Datastores

In order to reduce operational complexity, SpiceDB leverages existing, popular systems for persisting data.

AuthZed has standardized our managed services on CockroachDB, but we give self-hosted customers the option to pick the datastore that best suits their operational requirements.

• CockroachDB - Recomended for self hosted deployments with high throughput and/or multi-region requirements
• Cloud Spanner - Recommended for self-hosted Google Cloud deployments
• PostgreSQL - Recommended for self-hosted single-region deployments
• MySQL - Not recommended; only use if you cannot use PostgreSQL
• memdb - Recommended for local development and integration testing against applications

Migrations

Before a datastore can be used by SpiceDB or before running a new version of SpiceDB, you must execute all available migrations.

In order to migrate a datastore, run the following command with your desired values:

spicedb migrate head \
    --datastore-engine $DESIRED_ENGINE \
    --datastore-conn-uri $CONNECTION_STRING

For more information, refer to the Datastore Migration documentation.

CockroachDB

Usage Notes

• Recommended for multi-region deployments, with configurable region awareness
• Enables horizontal scalability by adding more SpiceDB and CockroachDB instances
• Resiliency to individual CockroachDB instance failures
• Query and data balanced across the CockroachDB
• Setup and operational complexity of running CockroachDB

Developer Notes

• Code can be found here (opens in a new tab)
• Documentation can be found here (opens in a new tab)
• Implemented using pgx (opens in a new tab) for a SQL driver and connection pooling
• Has a native changefeed
• Stores migration revisions using the same strategy as Alembic (opens in a new tab)

Configuration

Required Parameters

Optional Parameters

Understanding the New Enemy Problem with CockroachDB

CockroachDB is a Spanner-like datastore supporting global, immediate consistency, with the mantra "no stale reads." The CockroachDB implementation should be used when your SpiceDB service runs in multiple geographic regions, and Google's Cloud Spanner is unavailable (e.g. AWS, Azure, bare metal.)

In order to prevent the new-enemy problem, we need to make related transactions overlap. We do this by choosing a common database key and writing to that key with all relationships that may overlap. This tradeoff is cataloged in our blog post "The One Crucial Difference Between Spanner and CockroachDB (opens in a new tab)", and means we are trading off write throughput for consistency.

Overlap Strategies

CockroachDB datastore users that are willing to rely on more subtle guarantees to mitigate the New Enemy Problem can configure the overlap strategy with the flag --datastore-tx-overlap-strategy. The available strategies are:

Depending on your application, insecure may be acceptable, and it avoids the performance cost associated with the static and prefix options. If the New Enemy Problem is not a concern for your application, consider using the insecure strategy.

When is insecure overlap a problem?

Using insecure overlap strategy for SpiceDB with CockroachDB means that it is possible that timestamps for two subsequent writes will be out of order. When this happens, it's possible for the New Enemy Problem to occur.

Let's look at how likely this is, and what the impact might actually be for your workload.

When can timestamps be reversed?

Before we look at how this can impact an application, let's first understand when and how timestamps can be reversed in the first place.

• When two writes are made in short succession against CockroachDB
• And those two writes hit two different gateway nodes
• And the CRDB gateway node clocks have a delta D
• And the writes touch disjoint sets of relationships
• And those two writes are sent within the time delta D between the gateway nodes
• And the writes land in ranges whose followers are disjoint sets of nodes
• And other independent cockroach processes (heartbeats, etc) haven't coincidentally synced the gateway node clocks during the writes.

