We’re excited to announce the discharge of Zebra 3.0.0—probably the most feature-rich Zebra launch ever! Along with supporting NU6.1, the most recent community improve, we’ve additionally centered on making Zebra simpler to run in manufacturing, quicker to sync, and out there on extra platforms.
Zebra 3 at a Look
Right here’s what’s new on this launch:
- NU6.1 Community Improve – Prompts the most recent Zcash protocol improve on Mainnet
- Well being Verify Endpoints – Constructed-in HTTP endpoints for monitoring node well being and readiness in manufacturing deployments
- ARM64 Assist – Native assist for Apple Silicon Macs (M1/M2/M3), Raspberry Pi, and ARM cloud servers
- Dramatically Quicker Preliminary Sync – Full blockchain sync from genesis now takes 15-16 hours in CI and fewer than 9 hours on some machines, down from 24+ hours (representing a 35-65% discount in preliminary sync time)
- Enhanced Safety – Docker pictures now embody software program invoice of supplies (SBOM) and construct attestations
- Simpler Configuration – Configure Zebra utilizing atmosphere variables with out enhancing config recordsdata
- Improved Community Infrastructure – Zcash Basis now runs highly-available Zebra nodes to raised assist the ecosystem
- New RPC Strategies – Three new strategies for community information, mempool stats, and facet chain queries plus a number of new fields on the outputs of current RPC strategies.
- Quicker Testing – Modernized take a look at infrastructure for faster growth cycles
New Versioning method – Main model numbers now point out breaking adjustments, not simply community upgrades. This helps you perceive when updates want further consideration.
What This Means for You
If you happen to run Zebra in manufacturing
Now you can arrange well being monitoring with load balancers and orchestration instruments like Kubernetes. ARM64 assist opens up extra {hardware} choices together with cost-effective ARM cloud situations. The Zcash Basis can be now operating highly-available infrastructure which advantages all the ecosystem.
If you happen to’re establishing Zebra for the primary time
Zebra is now simpler to configure, permitting you to set atmosphere variables as an alternative of enhancing config recordsdata. Preliminary sync is dramatically quicker (15-16 hours as an alternative of 24+ hours)—and when you’re on an Apple Silicon Mac, you’ll get native ARM64 efficiency with none further steps.
If you happen to construct instruments that use Zebra
Three new RPC strategies present higher entry to community info, mempool knowledge, and facet chain queries. We’ve additionally improved transaction validation efficiency and made many enhancements to current RPC strategies.
If you happen to contribute to Zebra
Our modernized testing infrastructure means quicker CI runs and faster suggestions on pull requests. The codebase additionally advantages from quite a few code high quality enhancements and higher documentation.
The Particulars
Need to know extra? Right here’s a deeper have a look at what’s on this launch.
Community Improve & Protocol Modifications
NU6.1 Community Improve
Zebra 3.0.0 prompts NU6.1 on Mainnetwhich incorporates:
- Prolonged funding streams for community growth (1,260,000 extra blocks)
- One-time lockbox disbursements
- The newest community protocol model (170_140)
Operational Excellence
This launch consists of a number of options particularly designed to make operating Zebra in manufacturing simpler and extra dependable.
Well being Verify Endpoints for Manufacturing Operations
One of many standout options is the addition of HTTP well being test endpoints. These easy HTTP endpoints make it simple to combine Zebra with manufacturing monitoring and cargo balancing techniques.
Two endpoints can be found:
/wholesome– Returns 200 when Zebra is operating and has energetic peer connections, 503 in any other case/prepared– Returns 200 when Zebra is absolutely synced and able to serve requests, 503 in any other case
The distinction issues for manufacturing deployments: use /wholesome for fundamental liveness checks (is the method operating?) and /prepared to make sure you solely ship visitors to totally synchronized nodes with energetic peer connections.
Allow the well being test server by including a [health] part to your configuration:
[health]
listen_addr = "0.0.0.0:8080"
min_connected_peers = 1
ready_max_blocks_behind = 2
The endpoints are disabled by default and designed for inside infrastructure use (no authentication). They return easy plain-text responses like “okay” or temporary explanations when checks fail.
