> ## Documentation Index
> Fetch the complete documentation index at: https://ayakaleaf-pro.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Horizontal scaling

> Run several Ayakaleaf Pro replicas behind a load balancer with sticky sessions.

One replica uses the CPU of one node for every compile: on our test nodes with 2 vCPUs, six simultaneous compiles took three times as long as one, while the other nodes stayed idle. Several replicas spread users, and their compiles, over the nodes.

This follows Overleaf's [Horizontal scaling](https://ayakaleaf-pro.mintlify.site/on-premises/maintenance/horizontal-scaling.md) design, translated to Kubernetes:

| Requirement | Here |
| - | - |
| Shared MongoDB and Redis | `10-mongo.yaml` and `20-redis.yaml` from the installation |
| Shared secrets | the Secret `overleaf-secrets` from [step 2](https://ayakaleaf-pro.mintlify.site/on-premises/installation/using-kubernetes/2.-create-the-namespace-and-secrets.md) already has them |
| Project files and history in **S3** (not NFS) | any S3 service, see [S3](https://ayakaleaf-pro.mintlify.site/on-premises/configuration/kubernetes/s3.md); Garage there as an in-cluster example |
| A local disk per replica for compiles and uploads | `hostPath` `/srv/overleaf-ha` on each node |
| Persistent routing, WebSockets, 10 min timeouts, 50 MB uploads | HAProxy with Overleaf's sample configuration |
| git-bridge as a singleton next to one replica | git-bridge with `overleaf-0` as its sibling |

> [!WARNING]
>
> Plan S3 from the start. An instance that already keeps files on its volume needs the [S3 migration](https://ayakaleaf-pro.mintlify.site/on-premises/maintenance/s3-migration.md) first. This setup replaces the Deployment from step 5: delete it (`kubectl delete -f 40-overleaf.yaml`) before you apply the StatefulSet.

**1. S3 storage**

Set up the three buckets and two users as described in [S3](https://ayakaleaf-pro.mintlify.site/on-premises/configuration/kubernetes/s3.md), with your S3 service or with Garage in the cluster, and put the four keys into the Secret `overleaf-s3`. The configuration below uses the Garage example from that page.

**2. Configuration**

This ConfigMap replaces `30-config.yaml`. It adds S3 and git-bridge and points `OVERLEAF_SITE_URL` at the load balancer:

```yaml title="60-config.yaml" wrap theme={null}
apiVersion: v1
kind: ConfigMap
metadata:
  name: overleaf-env
  namespace: overleaf
data:
  OVERLEAF_APP_NAME: Ayakaleaf Pro
  OVERLEAF_SITE_URL: http://192.0.2.10:30081      # the load balancer (HAProxy) URL
  OVERLEAF_ADMIN_EMAIL: admin@example.com
  # Emails: add the sender and the SMTP server together, see Email delivery
  OVERLEAF_MONGO_URL: mongodb://mongo-0.mongo:27017,mongo-1.mongo:27017,mongo-2.mongo:27017/sharelatex?replicaSet=overleaf
  OVERLEAF_REDIS_HOST: redis
  REDIS_HOST: redis
  ENABLED_LINKED_FILE_TYPES: project_file,project_output_file,url
  ENABLE_CONVERSIONS: "true"
  EMAIL_CONFIRMATION_DISABLED: "true"
  OVERLEAF_TRUSTED_PROXY_IPS: loopback,10.244.0.0/16   # HAProxy runs in the pod CIDR
  # Sandboxed compiles, node paths: each replica owns /srv/overleaf-ha on its node
  SANDBOXED_COMPILES: "true"
  SANDBOXED_COMPILES_SIBLING_CONTAINERS: "true"
  DOCKER_RUNNER: "true"
  SANDBOXED_COMPILES_HOST_DIR_COMPILES: /srv/overleaf-ha/data/compiles
  SANDBOXED_COMPILES_HOST_DIR_OUTPUT: /srv/overleaf-ha/data/output
  SANDBOXED_COMPILES_HOST_DIR_CACHE: /srv/overleaf-ha/data/cache
  ALL_TEX_LIVE_DOCKER_IMAGES: ghcr.io/ayaka-notes/texlive-full:2026.1
  ALL_TEX_LIVE_DOCKER_IMAGE_NAMES: TeX Live 2026
  TEX_LIVE_DOCKER_IMAGE: ghcr.io/ayaka-notes/texlive-full:2026.1
  # S3 (required for more than one replica); keys are in the Secret overleaf-s3
  OVERLEAF_FILESTORE_BACKEND: s3
  OVERLEAF_FILESTORE_TEMPLATE_FILES_BUCKET_NAME: overleaf-template-files
  OVERLEAF_FILESTORE_S3_ENDPOINT: http://garage:3900
  OVERLEAF_FILESTORE_S3_PATH_STYLE: "true"
  OVERLEAF_FILESTORE_S3_REGION: garage
  OVERLEAF_HISTORY_BACKEND: s3
  OVERLEAF_HISTORY_PROJECT_BLOBS_BUCKET: overleaf-project-blobs
  OVERLEAF_HISTORY_CHUNKS_BUCKET: overleaf-chunks
  OVERLEAF_HISTORY_S3_ENDPOINT: http://garage:3900
  OVERLEAF_HISTORY_S3_PATH_STYLE: "true"
  OVERLEAF_HISTORY_S3_REGION: garage
  # Git integration: one git-bridge, its sibling is overleaf-0
  GIT_BRIDGE_ENABLED: "true"
  GIT_BRIDGE_HOST: 10.96.100.80      # the git-bridge Service's fixed ClusterIP, see 75-git-bridge.yaml
  GIT_BRIDGE_PORT: "8000"
```

