize and evaluate the new Object Lock action strings. + +--- + +#### 2. Action to Ranger Access Type Mapping + +Following the declarative contract in `ozone.json`, each S3 action requires a baseline primitive Ranger permission at the designated resource hierarchy: + +| S3 Action | Resource Level | Required Ranger Access Type | Description | +|:---------------------------------------|:---------------|:----------------------------|:--------------------------------------------------------------------| +| **`GetBucketObjectLockConfiguration`** | Bucket | `READ` | Retrieve default Object Lock settings on a bucket. | +| **`PutBucketObjectLockConfiguration`** | Bucket | `WRITE` | Configure default retention mode and period on a bucket. | +| **`GetObjectRetention`** | Key | `READ` | Read the retention mode and Retain-Until date of an object version. | +| **`PutObjectRetention`** | Key | `WRITE` | Set or extend object retention mode and duration. | +| **`GetObjectLegalHold`** | Key | `READ` | Query the current Legal Hold status (`ON` or `OFF`). | +| **`PutObjectLegalHold`** | Key | `WRITE` | Toggle the Legal Hold state (`ON` or `OFF`). | +| **`BypassGovernanceRetention`** | Key | `WRITE` and `DELETE` | Privileged entitlement to bypass retention in Governance Mode. | + +--- + +#### 3. Value-Level Authorization & Ranger Condition Evaluator + +While API invocation is controlled by the S3 Action (e.g., granting `PutObjectLegalHold`), real-world governance often requires **role segregation based on request values** (e.g., Compliance Officers can toggle `ON`, but only external Auditors can toggle `OFF`; or preventing operators from setting `COMPLIANCE` mode). + +To support this without fragmenting AWS-compatible Action definitions, Ozone and Ranger introduce a payload-aware **String Matcher Condition Evaluator**: + +1. **Context Extraction in Ozone S3G**: + * During request interception, S3G inspects the request body/headers and injects target attributes into the `RangerAccessRequest.getContext()`: + * `s3:object-lock-legal-hold`: Extracted from the XML body (`ON` or `OFF`). + * `s3:object-lock-mode`: Extracted from the XML body (`GOVERNANCE` or `COMPLIANCE`). + +2. **Ranger Custom Condition Evaluator**: + * Implement a dedicated condition evaluator extending `RangerAbstractConditionEvaluator` (registered in `ranger-servicedef-ozone.json`): + * Evaluates `s3:object-lock-legal-hold` against configured policy criteria (`ON` vs. `OFF`). + * Evaluates `s3:object-lock-mode` against configured policy criteria (`GOVERNANCE` vs. `COMPLIANCE`). + * Enables fine-grained policies such as: + * *Policy A (Evidence Preservation)*: Allow `PutObjectLegalHold` **ONLY IF** `s3:object-lock-legal-hold == "ON"`. + * *Policy B (Safe Retention)*: Allow `PutObjectRetention` **ONLY IF** `s3:object-lock-mode == "GOVERNANCE"`. + +--- + +#### 4. Retention Modes & Governance Bypass Enforcement + +* **Compliance Mode**: + * The strictest WORM protection tier. + * Once applied, the lock is immutable. **No role, administrator, or Ranger permission** (including `BypassGovernanceRetention`) can overwrite, shorten, or delete the object version until the retention period expires. +* **Governance Mode & `BypassGovernanceRetention`**: + * Protects objects from accidental deletion by general users while retaining operational flexibility for authorized personnel. + * **Dual-Check Enforcement**: + 1. **Client Declaration**: The client must explicitly set the `bypassGovernanceRetention` flag on the request. S3 Gateway sets it from the `x-amz-bypass-governance-retention: true` header; native Ozone clients pass it as a boolean argument. + It is carried in `KeyArgs` and `DeleteKeyArgs`, so it is available to every operation that checks Object Lock (delete, batch delete, overwrite, multipart upload, `PutObjectRetention`). + Because Ozone does not support versioning yet, this flag is an Ozone-specific design: when Object Lock is enabled without versioning, a Governance-locked key can only be deleted or overwritten with the flag. + 2. **Ranger Action Entitlement**: Ozone checks whether the principal has the **`BypassGovernanceRetention`** action granted in Ranger for the target key resource. + 3. If either check fails, the request is aborted with `403 Access Denied`. + + +> _**Warning**: Permanent Data Immutability in Compliance Mode_ +> +> Unlike AWS S3, where objects in Compliance mode cannot be deleted by any user (including root) and will only be destroyed if the entire AWS account itself is removed, Apache Ozone has no concept of a root or tenant account. +> Once applied, a Compliance lock cannot be overridden or deleted by any admin or Ranger policy until the retention period expires. + +### New Ozone APIs + +Surfacing objectLockEnabled, default retention settings, and active WORM protection via ozone shell will be addressed in a follow-up phase. + +**OzoneBucket** + +```java +public void setRetentionConfig(RetentionArgs retentionArgs); +public void setRetention(String keyName, ObjectRetention retention, boolean bypassGovernanceRetention); +public void setLegalHold(String keyName, boolean hold); +``` + +### Supported S3 APIs + +To ensure seamless integration with existing S3 clients (e.g., AWS CLI, Boto3) and applications, the Ozone S3 Gateway (S3G) will translate and support standard AWS S3 APIs. +In alignment with AWS S3, HeadObject and GetObject responses will be enriched with lock-related headers (`x-amz-object-lock-mode`, `x-amz-object-lock-retain-until-date`, and `x-amz-object-lock-legal-hold`) when the caller possesses the required GetObjectRetention or GetObjectLegalHold permissions. Without the permission the headers are omitted and the request still succeeds, so the extra authorization never fails a plain read. + +**Bucket-Level APIs:** +* `CreateBucket`: Supports enabling Object Lock at creation time via the x-amz-bucket-object-lock-enabled: true header. +* `GetBucketObjectLockConfiguration`: Retrieves the Object Lock status and the default retention configuration (if any) for a specified OBS bucket. +* `PutBucketObjectLockConfiguration`: Enables Object Lock and default retention on an existing bucket. In alignment with AWS S3, passing the `x-amz-bucket-object-lock-token` header is supported when enabling lock on existing buckets. + +**Object-Level APIs:** +* `PutObject / CreateMultipartUpload`: Supports setting retention and legal hold directly during object creation via `x-amz-object-lock-mode`, `x-amz-object-lock-retain-until-date`, and `x-amz-object-lock-legal-hold` request headers. +* `PutObjectRetention`: Places a retention configuration on an object, specifying the retention mode (COMPLIANCE or GOVERNANCE) and the retention duration (in days or years). +* `GetObjectRetention`: Retrieves the current retention configuration applied to an object. +* `PutObjectLegalHold`: Applies or removes a Legal Hold configuration to the specified object. +* `GetObjectLegalHold`: Retrieves the current Legal Hold status of an object. + +**Impacted Standard APIs:** +Standard data mutating APIs will intercept the Object Lock state and return the `403 Access Denied` (or specifically `WORMProtectionException`) if a modification is attempted on a locked object: +* `PutObject` / `CopyObject` (Overwrites will be rejected) +* `DeleteObject` / `DeleteObjects` (Deletions will be rejected) +* `CreateMultipartUpload` / `CompleteMultipartUpload` / `UploadPart` (Uploads will be rejected if the object is locked) +* `CopyObject` does not copy the Object Lock configuration of the source object (same as AWS S3 by default); the destination gets the bucket default retention or the `x-amz-object-lock-*` headers of the request. +* The rejection uses a new OM result code for `WORMProtectionException`, which S3 Gateway maps to `403 AccessDenied`; `DeleteObjects` reports it as a per-key `AccessDenied` error. +* Lifecycle expiration: the background lifecycle service deletes keys through the same `DeleteKeys` path, so locked keys are skipped by the delete check. + +> _**Note**: +> * **Object Lock Immutability & Future Versioning Binding**: +> - **Current State (No Versioning)**: Once Object Lock is enabled on a bucket (`objectLockEnabled = true`), it cannot be disabled. +> - **AWS S3 Alignment & Future Evolution**: In standard AWS S3, Object Lock inherently requires and automatically enables Versioning upon bucket creation. While Ozone temporarily decouples them because Ozone does not support Versioning yet, future implementations will align with S3 semantics: enabling Object Lock will automatically enforce Versioning, and Versioning can never be suspended or disabled thereafter. + +### Impact on Write and Delete Flows + +**Impact on Put Object Flow** + + + +During the Create Key phase, the system executes a preliminary WORM check in `preExecute` to enforce a fail-fast mechanism: + +1. Linearizability does not need to be guaranteed at this stage. +2. It avoids Raft consensus overhead, conserving system resources. + +During the Commit Key phase, the system performs final WORM validation in `validateAndUpdateCache`: + +1. It guarantees linearizability, ensuring only one write request can succeed simultaneously on a WORM-protected object. +2. Keys that fail to commit remain in an Open Key state, and their associated metadata and content will be reclaimed periodically by system background cleanup. + +**Impact on Multipart Upload Flow** + +Multipart Upload adopts the same design logic as Put Object to ensure operational consistency: +1. Similar to the creation key flow, WORM checks during the Initiate Multipart Upload and Put Part phases are executed in `preExecute` for fail-fast behavior (linearizability is not yet required, minimizing consensus overhead). +2. Similar to the commit key flow, WORM check during the Complete Multipart Upload phase is executed in `validateAndUpdateCache` to guarantee linearizability for writes under locked states. + +**Impact on Delete Object Flow** + +All delete operations must perform WORM validation during the `validateAndUpdateCache` phase to preserve linearizability and prevent accidental deletion of protected data. + +### Volume Force Delete + +A volume cannot be deleted while it contains locked keys (`VOLUME_NOT_EMPTY`), and `ozone sh volume delete -r` deletes keys Review Comment: It's not a good idea to support a `ozone sh volume delete -r` command, to minic the AWS account close situation. Whether and how Ozone should support an account close situation, is out of the scope of this feature. For this feature, the real problem is if Compliance Mode is put on a object, is there anyone can remove it? If we want to follow the AWS rule, then instead of provide the volume deletion command, provide a offline Compliance Mode update command per key or batch is more appropriate. @chungen0126 , there is this "ozone repair" command category, where commands are provided to fix the problems offline. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