Then it's possible that the second write will be assigned a timestamp earlier than the first write. In the next section we'll look at whether that matters for your application, but for now let's look at what makes the above conditions more or less likely:

• Clock skew. A larger clock skew gives a bigger window in which timestamps can be reversed. But note that CRDB enforces a max offset between clocks, and getting within some fraction of that max offset will kick the node from the cluster.
• Network congestion, or anything that interferes with node heartbeating. This increases the length of time that clocks can be desynchronized befor Cockroach notices and syncs them back up.
• Cluster size. When there are many nodes, it is more likely that a write to one range will not have follower nodes that overlap with the followers of a write to another range. It also makes it more likely that the two writes will have different gateway nodes. On the other side, a 3 node cluster with replicas: 3 means that all writes will sync clocks on all nodes.
• Write rate. If the write rate is high, it's more likely that two writes will hit the conditions to have reversed timestamps. If writes only happen once every max offset period for the cluster, it's impossible for their timestamps to be reversed.

The likelihood of a timestamp reversal is dependent on the cockroach cluster and the application's usage patterns.

When does a timestamp reversal matter?

Now we know when timestamps could be reversed. But when does that matter to your application?

The TL;DR is: only when you care about the New Enemy Problem.

Let's take a look at a couple of examples of how reversed timestamps may be an issue for an application storing permissions in SpiceDB.

Neglecting ACL Update Order

Two separate WriteRelationship calls come in:

• A: Alice removes Bob from the shared folder
• B: Alice adds a new document not-for-bob.txt to the shared folder

The normal case is that the timestamp for A < the timestamp for B.

But if those two writes hit the conditions for a timestamp reversal, then B < A.

From Alice's perspective, there should be no time at which Bob can ever see not-for-bob.txt. She performed the first write, got a response, and then performed the second write.

But this isn't true when using MinimizeLatency or AtLeastAsFresh consistency. If Bob later performs a Check request for the not-for-bob.txt document, it's possible that SpiceDB will pick an evaluation timestamp such that B < T < A, so that the document is in the folder and bob is allowed to see the contents of the folder.

Note that this is only possible if A - T < quantization window: the check has to happen soon enough after the write for A that it's possible that SpiceDB picks a timestamp in between them. The default quantization window is 5s.

Application Mitigations for ACL Update Order

This could be mitigated in your application by:

• Not caring about the problem
• Not allowing the write from B within the max_offset time of the CRDB cluster (or the quantization window).
• Not allowing a Check on a resource within max_offset of its ACL modification (or the quantization window).

Mis-apply Old ACLs to New Content

Two separate API calls come in:

• A: Alice remove Bob as a viewer of document secret
• B: Alice does a FullyConsistent Check request to get a ZedToken
• C: Alice stores that ZedToken (timestamp B) with the document secret when she updates it to say Bob is a fool.

Same as before, the normal case is that the timestamp for A < the timestamp for B, but if the two writes hit the conditions for a timestamp reversal, then B < A.

Bob later tries to read the document. The application performs an AtLeastAsFresh Check for Bob to access the document secret using the stored Zedtoken (which is timestamp B.)

It's possible that SpiceDB will pick an evaluation timestamp T such that B < T < A, so that bob is allowed to read the newest contents of the document, and discover that Alice thinks he is a fool.

Same as before, this is only possible if A - T < quantization window: Bob's check has to happen soon enough after the write for A that it's possible that SpiceDB picks a timestamp in between A and B, and the default quantization window is 5s.

Application Mitigations for Misapplying Old ACLs

This could be mitigated in your application by:

• Not caring about the problem
• Waiting for max_offset (or the quantization window) before doing the fully-consistent check.

When does a timestamp reversal not matter?

There are also some cases when there is no New Enemy Problem even if there are reversed timestamps.

Non-sensitive domain

Not all authorization problems have a version of the New Enemy Problem, which relies on there being some meaningful consequence of hitting an incorrect ACL during the small window of time where it's possible.

If the worst thing that happens from out-of-order ACL updates is that some users briefly see some non-sensitive data, or that a user retains access to something that they already had access to for a few extra seconds, then even though there could still effectively be a "New Enemy Problem," it's not a meaningful problem to worry about.