**3. Replicas**

```yaml title="70-overleaf.yaml" wrap theme={null}
apiVersion: v1
kind: Service
metadata:
  name: overleaf-hl
  namespace: overleaf
spec:
  clusterIP: None            # stable DNS per replica: overleaf-0.overleaf-hl, ...
  selector:
    app: overleaf-ha
  ports:
    - {name: http, port: 80}
    - {name: web, port: 3000}       # git-bridge API (sibling only)
    - {name: history, port: 3100}   # history-v1, used by git-bridge via V1_HISTORY_URL
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: overleaf
  namespace: overleaf
spec:
  serviceName: overleaf-hl
  replicas: 3
  # OrderedReady: overleaf-0 runs the migrations alone before the next replica starts
  podManagementPolicy: OrderedReady
  selector:
    matchLabels:
      app: overleaf-ha
  template:
    metadata:
      labels:
        app: overleaf-ha
    spec:
      terminationGracePeriodSeconds: 60
      # Keep the environment to the ConfigMap and Secrets: no REDIS_PORT=tcp://..., GIT_BRIDGE_PORT=tcp://...
      enableServiceLinks: false
      affinity:
        podAntiAffinity:            # one replica per node: each uses that node's Docker and disk
          requiredDuringSchedulingIgnoredDuringExecution:
            - labelSelector:
                matchLabels:
                  app: overleaf-ha
              topologyKey: kubernetes.io/hostname
      containers:
        - name: sharelatex
          image: ghcr.io/ayaka-notes/overleaf-pro:6.3.2
          env:
            - name: POD_NAME
              valueFrom:
                fieldRef:
                  fieldPath: metadata.name
            # Each replica hands out history URLs with its own stable name, so git-bridge can reach them
            - name: V1_HISTORY_URL
              value: http://$(POD_NAME).overleaf-hl.overleaf.svc.cluster.local:3100/api
          envFrom:
            - configMapRef: {name: overleaf-env}
            - secretRef: {name: overleaf-secrets}
            - secretRef: {name: overleaf-s3}
          ports:
            - containerPort: 80
          readinessProbe:
            httpGet: {path: /status, port: 80}
            periodSeconds: 10
          resources:
            requests: {cpu: 500m, memory: 1536Mi}
            limits: {memory: 3Gi}
          volumeMounts:
            - {name: local, mountPath: /var/lib/overleaf}
            - {name: docker-sock, mountPath: /var/run/docker.sock}
      volumes:
        # Local disk for ephemeral files (compiles, output, uploads); the HOST_DIR_* paths are under it
        - name: local
          hostPath: {path: /srv/overleaf-ha, type: DirectoryOrCreate}
        - name: docker-sock
          hostPath: {path: /var/run/docker.sock, type: Socket}
```

* **StatefulSet, not Deployment.** Each replica gets a stable name (`overleaf-0`, `overleaf-1`, ...) that the load balancer cookie and git-bridge refer to, and `OrderedReady` starts `overleaf-0` alone first, so only one replica runs the database migrations.
* **One replica per node.** Each replica starts its compiles through its node's Docker and keeps them in its node's `/srv/overleaf-ha`.
* **`V1_HISTORY_URL` per replica.** Each replica hands out history links with its own name, so that git-bridge can reach them.

