Merge pull request #58 from langchain-ai/infra/open-gpts

Add open-gpts helm chart

.github/workflows/helm_checks.yaml

@@ -37,4 +37,5 @@ jobs:
       - name: Run chart-testing (install)
         env:
           LANGSMITH_LICENSE_KEY: ${{ secrets.LANGSMITH_LICENSE_KEY }}
-        run: ct install --target-branch ${{ github.event.repository.default_branch }} --all --helm-extra-set-args "--set=frontend.service.type=ClusterIP --set=config.langsmithLicenseKey=$LANGSMITH_LICENSE_KEY"
+          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+        run: ct install --target-branch ${{ github.event.repository.default_branch }} --all --helm-extra-set-args "--set=frontend.service.type=ClusterIP --set=config.langsmithLicenseKey=$LANGSMITH_LICENSE_KEY --set=config.openaiApiKey=$OPENAI_API_KEY --set=config.tavilyApiKey=foo --set=config.kayApiKey=foo --set=config.ydcApiKey=foo --set=backend.service.type=ClusterIP"

charts/langsmith/templates/_helpers.tpl

@@ -93,7 +93,7 @@ the user or some other secret provisioning mechanism
 {{- end }}
 
 {{/*
-Name of the secret containing the secrets for postgres. This can be overriden by a secrets file created by
+Name of the secret containing the secrets for redis. This can be overriden by a secrets file created by
 the user or some other secret provisioning mechanism
 */}}
 {{- define "langsmith.redisSecretsName" -}}

charts/langsmith/templates/redis/stateful-set.yaml

@@ -134,4 +134,4 @@ spec:
         storageClassName: {{ .Values.redis.statefulSet.persistence.storageClassName }}
         {{- end }}
   {{- end }}
-{{- end }}
+{{- end }}

charts/open-gpts/.helmignore

@@ -0,0 +1,23 @@
+# Patterns to ignore when building packages.
+# This supports shell glob matching, relative path matching, and
+# negation (prefixed with !). Only one pattern per line.
+.DS_Store
+# Common VCS dirs
+.git/
+.gitignore
+.bzr/
+.bzrignore
+.hg/
+.hgignore
+.svn/
+# Common backup files
+*.swp
+*.bak
+*.tmp
+*.orig
+*~
+# Various IDEs
+.project
+.idea/
+*.tmproj
+.vscode/

charts/open-gpts/Chart.yaml

@@ -0,0 +1,9 @@
+apiVersion: v2
+name: open-gpts
+maintainers:
+  - name: Ankush
+    email: [email protected]
+description: Helm chart to deploy the open-gpts appplication and all services it depends on.
+type: application
+version: 0.1.0
+appVersion: "0.1.0"