Disjoint SpiceDB Graphs

The examples of the New Enemy Problem above rely on out-of-order ACLs to be part of the same permission graph. But not all ACLs are part of the same graph, for example:

definition user {}
 
definition blog {
    relation author: user
    permission edit = author
}
 
defintion video {
    relation editor: user
    permission change_tags = editor
}

A: Alice is added as an author of the Blog entry new-enemy B: Bob is removed from the editors of the spicedb.mp4 video

If these writes are given reversed timestamps, it is possible that the ACLs will be applied out-or-order and this would normally be a New Enemy Problem. But the ACLs themselves aren't shared between any permission computations, and so there is no actual consequence to reversed timestamps.

Garbage Collection Window

SpiceDB warns if the garbage collection window as configured in CockroachDB is smaller than the SpiceDB configuration.

If you need a longer time window for the Watch API or querying at exact snapshots, you can adjust the value in CockroachDB (opens in a new tab):

ALTER ZONE default CONFIGURE ZONE USING gc.ttlseconds = 90000;

Relationship Integrity

Relationship Integrity is a new experimental feature in SpiceDB that ensures that data written into the supported backing datastores (currently: only CockroachDB) is validated as having been written by SpiceDB itself.

• What does relationship integrity ensure? Relationship integrity primarily ensures that all relationships written into the backing datastore were written via a trusted instance of SpiceDB or that the caller has access to the key(s) necessary to write those relationships. It ensures that if someone gains access to the underlying datastore, they cannot simply write new relationships of their own invention.

• What does relationship integrity not ensure? Since the relationship integrity feature signs each individual relationship, it does not ensure that removal of relationships is by a trusted party. Schema is also currently unverified, so an untrusted party could change it as well. Support for schema changes will likely come in a future version.

Setting up relationship integrity

To run with relationship integrity, new flags must be given to SpiceDB:

spicedb serve ...existing flags...
--datastore-relationship-integrity-enabled
--datastore-relationship-integrity-current-key-id="somekeyid"
--datastore-relationship-integrity-current-key-filename="some.key"

Place the generated key contents (which must support an HMAC key) in some.key

Deployment Process

1. Start with a clean datastore for SpiceDB. At this time, migrating an existing SpiceDB installation is not supported.
2. Run the standard migrate command but with relationship integrity flags included.
3. Run SpiceDB with the relationship integrity flags included.

Cloud Spanner

Usage Notes

• Requires a Google Cloud Account with an active Cloud Spanner instance
• Take advantage of Google's TrueTime. The Spanner driver assumes the database is linearizable and skips the transaction overlap strategy required by CockroachDB.

Developer Notes

• Code can be found here (opens in a new tab)
• Documentation can be found here (opens in a new tab)
• Starts a background GC worker (opens in a new tab) to clean up old entries from the manually-generated changelog table

Configuration

• The Cloud Spanner docs (opens in a new tab) outline how to set up an instance
• Authentication via service accounts: The service account that runs migrations must have Cloud Spanner Database Admin; SpiceDB (non-migrations) must have Cloud Spanner Database User.

Required Parameters

Optional Parameters

PostgreSQL

Usage Notes

• Recommended for single-region deployments
• Postgres 15 or newer is required for optimal performance
• Resiliency to failures only when PostgreSQL is operating with a follower and proper failover
• Carries setup and operational complexity of running PostgreSQL
• Does not rely on any non-standard PostgreSQL extensions
• Compatible with managed PostgreSQL services (e.g. AWS RDS)
• Can be scaled out on read workloads using read replicas

Developer Notes

• Code can be found here (opens in a new tab)
• Documentation can be found here (opens in a new tab)
• Implemented using pgx (opens in a new tab) for a SQL driver and connection pooling
• Stores migration revisions using the same strategy as Alembic (opens in a new tab)
• Implements its own MVCC (opens in a new tab) model by storing its data with transaction IDs

Read Replicas

SpiceDB supports Postgres read replicas and does it while retaining consistency guarantees. Typical use cases are:

• scale read workloads/offload reads from the primary
• deploy SpiceDB in other regions with primarily read workloads

Read replicas are typically configured with asynchronous replication, which involves replication lag. That would be problematic to SpiceDB's ability to solve the new enemy problem but it addresses the challenge by checking if a revision has been replicated into the target replica. If missing, it will fall back to the primary.

