KAFKA_SASL_MECHANISM is read from auth.sasl_mechanism when the nested auth
map exists and from the top-level sasl_mechanism otherwise, so the pods start
both before and after kafka/apps/pm switches to creds delivered from the
kafka-topics terraform stack.
flows, issues, pm and message-hub read ca.crt from a copy of the Kafka TLS
secret in their own namespace instead of an inline PEM (flows) or a
per-namespace kafka-ca-cert ConfigMap (issues, pm, message-hub). Mount paths
and env names are unchanged. The secret must exist in each namespace before
the pods restart.
universal-chart already injects proxy.istio.io/config and
traffic.sidecar.istio.io/excludeOutboundPorts by default for every
service — explicitly setting them too produced a duplicate-key YAML
error in Flux's post-render step:
error while running post render on files: ... yaml: unmarshal errors:
line 42: mapping key "traffic.sidecar.istio.io/excludeOutboundPorts" already defined at line 25
line 41: mapping key "proxy.istio.io/config" already defined at line 26
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
New HelmRelease services.admin-frontend in apps/control-interface/base,
matching the live Deployment's image/port/resources (cpu 100m, memory
100Mi) and istio tracing podAnnotations. Downward-API envs (K8S_POD_UID/
K8S_POD_NAME/K8S_NAMESPACE/OTEL_RESOURCE_ATTRIBUTES) were left out — no
existing app in this repo uses valueFrom/fieldRef in the universal-chart
envs schema and the chart source isn't reachable to confirm support.
imagePullSecrets uses regcred (vad's actual convention) instead of the
source's dockerhub.
Since control-interface/vad and /uralkal both just inherit ../base
unmodified, this also shows up in uralkal as a side effect.
infrastructure/istio-config/vad: adds a plain admin-frontend route
(/admin-frontend/static/ -> admin-frontend-svc.control-interface, rewrite
/), matching the minimal style of the other sarex.vadroad.ru routes —
no cors block, per request.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds 13 new VirtualServices on the existing test.sarex.brusnika.tech host
and ingress-nginx/main-gateway, matching the active (non-commented)
location blocks in global-ingress's nginx-configmap (fetched live from the
cluster and cross-checked service/port/namespace names against what's
actually running). The root path (/) stays routed to
nginx-service.global-ingress as a fallback for anything not covered here.
Two deliberate deviations from literally replaying the nginx config:
- /comparisons/api/: nginx proxies to port 8080, but the real
backend-service.comparisons Service listens on 80 (targetPort 8080) —
used 80.
- /orchestrator/: nginx declares 4 location blocks, but the first
(bare ~^/orchestrator/) shadows the other three for any non-empty
path (nginx picks the first matching regex location, not the most
specific), so only one route (no rewrite) was ported, matching what
nginx actually does today.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Missing env vars copied from wb for ams-sync, attachments, bi/frontend,
checklists, django (celery, export-project, sarex-backend), documentations
(api, filestream, pdm), eav, flows (backend, celery, frontend), iam,
inspections, issues (backend, celery), rfi, transmittal/worker.
Image tags updated to match wb for flows (backend/celery/frontend), iam.
Known issue, not yet fixed in this commit: several of the copied env
values are wb-specific hosts (*.wb.ru, one uralmine.com) that don't apply
to brusnika-stage — ZITADEL_HOST/ZITADEL_DOMAIN (django, documentations,
iam), DATABASE_HOST/FLOWS_DB_HOST/ISSUES_DB_HOST (checklists, flows),
SUPERSET_HOST (bi), DJANGO_BASE_HOST/SAREX_BACKEND_URL (flows,
inspections), RESOURCES_INTERNAL_HOST/RESOURCE_URL (ams-sync, django,
flows, notes-related). To be corrected in a follow-up commit.
bi/backend.yaml, bim/backend.yaml and notes/backend.yaml were reverted
before this commit (image-tag updates for bi-backend/bim/notes and bi's
SUPERSET_* env additions are not included).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Every services.<svc>.deployment.resources.requests.{cpu,memory} across the
36 uralkal apps (69 HelmRelease/service entries total) is now nulled via a
kustomize patch, so Helm never renders a requests block for these pods on
uralkal — Kubernetes won't reserve CPU/memory for them there.
Where a uralkal patch already existed for that service (13 cases:
documentations api/filestream/pdf-markings-amqp, django backend, flows
backend/celery, pm backend/celery, transmittal backend/worker, bi,
document-link, stamp-verification, message-hub), the null block was added
into that same file. Where no uralkal patch existed yet (55 cases,
including all 13 cde workers and every plain frontend), a new minimal
patch file was added and wired into that app's kustomization.yaml.
vad and the other clusters are untouched — only apps/*/uralkal/* changed.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