**4. git-bridge**

```yaml title="75-git-bridge.yaml" wrap theme={null}
apiVersion: v1
kind: Service
metadata:
  name: git-bridge
  namespace: overleaf
spec:
  # Fixed ClusterIP (inside the service CIDR): the sharelatex image's nginx proxies /git/ with
  # `resolver 127.0.0.11` (Docker's DNS), which does not exist in Kubernetes, so GIT_BRIDGE_HOST is this IP
  clusterIP: 10.96.100.80
  selector:
    app: git-bridge
  ports:
    - {name: http, port: 8000}
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: git-bridge
  namespace: overleaf
spec:
  serviceName: git-bridge
  replicas: 1
  selector:
    matchLabels:
      app: git-bridge
  template:
    metadata:
      labels:
        app: git-bridge
    spec:
      # Service "git-bridge" would inject GIT_BRIDGE_PORT=tcp://..., which git-bridge reads as its port
      enableServiceLinks: false
      containers:
        - name: git-bridge
          image: ghcr.io/ayaka-notes/overleaf-pro/git-bridge:6.3.2   # same version as the sharelatex image
          command: ["/server-pro-start.sh"]
          securityContext:
            runAsUser: 0
          env:
            - {name: GIT_BRIDGE_API_BASE_URL, value: "http://overleaf-0.overleaf-hl.overleaf.svc.cluster.local:3000/api/v0/"}
            - {name: GIT_BRIDGE_OAUTH2_SERVER, value: "http://overleaf-0.overleaf-hl.overleaf.svc.cluster.local"}
            - {name: GIT_BRIDGE_POSTBACK_BASE_URL, value: "http://git-bridge.overleaf.svc.cluster.local:8000"}
            - {name: GIT_BRIDGE_ROOT_DIR, value: /data/git-bridge}
          ports:
            - containerPort: 8000
          resources:
            requests: {cpu: 100m, memory: 256Mi}
            limits: {memory: 768Mi}
          volumeMounts:
            - {name: data, mountPath: /data/git-bridge}
  volumeClaimTemplates:
    - metadata:
        name: data
      spec:
        accessModes: [ReadWriteOnce]
        resources:
          requests:
            storage: 5Gi
```

See [Git integration](https://ayakaleaf-pro.mintlify.site/on-premises/configuration/kubernetes/integrations/git-integration.md) for why `GIT_BRIDGE_HOST` is a fixed IP address.

**5. Load balancer**

The HAProxy configuration is Overleaf's sample with five changes: `/project/<id>/mcp` routed to `overleaf-0` like `/git/` (see [MCP Server](https://ayakaleaf-pro.mintlify.site/on-premises/configuration/kubernetes/integrations/mcp-server.md#several-replicas)), plain HTTP on port `8080` (the image runs as a non-root user, which cannot bind port 80 under containerd), servers found through cluster DNS (pod IPs change on every restart), a DaemonSet so that every node answers, and `balance leastconn` instead of the default `roundrobin` (see the Q\&A below).

```yaml title="80-haproxy.yaml" wrap theme={null}
apiVersion: v1
kind: ConfigMap
metadata:
  name: haproxy-config
  namespace: overleaf
data:
  haproxy.cfg: |
    global
      log stdout format raw local0 info

    resolvers kubedns
      parse-resolv-conf
      hold valid 10s

    defaults
      mode                    http
      # New sessions go to the replica with the fewest open connections (editors keep a WebSocket open)
      balance                 leastconn
      option                  httpchk HEAD /status
      http-check              expect status 200
      default-server          check resolvers kubedns init-addr none
      log                     global
      option                  httplog
      # Reroute to a different backend if the sticky one is down
      option                  redispatch 1
      retries                 3
      # Sticky session for 24h of inactivity -- compile output is deleted after 24h
      cookie                  server-pro-ha insert maxidle 24h
      timeout queue           1m
      timeout connect         15s
      timeout client          1m
      # Allow slow compiles -- hard-coded limit in clsi is 10min
      timeout server          10m
      # Disconnect the editor after 23h
      timeout tunnel          23h
      # Lower than NGINX_KEEPALIVE_TIMEOUT (65s) in the sharelatex container
      timeout http-keep-alive 60s

    listen overleaf
      bind :8080
      option forwardfor
      # When TLS ends in front of HAProxy, also pass:  http-request set-header X-Forwarded-Proto https
      # git requests go to the git-bridge sibling
      use-server overleaf-0 if { path_beg /git/ }
      # MCP clients send no cookie: keep compile, log and PDF download on one replica
      use-server overleaf-0 if { path_reg ^/project/[0-9a-f]{24}/mcp(/|$) }
      http-response add-header X-Served-By %s
      stats enable
      stats uri /haproxy
      server overleaf-0 overleaf-0.overleaf-hl.overleaf.svc.cluster.local:80 cookie overleaf-0
      server overleaf-1 overleaf-1.overleaf-hl.overleaf.svc.cluster.local:80 cookie overleaf-1
      server overleaf-2 overleaf-2.overleaf-hl.overleaf.svc.cluster.local:80 cookie overleaf-2
---
apiVersion: apps/v1
# DaemonSet: one HAProxy per node, so the NodePort answers on every node with externalTrafficPolicy
# Local. The cookie holds the replica name, so stickiness works whichever HAProxy a request hits.
kind: DaemonSet
metadata:
  name: haproxy
  namespace: overleaf
spec:
  selector:
    matchLabels:
      app: haproxy
  template:
    metadata:
      labels:
        app: haproxy
      annotations:
        config-hash: "1"   # bump to restart after editing the ConfigMap
    spec:
      containers:
        - name: haproxy
          image: haproxy:lts-alpine
          ports:
            - containerPort: 8080
          resources:
            requests: {cpu: 50m, memory: 32Mi}
            limits: {memory: 256Mi}
          volumeMounts:
            - {name: config, mountPath: /usr/local/etc/haproxy}
      volumes:
        - name: config
          configMap:
            name: haproxy-config
---
apiVersion: v1
kind: Service
metadata:
  name: haproxy
  namespace: overleaf
spec:
  type: NodePort
  externalTrafficPolicy: Local    # keep the client IP (no SNAT); only nodes running HAProxy answer
  selector:
    app: haproxy
  ports:
    - {name: http, port: 80, targetPort: 8080, nodePort: 30081}
```

```bash theme={null}
kubectl delete -f 40-overleaf.yaml --ignore-not-found
kubectl apply -f 60-config.yaml -f 70-overleaf.yaml -f 75-git-bridge.yaml -f 80-haproxy.yaml
kubectl -n overleaf rollout status statefulset/overleaf --timeout=20m
curl -sI http://192.0.2.10:30081/login | grep -i x-served-by
```

