31ee612f27
Scale write throughput across data-shard groups while keeping a single unified read surface: - scatter_gather.rs: pooled fan-out across shard groups (replaces per-request client construction); cross-shard query results merged on one node - cluster/node.rs: cross-shard read routing — a read on any node gathers from every shard group's leader and unions results - cluster/forward.rs: fix h2 204 forward-relay bug (relay_forwarded skips body for 1xx/204/304 — synthesized JSON body on a 204 triggered HTTP/2 RST_STREAM on the real mTLS plane) - dto.rs: cross-shard query/result DTOs - k8s/cluster/: enable 3-group `shards:` topology (statefulset, service-peers, topology-configmap) - k8s/cluster-local-kind/: local-kind overlay to run the T5 gate without Ref-A - tidal-stress/k8s/stress-job-t5.yaml: 2-generator sharded throughput job - tests: cluster_cross_shard_reads.rs + multiproc support; ran real on kind - docs/profiling/m12p4-t5-sharded-throughput.md: T5 throughput findings
54 lines
2.1 KiB
YAML
54 lines
2.1 KiB
YAML
# Headless PEER-discovery Service (m11p5 §4).
|
|
#
|
|
# Gives every StatefulSet pod a stable per-pod DNS name
|
|
# tidaldb-N.tidaldb-peers.tidaldb-cluster.svc.cluster.local
|
|
# which is exactly what the topology ConfigMap advertises as each region's
|
|
# grpc_addr/http_addr. Peers dial these names; tonic re-resolves DNS on every
|
|
# reconnect so a pod rescheduled onto a new IP becomes reachable with no peer
|
|
# restart (the whole point of the m11p5 bind/advertise split).
|
|
#
|
|
# publishNotReadyAddresses: true is LOAD-BEARING. The m11p5 readiness predicate
|
|
# 503s a joiner/quarantined/reseeding pod (so the CLIENT Service drops it from
|
|
# load-balancing — see service-client.yaml). But peer discovery, snapshot fetch,
|
|
# and the catch-up stream must still RESOLVE a not-ready joiner's DNS, or the
|
|
# joiner can never reach a seed to converge and become ready — a deadlock. This
|
|
# Service keeps NOT-ready pods in its DNS A-records for exactly that reason.
|
|
#
|
|
# This Service is also `serviceName` for the StatefulSet (stable network ids).
|
|
apiVersion: v1
|
|
kind: Service
|
|
metadata:
|
|
name: tidaldb-peers
|
|
namespace: tidaldb-cluster
|
|
labels:
|
|
app.kubernetes.io/name: tidaldb
|
|
app.kubernetes.io/part-of: tidaldb
|
|
spec:
|
|
clusterIP: None # headless: per-pod DNS, no VIP
|
|
publishNotReadyAddresses: true # keep not-ready joiners resolvable for peers
|
|
selector:
|
|
app.kubernetes.io/name: tidaldb
|
|
app.kubernetes.io/component: cluster-node
|
|
ports:
|
|
- name: http
|
|
port: 9500
|
|
targetPort: http
|
|
# One gRPC port per hosted shard group (m11p6/m12p4 3-group topology). A
|
|
# headless Service routes by pod DNS → pod IP, so peers dial the derived
|
|
# per-group ports (9601/9602/9603) on the pod directly regardless of this
|
|
# list; the entries are declared so the named ports stay discoverable and
|
|
# match statefulset.yaml. With the legacy single group only `grpc` (9601)
|
|
# is bound.
|
|
- name: grpc
|
|
port: 9601
|
|
targetPort: grpc
|
|
- name: grpc-1
|
|
port: 9602
|
|
targetPort: grpc-1
|
|
- name: grpc-2
|
|
port: 9603
|
|
targetPort: grpc-2
|
|
- name: metrics
|
|
port: 9091
|
|
targetPort: metrics
|