727fbfcb6b
Root-caused and fixed five sharding bugs exposed on the real k3s 3-shard cluster (rc5→rc7), plus a divergent-rejoin reseed loop found in rc9: 1. reseed shard-awareness (Bug 3, keystone): `run_boot_install_for_region` visits each hosted group's own shard subdir; per-group leader discovery appends `?shard=N` so a divergent shard heals from its own leader (not shard-0's WAL/term — cross-shard contamination). 2. leader self-join term (Bug 4): `become_leader_for_term` now calls `note_self_won_term` so the elected shard's `joined_term` is set and `cluster_promote` routes rebalances correctly (was: topology-era mis-read → legacy fenced promote → 500). 3. boot self-heal self-pull guard (Bug 2): `leader_shard != my_shard` gate prevents a node pulling its own stream (its stream isn't a registered peer) → eliminates the `PeerUnreachable(self)` loop. 4. scatter-merge degraded partial (Bug 1): failed shard logs + continues instead of `?`-failing the whole read; bounded read-admission semaphore (`offload.rs`) sheds as 429 instead of piling into a 36s p99. 5. WAL retention (Bug 5): `compact_wal_retained` keeps `WAL_RETENTION_SEGMENTS=4` most-recent sealed segments; online path gets the same retention clamp. Prevents brief-restart forced-reseed. 6. divergent-rejoin reseed loop (Bug 6, rc9): `note_quarantined` latches `from_seqno = stream_baseline` (not `frontier + 1`) so `wal_covers` returns `needed=true` and the snapshot installs instead of looping. Also: `TidalDb::close_shared` for deterministic HNSW save on cluster SIGTERM (HNSW graph was not saved when request-scoped Arc clones were alive at shutdown); updated profiling doc with full rc8/rc9 fix narrative; k8s recall job YAMLs.
112 lines
3.9 KiB
YAML
112 lines
3.9 KiB
YAML
# T4 (m12p5) — sustained load during the 3→5→3 scale (Phase B).
|
|
#
|
|
# A single generator holds a steady ack=quorum read+write mix for ~5 min, long
|
|
# enough to span a scale-down (5→3) and scale-up (3→5) under load. `--skip-seed`
|
|
# (the corpus is already registered by t4-seed-job) keeps the id space; the mix
|
|
# carries item/embed WRITES so the transition is exercised by live quorum writes,
|
|
# not just reads. `--max-error-pct 0` is the zero-acked-loss gate: a quorum write
|
|
# that cannot reach its majority during a membership change fails to ack → error
|
|
# → the Job exits non-zero. `--json-summary` captures the per-stage p99 envelope
|
|
# (before / during / after the scale) for the evidence table.
|
|
#
|
|
# NO --poll-status: this phase is UNDER LOAD (the idle-readiness isolation is the
|
|
# separate idle Phase A driven from the host), so status polling is unneeded and
|
|
# omitted to keep the offered load pure.
|
|
#
|
|
# Apply: kubectl apply -f tidal-stress/k8s/t4-load-job.yaml
|
|
# Watch: kubectl logs -f job/tidal-stress-t4-load -n tidaldb-cluster
|
|
apiVersion: batch/v1
|
|
kind: Job
|
|
metadata:
|
|
name: tidal-stress-t4-load
|
|
namespace: tidaldb-cluster
|
|
labels:
|
|
app.kubernetes.io/name: tidal-stress
|
|
app.kubernetes.io/part-of: tidaldb
|
|
spec:
|
|
backoffLimit: 0
|
|
ttlSecondsAfterFinished: 7200
|
|
template:
|
|
metadata:
|
|
labels:
|
|
app.kubernetes.io/name: tidal-stress
|
|
app.kubernetes.io/part-of: tidaldb
|
|
spec:
|
|
restartPolicy: Never
|
|
automountServiceAccountToken: false
|
|
securityContext:
|
|
runAsNonRoot: true
|
|
runAsUser: 1000
|
|
runAsGroup: 1000
|
|
seccompProfile:
|
|
type: RuntimeDefault
|
|
containers:
|
|
- name: stress
|
|
image: registry.threesix.ai/tidal/stress@sha256:4b21c1b89790f7a995f4d9e754fe0faf530079d7ad67a5ad74c15377dcc18ac2 # m12 recall/stress harness (registry)
|
|
imagePullPolicy: IfNotPresent
|
|
args:
|
|
- --target
|
|
- https://tidaldb-0.tidaldb-peers.tidaldb-cluster.svc.cluster.local:9500
|
|
- --target
|
|
- https://tidaldb-1.tidaldb-peers.tidaldb-cluster.svc.cluster.local:9500
|
|
- --target
|
|
- https://tidaldb-2.tidaldb-peers.tidaldb-cluster.svc.cluster.local:9500
|
|
- --ca-cert
|
|
- /etc/tidaldb/tls/ca.crt
|
|
- --ack
|
|
- quorum
|
|
- --skip-seed
|
|
- --corpus
|
|
- "100000"
|
|
- --embedding-dim
|
|
- "1536"
|
|
- --users
|
|
- "100000"
|
|
# ~5 min steady hold: three 100s stages so the per-stage JSON gives a
|
|
# before / during / after p99 read across the scale window.
|
|
- --ramp
|
|
- "400:100,400:100,400:100"
|
|
- --mix
|
|
- "feed=4,search=1,view=2,like=1,item=1,embed=1"
|
|
- --max-inflight
|
|
- "2000"
|
|
- --json-summary
|
|
- /tmp/t4-load-summary.json
|
|
- --max-error-pct
|
|
- "0"
|
|
env:
|
|
- name: TIDAL_API_KEY
|
|
valueFrom:
|
|
secretKeyRef:
|
|
name: tidaldb-credentials
|
|
key: TIDAL_API_KEY
|
|
- name: TIDAL_STRESS_LOG
|
|
value: warn
|
|
resources:
|
|
requests:
|
|
cpu: 500m
|
|
memory: 256Mi
|
|
limits:
|
|
cpu: "3"
|
|
memory: 1Gi
|
|
securityContext:
|
|
allowPrivilegeEscalation: false
|
|
readOnlyRootFilesystem: true
|
|
capabilities:
|
|
drop: ["ALL"]
|
|
volumeMounts:
|
|
- name: cluster-tls
|
|
mountPath: /etc/tidaldb/tls
|
|
readOnly: true
|
|
- name: tmp
|
|
mountPath: /tmp
|
|
volumes:
|
|
- name: cluster-tls
|
|
secret:
|
|
secretName: tidaldb-cluster-tls
|
|
items:
|
|
- key: ca.crt
|
|
path: ca.crt
|
|
- name: tmp
|
|
emptyDir: {}
|