...
For ECS environments: Shared entities' version history should be tracked only in the central tenant.
All changes in the system related to inventory entities (instances, items, holdings, bibs) generate Domain events.
Domain events related to update action have old and new versions of the entity
Baseline Architecture
In existing architecture, mod-inventory-storage is responsible for persisting such entities as instances, holdings, and items. mod-entities-links is responsible for authorities. Both modules produce domain events on create/update/delete actions from different sources.
...
Audit Consumers Sequence Diagram
| Drawio | ||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
| Expand | ||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ||||||||||||||||||||||||||||||||||||
Audit Consumers with Outbox Sequence Diagram
The implementation can follow a Transactional outbox pattern. The approach allows enhanced guarantee for persisting the audit event but the trade-off is that this approach will negatively affect the performance of flows related to domain events. |
Solution Summary
The process is split into two main parts
1. Persistence: There are storage options that can be implemented.
Option | Description | Pros & Cons | |
|---|---|---|---|
| 1 | RDBMS | The audit database should persist a |
...
diff of the entity. The queries are made mostly by the entity's unique |
...
identifier. Thus partitioning by UUID and subpartitioning by date ranges can be applied to audit tables | Pros:
Cons:
| ||
| 2 | Object Storage | AWS S3-like storage can be used to persist diffs because audit events can be stored as plain-text (JSON) documents | Pros:
Cons:
|
2. Version history display: This should be done on demand comparing each
...
diff of the entity to the previous
3. Changes to be tracked:
Changes in entity fields
Changes related to an update of the parent entity, e.g. holding update cascades to updates in items’ effective shelving order
Changes in metadata should be ignored
Key implementation aspects:
The Kafka default delivery semantics is “AT_LEAST_ONCE”. Ensure that domain events have their unique identifiers to be able to handle consuming messages in an idempotent manner
Add new consumers in
mod-auditto inventory domain events for instances, items, and holdings.Add new consumers in
mod-auditto authority domain events for authorities.Add new consumers in
mod-auditto source record domain events for marc-bib records.(see: Source Record Domain Eventing)mod-auditshould support the following configurations on the tenant level:Retention period in years (with default value - 0 for indefinite retention)
Feature flag to enable audit capability. In case when the audit is disabled no consumers and logs should be persisted.
Anonymizing flag that indicates whether the records in the database should be anonymized before persistence to the database (To be confirmed).
mod-auditshould have the following scheduled jobs:Daily: to remove records that exceed the retention period
Monthly: to create subpartitions for audit tables
Persist audit events in an event storage. A single table in DB per entity type with partitioning by UUID (hash) and subpartitioning by date range.
Create REST API
4. to provide information on a list of changes related to a particular entity
5. to provide detailed information on the particular change - this API should use the Object diff library to return a verbose description of the difference between current and previous snapshots of changes related to the entity
ERD
With data size implications, creating separate tables per each entity type is required. The default table structure is listed below:
Column | Type | required | unique | Description | |
|---|---|---|---|---|---|
| 1 | EventID | UUID | y | y | unique event identifier |
| 2 | EventDate | timestamp | y | n | date when the event appeared in the event log |
| 3 | Origin | varchar | y | n | Origin of the event: data-import, batch-update, user, etc. |
| 4 | Action | varchar | y | n | what action was performed |
| 5 | ActionDate | timestamp | y | n | when action was performed |
| 6 | EntityID | UUID | y | n | entity identifier |
| 7 | UserId | UUID | y | n | user who did the action, fixed UUID for anonymized user |
| 8 | Snapshot Diff | jsonb | y | n | Difference between “old” and “new” body of the entity |
WBS
| Expand | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
...
Risk | Description | Probability | Impact | Mitigation | |
|---|---|---|---|---|---|
| 1 | Long period for audit records retention | The number of records could overwhelm the capability of the Postgres database both from a computational point of view and cost | High | High | Introduce separate storage for audit-events |
| 2 | Cascade Updates will create redundant copies in the audit log | The update to holdings causes updates to all related items. Some holdings may contain ~15000 records | High | Medium | Collapse or filter out events that only change the parent entity |
| 3 | Some flows could update inventory entities without using the Domain-events mechanism | With different capabilities of the system including UI, data import, bulk edit, etc some of the flows might skip sending Domain events and/or edit entities directly | Low | Medium | List those cases and add domain events to flows that has no this capability |
| 4 | Linked data | The flow and integration with inventory are not clear for the BIBFRAME format | Low | Low | Adjust BIBFRAME flow to follow the proposed solution for other inventory entities |
...
Question | Answer | Comment | |
|---|---|---|---|
| 1 | Should failure in sending audit message block the create/update/delete operation? | Hey Kalibek Turgumbayev - what happens today when an update is made and the create/update date and time stamp is not updated? | The question is related to transactional outbox pattern implementation |
| 2 | What would be the period of retention for Audit records? | Dennis Bridges has this requirement come up for you with respect to Acq’s change tracker? | The storage options depend on this question:
|
| 3 | In what form should we show the changes to non-marc fields (e.g. staffSuppress, administrative Notes, etc.) in MARC instances? | Kalibek Turgumbayev - I am unsure I understand this question. Can you review this mockup of how to display updates made to a FOLIO instance record? | Instance with source FOLIO or MARC from inventory is a separate object than SRS record and should be tracked separately |
| 4 | If only the order of fields in a MARC record is changed, should it be logged? | Kalibek Turgumbayev - Good question - I need to ask users but unless it is a significant to implement, answer is Yes. Dennis Bridges has this requirement come up for you with respect to Acq’s change tracker? | |
| 5 | Do we have scenarios where the audit log is exported in batches for some period of dates? | It is possible that a library may want to do so but I do not think it is a requirement for Sunflower. Dennis Bridges has this requirement come up for you with respect to Acq’s change tracker? | If the solution uses Postgres with partitioning by entity key, such exports would cause significant performance issues. |
| 6 | Do we need to have remapping or other technical updates as auditable events? | Not for Sunflower | We might need a list of actions leading to an event in the audit log. |
| 7 | Can we show only the last 20 records (last 3 months)? | Yes. Older records can be fetched with “see more” button and alert that it can take time. | Display of the whole history of the entity will impact performance negatively. |
| 8 | What should be tracked with BIBFRAME? Should we track the original BIBFRAME record or related MARC records is enough. | For Sunflower only track related MARC records. |
Links
Acquisition event log - data retention period is 20 years
Javers - Java object diff library