All API consistency options will leverage replicas, but the ones that benefit the most are those that involve some level of staleness as it increases the odds a revision has replicated. minimize_latency, at_least_as_fresh, and at_exact_snapshot consistency modes have the highest chance of being redirected to a replica.

SpiceDB supports Postgres replicas behind a load-balancer, and/or individually listing replica hosts. When multiple URIs are provided, they will be queried using a round-robin strategy.
Please note that the maximum number of replica URIs to list is 16.

Read replicas are configured with the --datastore-read-replica-* family of flags.

SpiceDB supports PgBouncer (opens in a new tab) connection pooler and is part of the test suite.

Transaction IDs and MVCC

The Postgres implementation of SpiceDB's internal MVCC mechanism involves storing the internal transaction ID count associated with a given transaction in the rows written in that transaction. Because this counter is instance-specific, there are ways in which the data in the datastore can become desynced with that internal counter. Two concrete examples are the use of pg_dump and pg_restore to transfer data between an old instance and a new instance and setting up logical replication between a previously-existing instance and a newly-created instance.

If you encounter this, SpiceDB can behave as though there is no schema written, because the data (including the schema) is associated with a future transaction ID and therefore isn't "visible" to Spicedb. If you run into this issue, the fix is documented here (opens in a new tab)

Configuration

Required Parameters

Optional Parameters

MySQL

Usage Notes

• Recommended for single-region deployments
• Setup and operational complexity of running MySQL
• Does not rely on any non-standard MySQL extensions
• Compatible with managed MySQL services
• Can be scaled out on read workloads using read replicas

Developer Notes

• Code can be found here (opens in a new tab)
• Documentation can be found here (opens in a new tab)
• Implemented using Go-MySQL-Driver (opens in a new tab) for a SQL driver
• Query optimizations are documented here (opens in a new tab)
• Implements its own MVCC (opens in a new tab) model by storing its data with transaction IDs

Read Replicas

SpiceDB supports MySQL read replicas and does it while retaining consistency guarantees. Typical use cases are:

• scale read workloads/offload reads from the primary
• deploy SpiceDB in other regions with primarily read workloads

Read replicas are typically configured with asynchronous replication, which involves replication lag. That would be problematic to SpiceDB's ability to solve the new enemy problem but it addresses the challenge by checking if the revision has been replicated into the target replica. If missing, it will fall back to the primary. Compared to the Postgres implementation, MySQL support requires two roundtrips instead of one, which means it adds some extra latency overhead.

All API consistency options will leverage replicas, but the ones that benefit the most are those that involve some level of staleness as it increases the odds a revision has replicated. minimize_latency, at_least_as_fresh, and at_exact_snapshot consistency modes have the highest chance of being redirected to a replica.

SpiceDB does not support MySQL replicas behind a load-balancer: you may only list replica hosts individually. Failing to adhere to this would compromise consistency guarantees. When multiple host URIs are provided, they will be queried using round-robin. Please note that the maximum number of MySQL replica host URIs to list is 16.

Read replicas are configured with the --datastore-read-replica-* family of flags.

Configuration

Required Parameters

Optional Parameters

memdb

Usage Notes

• Fully ephemeral; all data is lost when the process is terminated
• Intended for usage with SpiceDB itself and testing application integrations
• Cannot be ran highly-available as multiple instances will not share the same in-memory data

Developer Notes

• Code can be found here (opens in a new tab)
• Documentation can be found here (opens in a new tab)
• Implements its own MVCC (opens in a new tab) model by storing its data with transaction IDs

Configuration

Required Parameters

Optional Parameters