ARM64 Assist: Native Efficiency In every single place
Zebra now runs natively on ARM64 platforms, which implies:
- Apple Silicon Macs (M1, M2, M3) – No extra emulation or Rosetta overhead
- ARM cloud situations – AWS Graviton, Google Cloud Tau T2A, and different ARM servers
- Single-board computer systems – Raspberry Pi and related units
Docker routinely handles platform detection:
docker pull zfnd/zebra:newest
# Routinely will get the fitting model on your platformThis transformation takes benefit of GitHub’s free ARM64 runners for open supply tasks, so we will construct and take a look at ARM64 releases with out extra infrastructure prices. The result's native efficiency on ARM platforms with none setup complexity.
Enhanced Docker Picture Safety
For organizations that must audit their software program provide chain, Zebra’s Docker pictures now embody provenance attestations and Software program Invoice of Supplies (SBOM).
What this implies:
- Provenance attestations observe precisely how the picture was constructed, the place, and by whom
- SBOM supplies a whole stock of all software program parts within the picture
- Each allow safety scanning, vulnerability monitoring, and compliance reporting
These options combine with instruments like Docker Scout, Snyk, and different safety platforms that may routinely scan for recognized vulnerabilities.
Improved Community Infrastructure
The Zcash Basis has upgraded its deployment infrastructure to run long-lived, highly-available Zebra nodes. These enhancements straight profit the Zcash ecosystem:
- DNS Seeder Assist – Lengthy-lived nodes with static IP addresses present dependable peer discovery for the community
- Multi-zone Excessive Availability – Operating 2-3 situations throughout availability zones ensures constant uptime
- Automated Well being Monitoring – Makes use of the brand new
/wholesomeendpoint with auto-healing to keep up service reliability - Persistent Storage – Disk reuse between deployments eliminates prolonged re-syncs, making certain nodes keep present
These long-lived nodes with static IP addresses assist assist the broader Zcash ecosystem by offering steady infrastructure for DNS seeders and enhancing total community reliability. Whereas this work is restricted to the Basis’s infrastructure, the underlying enhancements to Zebra’s deployment tooling can be found for anybody operating manufacturing nodes.
Efficiency Enhancements
Dramatically Quicker Preliminary Sync: From 24+ Hours to Beneath 16 Hours
This launch delivers a big efficiency achievement: full blockchain synchronization from genesis now completes in 15-16 hours, down from 24+ hours in earlier variations. This represents a 35-40% discount in sync time.
The development comes primarily from PR #9973, which fastened a regression in how Zebra updates handle balances within the database. The repair avoids costly RocksDB merge operations when the database format is already up-to-date, dramatically enhancing write efficiency throughout preliminary sync.
That is mixed with batch validation enhancements for each Orchard and Sapling proofs, which scale back computational overhead throughout block validation.
Actual-world impression: Our steady integration checks present constant sync instances of 15-16 hours for full Mainnet sync from genesis, in comparison with 20-28 hours earlier than these optimizations. If you happen to’re operating a brand new Zebra node or syncing from scratch, you’ll see this enchancment instantly.
Higher Efficiency By means of Batch Validation
Transaction validation is now extra environment friendly because of:
Batch validation reduces computational overhead throughout block processing by validating teams of proofs collectively, which is extra environment friendly than particular person verification. The batch dimension for Orchard validation elevated from 2 to 64 actions per batch.
Developer Expertise
Simpler Configuration with Setting Variables
Zebra now makes use of a layered configuration system that makes it simpler to configure by means of atmosphere variables:
export ZEBRA_NETWORK__NETWORK=Mainnet
export ZEBRA_STATE__CACHE_DIR=/var/cache/zebra
export ZEBRA_RPC__LISTEN_ADDR=0.0.0.0:8232
zebrad beginThe configuration masses in layers: built-in defaults, then optionally available TOML file, then atmosphere variables. This implies you may:
- Run Zebra with zero configuration recordsdata for easy setups
- Use a base configuration file and override particular values through atmosphere variables
- Hold secrets and techniques in atmosphere variables as an alternative of config recordsdata
The sample for atmosphere variables is ZEBRA_SECTION__KEYmatching the TOML construction. For instance, [network] part’s community = "Mainnet" turns into ZEBRA_NETWORK__NETWORK=Mainnet.
