fix(infra): cap celery workers at 2 so the shared node stops OOM-killing CI (refs #182)

Unset CELERY_WORKER_CONCURRENCY makes oz-celery and nrc-celery fork one
process per CPU — 22 each on the lab node, ~225 MB apiece — and Talos'
OOM controller was killing the runner mid-verify-stack. Same lever as the
uWSGI caps (#144/#145/#147); objecten-celery already defaults to 1.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

infra/docker-compose.yml

@@ -57,6 +57,9 @@ services:
       # share this anchor and ignore it — they don't run uwsgi.
       UWSGI_PROCESSES: "1"
       UWSGI_THREADS: "2"
+      # Same lever for oz-celery: unset, the worker forks one process per CPU (22 on the lab node,
+      # ~225 MB each), which OOM-killed the shared runner mid-verify-stack. Only celery reads it.
+      CELERY_WORKER_CONCURRENCY: "2"
       DJANGO_SETTINGS_MODULE: openzaak.conf.docker
       SECRET_KEY: ${OZ_SECRET_KEY:-dev-only-not-for-production}
       DB_HOST: oz-db
@@ -144,6 +147,8 @@ services:
       # 1 uWSGI worker, not the image default of 4×4 (#147) — see the oz-env note above.
       UWSGI_PROCESSES: "1"
       UWSGI_THREADS: "2"
+      # Two celery workers, not one per CPU — see the oz-env note above.
+      CELERY_WORKER_CONCURRENCY: "2"
       DJANGO_SETTINGS_MODULE: nrc.conf.docker
       SECRET_KEY: ${NRC_SECRET_KEY:-dev-only-not-for-production}
       DB_HOST: nrc-db

infra/helm/big-reference/values.yaml

@@ -76,6 +76,7 @@ envGroups:
   oz:
     UWSGI_PROCESSES: "1"
     UWSGI_THREADS: "2"
+    CELERY_WORKER_CONCURRENCY: "2"
     DJANGO_SETTINGS_MODULE: openzaak.conf.docker
     SECRET_KEY: dev-only-not-for-production
     DB_HOST: oz-db
@@ -98,6 +99,7 @@ envGroups:
   nrc:
     UWSGI_PROCESSES: "1"
     UWSGI_THREADS: "2"
+    CELERY_WORKER_CONCURRENCY: "2"
     DJANGO_SETTINGS_MODULE: nrc.conf.docker
     SECRET_KEY: dev-only-not-for-production
     DB_HOST: nrc-db
@@ -275,11 +277,6 @@ workloads:
       # this issuer back, which is what browser tokens carry (infra/host-browser.yml).
       KC_HOSTNAME: '{{ include "big.keycloakUrl" . }}'
       KC_HOSTNAME_BACKCHANNEL_DYNAMIC: "true"
-      # Behind a TLS proxy (keycloakUrl) the dynamic backchannel URLs — token,
-      # userinfo, certs — take their scheme from the request, which reaches Keycloak
-      # as plain http; trusting X-Forwarded-Proto keeps them https so the browser
-      # doesn't block them as mixed content. In-cluster calls send no such header.
-      KC_PROXY_HEADERS: xforwarded
     ports: [{ name: http, port: 8080 }]
     # TCP, not /health/ready on the management port: nothing here gates on realm
     # import, and a wrong health path would leave the Service with no endpoints.