`externalTrafficPolicy: Local` keeps the users' IP addresses in `X-Forwarded-For`.

**6. TLS**

Terminate HTTPS in HAProxy as described in [TLS proxy](https://ayakaleaf-pro.mintlify.site/on-premises/configuration/kubernetes/tls-proxy.md), then set `OVERLEAF_SITE_URL` to the `https://` address and `OVERLEAF_SECURE_COOKIE: "true"` in `60-config.yaml`.

### What to expect

We tested three replicas on three nodes with 2 vCPUs and 4 GB each, with MongoDB, Redis, Garage, git-bridge and HAProxy on the same nodes:

* 12 users compiling a 600-page document at the same time: all 12 compiled and fetched their PDF, 4 users per replica; a round took 8 to 18 seconds. Each node kept at least 1.3 GB of memory available.
* Every user stays on one replica (cookie `server-pro-ha`), including the PDF download that must come from the replica that compiled it. The same holds over HTTPS, including the editor's WebSocket.
* When a replica stops, its users get errors for about 15 to 20 seconds, until HAProxy marks it down; then they move to another replica and stay logged in, because sessions live in Redis.

### Q\&A

> [accordion] Why `balance leastconn`?
>
> **Answer:** With Overleaf's sample (`roundrobin`, one HAProxy per node), 12 new users ended up 8, 3 and 1 on the three replicas in our test, and the busiest replica made the slowest users wait 29 seconds. With `leastconn`, which prefers the replica with the fewest open connections (each editor keeps a WebSocket open), the same test gave 4, 4 and 4 and 18 seconds. Stickiness is unchanged: it comes from the cookie.

> [accordion] Can I use my ingress controller instead of HAProxy?
>
> **Answer:** Yes, if it routes by a cookie to a stable replica, supports WebSockets, allows request times of 10 minutes and uploads of 50 MB, keeps its keep-alive timeout below 65 seconds, sends `X-Forwarded-For` and `X-Forwarded-Proto`, and sends `/git/` and `/project/<id>/mcp` to `overleaf-0`. Route to the replicas' own names (`overleaf-N.overleaf-hl`), not to a Service that balances every request.

> [accordion] How do I add a fourth replica?
>
> **Answer:** Add a node with Docker and the TeX Live images, set `replicas: 4` in `70-overleaf.yaml`, and add a `server overleaf-3 ...` line to the HAProxy configuration (and bump `config-hash` so HAProxy restarts).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.