OpenEMR on GKE Autopilot

OpenEMR is the world's most widely adopted open-source Electronic Health Records (EHR) and practice management system, used by 100,000+ healthcare providers across 100+ countries. This module deploys OpenEMR on GKE Autopilot on top of the App_GKE foundation, which provisions and manages the shared Google Cloud and Kubernetes infrastructure.

This guide focuses on the cloud services OpenEMR uses and how to explore and operate them from the Google Cloud Console and the command line. For the mechanics common to every GKE application — Workload Identity, ingress, autoscaling, CI/CD, Cloud Armor, IAP, Binary Authorization, VPC Service Controls, backups, and the deployment lifecycle — refer to the App_GKE foundation guide rather than repeating them here.

---

1. Overview

OpenEMR runs as an Apache/PHP 8.3 FPM workload on Alpine 3.20. The deployment wires together a focused set of Google Cloud services:

Capability	Google Cloud service	Notes
Compute	GKE Autopilot	Apache/PHP pods, 2 vCPU / 4 GiB by default, horizontally autoscaled
Database	Cloud SQL for MySQL 8.0	Required — OpenEMR does not support PostgreSQL
Patient documents	Filestore (NFS)	sites/ directory with patient documents, session cache, and application state shared across all replicas
Object storage	Cloud Storage	A general-purpose data bucket
Session store	Redis	Enabled by default; falls back to the NFS server IP when no Redis host is given
Secrets	Secret Manager	Auto-generated admin password (OE_PASS) and database password (MYSQL_PASS)
Ingress	Cloud Load Balancing	External LoadBalancer, optional custom domain + managed certificate

Sensible defaults worth knowing up front:

• MySQL 8.0 is mandatory. The database engine is fixed; selecting PostgreSQL or NONE breaks startup.
• NFS is mandatory. OpenEMR's sites/ directory — containing sqlconf.php, patient documents, Twig/Smarty caches, and uploaded files — must be on a shared NFS volume. The application cannot function without it.
• Redis is enabled by default. When max_instance_count > 1, a shared session store is required to prevent PHP session loss across pods.
• Session affinity is ClientIP. OpenEMR relies on PHP sessions, so requests from a browser are pinned to one pod.
• First-boot installation is automated and slow. On first deploy, three initialization jobs run in sequence: nfs-init (NFS directory setup), db-init (MySQL user and database creation), and openemr-install (schema installation via auto_configure.php). The startup probe allows up to 120 seconds for the application to become ready after jobs complete.
• The OpenEMR admin password is generated automatically and stored in Secret Manager; you never set it in plain text.
• min_instance_count defaults to 1. Scale-to-zero is not recommended for clinical EHR systems — cold starts add 20–40 seconds of latency that clinicians may interpret as a system failure.

---

2. Google Cloud Services & How to Explore Them

All commands assume you have run gcloud container clusters get-credentials <cluster> --region <region> --project <project> and that PROJECT, REGION, and NAMESPACE are set. The namespace and other identifiers are reported in the deployment Outputs.

A. GKE Autopilot — the OpenEMR workload

OpenEMR pods are scheduled on Autopilot, which bills for the CPU/memory the pods actually request. Horizontal Pod Autoscaling sizes the deployment between the minimum and maximum replica counts.

• Console: Kubernetes Engine → Workloads → select the OpenEMR workload to see pods, events, and probe status. Kubernetes Engine → Services & Ingress shows the external IP.
• CLI:

  kubectl get pods,svc,hpa -n "$NAMESPACE"
  kubectl logs -n "$NAMESPACE" deploy/<service-name> --tail=100
  kubectl describe hpa -n "$NAMESPACE"          # current vs target utilisation

See App_GKE for how Autopilot, scaling, and the workload type (Deployment vs StatefulSet) are managed.