Necessary: This transformation consists of breaking updates to atmosphere variable names. See the Migration Notes part under.
New RPC Strategies
Three new RPC strategies enhance compatibility with current instruments:
RPC Enhancements
Many current RPC strategies obtained enhancements:
Quicker Improvement with Improved Testing
This launch features a full modernization of the testing infrastructure utilizing cargo-nextest:
- Quicker take a look at execution – Higher parallel take a look at operating
- Centralized take a look at configuration – 17 specialised take a look at profiles in
.config/nextest.toml - Clearer take a look at outcomes – Higher progress reporting and timeout dealing with
- Simplified CI workflows – Decreased complexity in GitHub Actions
For contributors, this implies faster suggestions on pull requests and fewer time ready for CI to finish.
New Versioning Technique
Beginning with Zebra 3.0.0, we’re altering how we model releases. Beforehand, main model bumps had been primarily tied to community upgrades. Going ahead, we’ll launch a brand new main model every time there are breaking adjustments, which incorporates:
- Modifications to the Zebra API
- Modifications to operational necessities (like configuration format)
- Different modifications that require motion from node operators or builders
This method provides us a clearer strategy to talk impression. While you see a significant model bump, evaluate the discharge notes fastidiously and plan for potential updates to your deployment or integration. Minor model bumps will proceed to be backward suitable.
Migration Notes
If you happen to’re upgrading from Zebra 2.x, please notice these breaking adjustments.
Setting Variable Modifications
The atmosphere variable naming scheme modified. Key updates:
| Outdated Variable | New Variable |
|---|---|
ZEBRA_CACHE_DIR |
STATE_CACHE_DIR |
ZEBRA_FORCE_USE_COLOR |
FORCE_USE_COLOR |
RUST_LOG or ZEBRA_RUST_LOG |
ZEBRA_TRACING__FILTER |
For configuration values, use the sample ZEBRA_SECTION__KEY:
ZEBRA_NETWORK__NETWORK=MainnetZEBRA_RPC__LISTEN_ADDR=0.0.0.0:8232ZEBRA_STATE__CACHE_DIR=/var/cache/zebra
Docker Entrypoint Modifications
The Docker entrypoint now not generates TOML configuration recordsdata routinely. As a substitute:
- Configure utilizing atmosphere variables (advisable)
- Mount a configuration file into the container
- Use a mix of each approaches
Acknowledgments
An enormous thanks to all of the contributors who made this launch attainable: @arya2, @oxarbitrage, @conradoplg, @gustavovalverde, @syszery, @upbqdn, @str4d, @nuttycom, @zancas, @natalieesk, @gap-editor, @elijahhampton, @dorianvp, @AloeareV, @sashass1315, @radik878, @GarmashAlex, @Galoretka and @Fibonacci747.
Particular recognition goes to Least Authority for the safety audit of the NU6.1 implementation which supplied beneficial strategies for enhancements.
Thanks to everybody who filed points, examined launch candidates, and supplied suggestions all through this launch cycle.
The Greater Image
The Z3 Stack
Zebra 3.0.0 is designed to combine seamlessly with the opposite Z3 parts. We’ve been creating the Z3 stackwhich is a whole deployment answer for Zcash infrastructure. The stack consists of:
- Zebra – The consensus node (this launch)
- Backpack – The indexer for handle and transaction knowledge
- Zallet – The CLI pockets service
These parts work collectively to offer a whole, production-ready Zcash infrastructure that you may deploy with a single configuration.
Zebra 3.0.0 represents a big step ahead in making Zebra production-ready. The options on this launch—well being checks, ARM64 assist, and improved configuration—present a robust basis for dependable node operations.
We’re persevering with to develop the Z3 stack to offer a whole, easy-to-deploy answer for Zcash infrastructure. We’re additionally engaged on future community improve assist, together with early implementations of proposals for NU7.
Get Concerned
Questions or suggestions? We’d love to listen to about your expertise with Zebra 3.0.0. Share your ideas on the Zcash Neighborhood Discussion board, join with us on Discord, or open a difficulty within the Zebra repository:
