Turbine randomly shuffles cluster nodes on a broadcast tree for each
shred. This requires knowing the stakes and nodes' contact-infos (from
gossip).
However gossip is subject to partitioning and propogation delays.
Additionally unstaked nodes may join and leave the cluster at any
moment, changing the cluster view from one node to another.
This commit:
* Always arranges the unstaked nodes at the bottom of turbine broadcast
tree.
* Staked nodes are always included regardless of if their contact-info
is available in gossip or not.
* Uses the unbiased WeightedShuffle construct for shuffling nodes.
Summary of Changes
Create a plugin mechanism in the accounts update path so that accounts data can be streamed out to external data stores (be it Kafka or Postgres). The plugin mechanism allows
Data stores of connection strings/credentials to be configured,
Accounts with patterns to be streamed
PostgreSQL implementation of the streaming for different destination stores to be plugged in.
The code comprises 4 major parts:
accountsdb-plugin-intf: defines the plugin interface which concrete plugin should implement.
accountsdb-plugin-manager: manages the load/unload of plugins and provide interfaces which the validator can notify of accounts update to plugins.
accountsdb-plugin-postgres: the concrete plugin implementation for PostgreSQL
The validator integrations: updated streamed right after snapshot restore and after account update from transaction processing or other real updates.
The plugin is optionally loaded on demand by new validator CLI argument -- there is no impact if the plugin is not loaded.
Shreds recovered from erasure codes have not been received from turbine
and have not been retransmitted to other nodes downstream. This results
in more repairs across the cluster which is slower.
This commit channels through recovered shreds to retransmit stage in
order to further broadcast the shreds to downstream nodes in the tree.
Working towards sending shreds (instead of packets) to retransmit stage
so that shreds recovered from erasure codes are as well retransmitted.
Following commit will add these metrics back to window-service, earlier
in the pipeline.
bank.get_leader_schedule_epoch(shred_slot)
is one epoch after epoch_schedule.get_epoch(shred_slot).
At epoch boundaries, shred is already one epoch after the root-slot. So
we need epoch-stakes 2 epochs ahead of the root. But the root bank only
has epoch-stakes for one epoch ahead, and as a result looking up epoch
staked-nodes from the root-bank fails.
To be backward compatible with the current master code, this commit
implements a fallback on working-bank if epoch staked-nodes obtained
from the root-bank is none.
The new cluster-nodes cache will:
* ensure cluster-nodes are recalculated if the epoch (and so the epoch
staked nodes) changes.
* encapsulate time-to-live eviction policy.
Broadcast stage and retransmit stage should arrange nodes on turbine
broadcast tree in exactly same order. Additionally any changes to this
ordering (e.g. updating how unstaked nodes are handled) requires feature
gating to keep the cluster in sync.
Current implementation is scattered out over several public methods and
exposes too much of implementation details (e.g. usize indices into
peers vector) which makes code changes and checking for feature
activations more difficult.
This commit encapsulates turbine peer computations into a new struct,
and only exposes two public methods, get_broadcast_peer and
get_retransmit_peers, for call-sites.
* Move gossip modules to solana-gossip
* Update Protocol abi digest due to move
* Move gossip benches and hook up CI
* Remove unneeded Result entries
* Single use statements
In order to remove port-based forwarding logic in turbine, we need to
first track how often the turbine retransmit/broadcast trees mismatch
across nodes.
One consistency condition is that if the node is on the critical path
(i.e. the first node in each neighborhood), then we expect that the
packet arrives at tvu socket as opposed to tvu-forwards.
This commit adds a metric to track how often above condition is not met.
* Upgrade Rust to 1.52.0
update nightly_version to newly pushed docker image
fix clippy lint errors
1.52 comes with grcov 0.8.0, include this version to script
* upgrade to Rust 1.52.1
* disabling Serum from downstream projects until it is upgraded to Rust 1.52.1
* Require that blockstore block-time only be recognized slot, instead of root
* Move cache_block_time to after Bank freeze
* Single use statement
* Pass transaction_status_sender by reference
* Remove unnecessary slot-existence check before caching block time altogether
* Move block-time existence check into Blockstore::cache_block_time, Blockstore no longer needed in blockstore_processor helper
CodingShredHeader.position is equal to
ShredCommonHeader.index - ShredCommonHeader.fec_set_index
and is so redundant. The extra position field can add bugs if not
consistent with index and fec_set_index.
behzad nouri
Lijun Wang
Brooks Prumo
Brooks Prumo
Ryo Onodera
sakridge
carllin
Jeff Washington (jwash)
Michael Vines
Tyera Eulberg
Tao Zhu
Justin Starry