B. Cloud SQL for MySQL 8.0

OpenEMR stores all clinical data (patient records, scheduling, billing) in a managed Cloud SQL for MySQL 8.0 instance. Pods reach it privately through the Cloud SQL Auth Proxy sidecar over a Unix socket, so no public IP is exposed. On first deploy the db-init job creates the application database and user before the openemr-install job runs the OpenEMR schema installer.

• Console: SQL → select the instance for connections, backups, flags, and metrics.
• CLI:

  gcloud sql instances list --project "$PROJECT"
  gcloud sql instances describe <instance-name> --project "$PROJECT"
  # Open an interactive shell to inspect schema/data:
  gcloud sql connect <instance-name> --user=<db-user> --project "$PROJECT"

The instance name, database name, user, and the Secret Manager secret holding the password are all surfaced in the Outputs. For the connection model, automated backups, and password rotation, see App_GKE.

C. Filestore (NFS) and Cloud Storage

OpenEMR's sites/ directory is written to a Filestore (NFS) share mounted into every pod at /var/www/localhost/htdocs/openemr/sites. This directory contains sqlconf.php (which signals installation completion), patient-uploaded documents, Twig/Smarty template caches, and session data. All replicas must share the same NFS mount. A general-purpose Cloud Storage bucket is also provisioned.

• Console: Filestore → Instances for the NFS share; Cloud Storage → Buckets for the data bucket.
• CLI:

  gcloud filestore instances list --project "$PROJECT"
  gcloud storage buckets list --project "$PROJECT"
  # Confirm the NFS share is mounted and sites directory exists:
  kubectl exec -n "$NAMESPACE" deploy/<service-name> -- \
    ls /var/www/localhost/htdocs/openemr/sites/default/

See App_GKE for NFS provisioning, GCS Fuse, and CMEK options.

D. Redis session store

Redis backs OpenEMR's PHP session store. When redis_host is left empty and NFS is enabled, the NFS server's co-located Redis instance is used automatically. In multi-replica deployments, a shared session store is required to prevent session loss.

• Console: Memorystore → Redis (if using a managed Memorystore instance).
• CLI:

  redis-cli -h <redis-host> ping        # from a host with network access
  redis-cli -h <redis-host> info keyspace
  # Confirm REDIS_SERVER is set inside the pod:
  kubectl exec -n "$NAMESPACE" deploy/<service-name> -- env | grep REDIS

E. Secret Manager

The OpenEMR admin password (OE_PASS) and the MySQL database password (MYSQL_PASS) are stored as Secret Manager secrets and injected into pods at runtime via the Secret Store CSI driver. Plaintext never appears in configuration.

• Console: Security → Secret Manager.
• CLI:

  gcloud secrets list --project "$PROJECT"
  # Retrieve the admin password to log in for the first time:
  gcloud secrets versions access latest \
    --secret=<admin-password-secret-id> --project "$PROJECT"

The admin password secret ID is exposed as the admin_password_secret_id output. The database password secret name is in database_password_secret. See App_GKE for the Secret Store CSI integration and rotation.

F. Networking & ingress

By default the workload is exposed through an external Cloud Load Balancing IP. A custom domain with a Google-managed certificate can be enabled, and a static IP can be reserved so the address survives redeploys.

• Console: Network services → Load balancing; VPC network → IP addresses.
• CLI:

  kubectl get ingress,svc -n "$NAMESPACE"
  gcloud compute addresses list --project "$PROJECT"
  # Test the login page from within the cluster:
  kubectl exec -n "$NAMESPACE" deploy/<service-name> -- \
    curl -s -o /dev/null -w "%{http_code}" http://localhost/interface/login/login.php

See App_GKE for custom domains, Cloud CDN, and static IP details.

G. Cloud Logging & Monitoring

Pod stdout/stderr flow to Cloud Logging; GKE and Cloud SQL metrics flow to Cloud Monitoring. Optional uptime checks and alert policies are available.

• Console: Logging → Logs Explorer; Monitoring → Dashboards / Alerting.
• CLI:

  gcloud logging read \
    'resource.type="k8s_container" AND resource.labels.namespace_name="'"$NAMESPACE"'"' \
    --project "$PROJECT" --limit 50

---

3. OpenEMR Application Behaviour

• Three-stage first-deploy initialization. The following Kubernetes Jobs run in sequence on every apply:

  Job	Purpose	Depends on
  nfs-init	Prepares the NFS sites/ directory structure, sets ownership to UID 1000 (Apache), and optionally restores a backup	—
  db-init	Creates the MySQL database and application user	—
  openemr-install	Runs auto_configure.php in K8S=admin mode to install the database schema and create the admin account; writes $config=1 to sqlconf.php on NFS	nfs-init, db-init

  The main application pod starts only after openemr-install completes and sees $config=1 in sqlconf.php — it then skips the installer and begins serving. Inspect the jobs:

  kubectl get jobs -n "$NAMESPACE"
  kubectl logs -n "$NAMESPACE" job/nfs-init
  kubectl logs -n "$NAMESPACE" job/openemr-install

• Startup can take 5–20 minutes on first boot. The openemr-install job runs the full PHP schema installer, which is slow. The startup probe (TCP on port 80, 12 failure threshold) allows 120 seconds for the pod to become ready after jobs complete. On a truly fresh install consider raising failure_threshold in the startup probe.

• Version-aware upgrades. On subsequent deployments the startup script compares the image version against the NFS-stored version and runs the appropriate upgrade scripts (fsupgrade-N.sh) automatically.

• Temporary health probe server. During the installation phase openemr.sh starts a PHP built-in web server on port 80 that returns HTTP 200 on the health probe path, preventing the pod from being killed while the installer runs.

• Admin login. The initial administrator username is admin. The password is auto-generated and stored in Secret Manager — retrieve it with:

  gcloud secrets versions access latest \
    --secret=<admin_password_secret_id> --project "$PROJECT"

  If the admin account is locked after failed login attempts, use the /root/unlock_admin.sh <new_password> utility inside the container.

• K8S=yes environment variable. The application receives K8S=yes at runtime, which instructs openemr.sh to use the Kubernetes-aware startup path (skipping slow recursive chown operations that would cause timeout failures).

---

4. Configuration Variables

Variables are grouped exactly as they appear on the deployment platform. Only settings specific to or notable for OpenEMR are listed; every other input is inherited from App_GKE with its standard behaviour and defaults.

Group 1 — Project & Identity

Variable	Default	Description
project_id	(required)	Target Google Cloud project.
region	us-central1	Region for the workload and regional resources.

Group 2 — Deployment Environment

Variable	Default	Description
tenant_deployment_id	demo	Short suffix that makes resource names unique per environment.
support_users	[]	Emails granted project access and monitoring alerts.
resource_labels	{}	Labels applied to all resources for cost/ownership tracking.

Group 3 — Application Identity

Variable	Default	Description
application_name	openemr	Base name for resources. Do not change after first deploy.
application_version	7.0.4	OpenEMR image version tag; increment to roll out a new version.
display_name	OpenEMR	Friendly name shown in the Console and dashboards.
description	(set)	Workload description annotation.

Group 4 — Runtime & Scaling

Variable	Default	Description
deploy_application	true	Set false to provision infrastructure only.
cpu_limit	2000m	CPU per pod; 2 vCPU recommended for concurrent clinical workloads.
memory_limit	4Gi	Memory per pod; 4 GiB minimum for production.
ephemeral_storage_limit	8Gi	Ephemeral storage for PHP opcache, Apache logs, and temp files. GKE Autopilot caps total pod ephemeral storage at 10 GiB; the Auth Proxy sidecar uses ~1 GiB, leaving a maximum of 9 GiB.
min_instance_count	1	Minimum replicas. Keep ≥ 1 to avoid cold-start delays for clinical users.
max_instance_count	1	Increase only after confirming Redis session sharing is operational.
enable_cloudsql_volume	true	Cloud SQL Auth Proxy sidecar. Not user-configurable in this module.
enable_vertical_pod_autoscaling	false	Let Autopilot tune resource requests automatically.
deployment_timeout	1800	Seconds Terraform waits for rollout. Extended for OpenEMR's long first-boot install.

Group 5 — Environment Variables & Secrets

Variable	Default	Description
environment_variables	{}	Extra plain-text settings. Core MYSQL_* and OE_* values are set automatically. Common additions: PHP_MEMORY_LIMIT, SMTP_HOST, SMTP_PORT.
secret_environment_variables	{}	Map of env var → Secret Manager secret name. Use for sensitive values such as SMTP credentials.
secret_propagation_delay	30	Seconds to wait after secret creation before proceeding.

Group 6 — GKE Backend & Cluster

Variable	Default	Description
service_type	LoadBalancer	How the Service is exposed.
session_affinity	ClientIP	Sticky routing required for OpenEMR PHP sessions.
workload_type	null	Auto-resolves to StatefulSet when per-pod storage is enabled.
network_tags	['nfsserver']	Required for NFS connectivity via VPC firewall rules. Do not remove.

Group 7 — StatefulSet

Variable	Default	Description
stateful_pvc_enabled	null	Enable PVC templates. OpenEMR uses NFS, not PVCs — leave unset.
stateful_pvc_size / stateful_pvc_mount_path / stateful_pvc_storage_class	(set)	PVC options when StatefulSet mode is explicitly needed.

Group 8 — Resource Quota

Variable	Default	Description
enable_resource_quota	false	Cap namespace CPU/memory/object counts.
quota_memory_requests / quota_memory_limits	""	Must use binary units (4Gi, 8192Mi) — bare integers are read as bytes and block scheduling.

Group 9 — Reliability Policies

Variable	Default	Description
enable_pod_disruption_budget	false	Disabled by default because max_instance_count = 1 — a PDB with min_available = 1 blocks node drains on a single pod. Enable only when running 2+ replicas.
pdb_min_available	1	Raise min_instance_count above 1 if you need eviction headroom.
enable_topology_spread	false	Spread pods across zones.

Group 10 — Observability & Health

Variable	Default	Description
startup_probe	TCP on port 80, 12 failures × 10s	TCP probe; allows up to 120 seconds for startup. Increase failure_threshold for first-time deploys with large databases.
liveness_probe	HTTP GET /interface/login/login.php, 10 failures × 30s	Login page returns HTTP 200 only when Apache, PHP-FPM, and the database connection are all operational.
uptime_check_config	disabled	Optional Cloud Monitoring uptime check.
alert_policies	[]	Optional metric alert policies.

Group 11 — Workload Automation

Variable	Default	Description
initialization_jobs	[]	Leave empty to use the built-in nfs-init / db-init / openemr-install sequence.
cron_jobs	[]	Scheduled CronJobs (e.g., backup, report generation).

Group 12 — CI/CD & GitHub Integration

Standard App_GKE Cloud Build / Cloud Deploy integration — see App_GKE. Key inputs: enable_cicd_trigger, github_repository_url, github_token, enable_cloud_deploy.

Group 13 — Filesystem (NFS)

Variable	Default	Description
enable_nfs	true	Must remain true. OpenEMR requires NFS for the sites/ directory.
nfs_mount_path	/var/www/localhost/htdocs/openemr/sites	Mount path inside the container. Must match the OpenEMR sites directory path.

Group 14 — Cloud Storage & Artifact Registry

Variable	Default	Description
create_cloud_storage	true	Provision the data bucket.
storage_buckets	[{name_suffix="data"}]	Additional buckets.
manage_storage_kms_iam / enable_artifact_registry_cmek	false	CMEK options.

Group 15 — Redis Session Store

Variable	Default	Description
enable_redis	true	Use Redis for PHP session storage.
redis_host	""	Leave empty to use the NFS server IP; set explicitly for a dedicated Memorystore instance.
redis_port	6379	Redis port.
redis_auth	""	Optional Redis auth password (sensitive).

Group 16 — Database Backend

Variable	Default	Description
db_name	openemr	MySQL database name. Immutable after first deploy.
db_user	openemr	Application user. Immutable after first deploy.
database_password_length	32	Generated password length (16–64).
enable_auto_password_rotation	false	Zero-downtime DB password rotation. Requires pod restart to pick up the new secret.

Group 17 — Backup & Maintenance

Variable	Default	Description
backup_schedule	0 2 * * *	Automated backup cron (UTC). Do not disable for HIPAA-regulated deployments.
backup_retention_days	7	Retention; raise to 30–90 for production/compliance.
enable_backup_import	false	Restore from a backup on deploy.
backup_source	gcs	Import source: gcs or gdrive.
backup_uri	""	GCS URI (gs://bucket/path) or Google Drive file ID. When set, injected into nfs-init as BACKUP_FILEID.
backup_format	sql	Backup file format: sql, tar, gz, tgz, tar.gz, or zip.

Group 18 — Custom SQL Scripts

enable_custom_sql_scripts, custom_sql_scripts_bucket, custom_sql_scripts_path, custom_sql_scripts_use_root — run SQL from a GCS bucket after provisioning. See App_GKE.

Group 19 — Custom Domain, Static IP & Networking

Variable	Default	Description
enable_custom_domain	true	Provision Kubernetes Gateway for custom hostnames + managed certificate (a Gateway with a static IP is provisioned automatically).
application_domains	[]	Hostnames to serve.
reserve_static_ip	true	Stable external IP across redeploys.

Group 20 — Identity-Aware Proxy (IAP)

Variable	Default	Description
enable_iap	false	Require Google sign-in in front of OpenEMR. Recommended for restricting access to clinical staff.
iap_authorized_users / iap_authorized_groups	[]	Who may access.
iap_oauth_client_id / iap_oauth_client_secret	""	Required when IAP is enabled (sensitive).
iap_support_email	""	Shown on the OAuth consent screen.

Group 21 — Cloud Armor

Variable	Default	Description
enable_cloud_armor	false	Attach a Cloud Armor (WAF) policy to the Ingress backend.
admin_ip_ranges	[]	CIDRs allowed privileged access.
cloud_armor_policy_name	default-waf-policy	Policy name.

Group 22 — VPC Service Controls & Audit Logging

Variable	Default	Description
enable_vpc_sc	false	Enforce a VPC-SC perimeter. Requires organization_id to be set explicitly. Recommended for HIPAA environments.
vpc_cidr_ranges / vpc_sc_dry_run	(set)	Access level CIDRs / dry-run mode.
enable_audit_logging	false	Detailed Cloud Audit Logs (DATA_READ, DATA_WRITE). Recommended for HIPAA compliance.

---

5. Outputs

These values are returned on a successful deployment and are the quickest way to locate and explore the running resources.

Output	Description
service_name	Kubernetes Service name.
namespace	Namespace the workload runs in.
service_cluster_ip	In-cluster ClusterIP.
stage_service_cluster_ips	Map of ClusterIPs for stage-specific services (Cloud Deploy).
service_external_ip	External LoadBalancer IP (when a static IP is reserved).
service_url	URL to reach OpenEMR.
admin_password_secret_id	Secret Manager secret ID for the OpenEMR admin password (OE_PASS).
database_instance_name	Cloud SQL instance name.
database_name	Application database name.
database_user	Application database user.
database_password_secret	Secret Manager secret holding the DB password (MYSQL_PASS).
database_host / database_port	DB endpoint (127.0.0.1 via the Auth Proxy) / port.
nfs_server_ip	Internal IP of the NFS server (sensitive).
nfs_mount_path	NFS mount path inside the container.
nfs_share_path	NFS share path on the server.
nfs_setup_job	Name of the NFS setup job.
storage_buckets	Created Cloud Storage buckets.
network_name / network_exists / regions	VPC network, presence, available regions.
container_image / container_registry	Deployed image and Artifact Registry repo.
monitoring_enabled / monitoring_notification_channels	Monitoring status and channels.
initialization_jobs / db_import_job	Names of the setup and (optional) import jobs.
deployment_id / tenant_id / resource_prefix	Naming identifiers.
project_id / project_number	Project identifiers.
cicd_enabled / cicd_configuration	CI/CD status and details (repo, trigger, registry).
github_repository_url / github_repository_owner / github_repository_name	GitHub repo details.
artifact_registry_repository / cloudbuild_trigger_name / cloudbuild_trigger_id	Registry and build trigger.
kubernetes_ready	Whether the cluster/workload is ready.
vpc_sc_enabled / vpc_sc_perimeter_name / vpc_sc_dry_run_mode	VPC-SC status.
audit_logging_enabled / artifact_registry_cmek_enabled	Audit logging and CMEK status.

---

6. Configuration Pitfalls & Sensible Defaults

Risk: Critical (data loss / outage / security) — High (service degraded) — Medium (cost or partial degradation) — Low (minor).

Setting	Sensible value	Risk	Consequence if wrong
enable_nfs	true	Critical	OpenEMR cannot function without NFS. The sites/ directory, sqlconf.php, and patient documents all live on NFS. Disabling causes immediate startup failure.
nfs_mount_path	/var/www/localhost/htdocs/openemr/sites	Critical	Must match the OpenEMR sites directory path. A mismatch means openemr-install writes sqlconf.php to a location the main pod never checks — the pod waits indefinitely for setup completion.
database_type (via OpenEMR_Common)	MYSQL_8_0	Critical	OpenEMR requires MySQL; PostgreSQL or NONE breaks the installer and all PHP database calls.
db_name / db_user	set once	Critical	Immutable after first deploy; renaming recreates the DB/user and destroys all patient data.
enable_backup_import	false unless restoring	Critical	Enabling without a valid backup_uri fails the import job and can corrupt the NFS sites directory.
quota_memory_requests / _limits	binary units (4Gi)	Critical	Bare integers are bytes and block all pod scheduling.
backup_schedule	0 2 * * *	Critical	Disabling backups for an EHR containing PHI is a HIPAA compliance violation.
ephemeral_storage_limit	8Gi	Critical	OpenEMR writes PHP opcache, Apache logs, and temp files to the container layer. GKE Autopilot's default 1 GiB is insufficient — the pod is evicted during startup.
enable_redis	true	High	With >1 replica, isolated per-pod PHP sessions cause login failures and session loss.
redis_host	"" (NFS) or explicit	High	An unreachable Redis host causes PHP session failures and prevents all logins.
memory_limit	4Gi	High	OpenEMR PDF generation and billing reports are memory-intensive. Less than 2 GiB causes OOM kills mid-request.
session_affinity	ClientIP	High	Without stickiness, multi-replica deployments lose session state between requests.
min_instance_count	1	High	Scale-to-zero causes cold-start delays of 20–40 seconds — unacceptable for clinical access.
enable_pod_disruption_budget	enable when min_instance_count > 1	High	Disabled by default because max_instance_count = 1 — a PDB would permanently block node drains on a single-pod deployment. Enable when scaling beyond one replica.
backup_retention_days	7 (raise for prod)	Medium	HIPAA-regulated environments should retain at least 90 days.
enable_iap / enable_cloud_armor	enable for healthcare	Medium	The OpenEMR admin interface is publicly reachable without these controls.
enable_audit_logging	true for HIPAA	Medium	HIPAA requires audit logging of access to PHI.
enable_vpc_sc	set organization_id explicitly	Medium	Without an explicit org ID, VPC-SC silently skips perimeter creation — leaving a false sense of security.

---

For the foundation behaviour referenced throughout — IAM and Workload Identity, autoscaling, ingress and certificates, CI/CD, Cloud Armor, IAP, Binary Authorization, VPC-SC, backups, and image mirroring — see App_GKE. OpenEMR-specific application configuration shared with the Cloud Run variant is described in OpenEMR_Common.