charts/open-gpts/README.md

@@ -0,0 +1,282 @@
+# open-gpts
+
+![Version: 0.1.0](https://img.shields.io/badge/Version-0.1.0-informational?style=flat-square) ![Type: application](https://img.shields.io/badge/Type-application-informational?style=flat-square) ![AppVersion: 0.1.0](https://img.shields.io/badge/AppVersion-0.1.0-informational?style=flat-square)
+
+Helm chart to deploy the open-gpts appplication and all services it depends on.
+
+## Deploying OpenGPTs with Helm
+
+### Prerequisites
+
+Ensure you have the following tools/items ready.
+
+1. A working Kubernetes cluster that you can access via `kubectl`
+    1. Recommended: Atleast 1 vCPUs, 4GB Memory available
+        1. You may need to tune resource requests/limits for all of our different services based off of organization size/usage
+    2. Valid Dynamic PV provisioner or PVs available on your cluster. You can verify this by running:
+
+        ```jsx
+        kubectl get storageclass
+        ```
+2. `Helm`
+    1. `brew install helm`
+3. SSL(optional)
+    1. This should be attachable to a load balancer that the chart will provision
+4. API Keys(optional).
+    1. Used by various SDKS. Configure based on the tools you are using.
+5. External Redis(optional).
+    1. You can configure external redis using the `values.yaml` file. You will need to provide a connection url for your redis instance.
+    2. Currently, we do not support using Redis with TLS. We will be supporting this shortly.
+    3. We only official support Redis versions >= 6.
+    4. We rely on the vector store module for Redis. Ensure that your Redis instance has this module installed.
+
+### Configure your Helm Charts:
+
+1. Create a copy of `values.yaml`
+2. Override any values in the file. Refer to the `values.yaml` documentation below to see all configurable values. Some values we recommend tuning:
+    1. Resources
+    2. SSL
+        1. Add an annotation to the `backend.service` object to tell your cloud provider to provision a load balancer with said certificate attached.
+        2. This will vary based on your cloud provider. Refer to their documentation for more information.
+        3. To support the above, we e
+    3. Api Keys
+    4. Images
+
+Bare minimum config file `open_gpts_config.yaml`:
+
+```yaml
+config:
+  openaiApiKey: "foo"
+  tavilyApiKey: "foo"
+  kayApiKey: "foo"
+  ydcApiKey: "foo"
+```
+
+Example `EKS` config file with certificates setup using ACM:
+
+```jsx
+config:
+  openaiApiKey: "foo"
+  tavilyApiKey: "foo"
+  kayApiKey: "foo"
+  ydcApiKey: "foo"
+
+backend:
+  service:
+    annotations:
+      service.beta.kubernetes.io/aws-load-balancer-type: "nlb"
+      service.beta.kubernetes.io/aws-load-balancer-ssl-ports: "https"
+      service.beta.kubernetes.io/aws-load-balancer-ssl-cert: "<certificate arn>"
+```
+
+Example config file with external redis:
+
+```jsx
+config:
+  openaiApiKey: "foo"
+  tavilyApiKey: "foo"
+  kayApiKey: "foo"
+  ydcApiKey: "foo"
+redis:
+  external:
+    enabled: true
+    connectionUrl: "redis://<url>:6379"
+```
+
+You can also use existingSecretName to avoid checking in secrets. This secret will need to follow
+the same format as the secret in the corresponding `secrets.yaml` file.
+
+### Deploying to Kubernetes:
+
+1. Verify that you can connect to your Kubernetes cluster(note: We highly suggest installing into an empty namespace)
+    1. Run `kubectl get pods`
+
+        Output should look something like:
+
+        ```bash
+        kubectl get pods
+        No resources found in default namespace.
+        ```
+
+2. Ensure you have the Langchain Helm repo added. (skip this step if you are using local charts)
+
+        helm repo add langchain https://langchain-ai.github.io/helm/
+        "langchain" has been added to your repositories
+
+3. Run `helm install open-gpts langchain/open-gpts --values open_gpts_config.yaml`
+4. Run `kubectl get pods`
+    1. Output should now look something like:
+
+    ```bash
+    NAME                                 READY   STATUS    RESTARTS      AGE
+    open-gpts-backend-849c5c49bd-5hrth   1/1     Running   0             3m17s
+    open-gpts-redis-0                    1/1     Running   0             3m17s
+    ```
+
+### Validate your deployment:
+
+1. Run `kubectl get services`
+
+    Output should look something like:
+
+    ```bash
+    NAME                TYPE           CLUSTER-IP       EXTERNAL-IP                                                               PORT(S)        AGE
+    open-gpts-backend   LoadBalancer   172.20.158.106   <ip>                                                                      80:31767/TCP   7m22s
+    open-gpts-redis     ClusterIP      172.20.181.57    <none>                                                                    6379/TCP       7m22s
+    ```
+
+3. Visit the external ip for the `open-gpts-backend` service on your browser
+
+    The Open-GPTs UI should be visible/operational
+
+    ![./open_gpts_ui.png](open_gpts_ui.png)
+
+## FAQ:
+
+1. How can we upgrade our application?
+    - To upgrade, you will need to follow the upgrade instructions in the Helm README and run a `helm upgrade open-gpts --values <values file>`
+2. How can we backup our application?
+    - Currently, we rely on PVCs/PV to power storage for our application. We strongly encourage setting up `Persistent Volume` backups or moving to a managed service for `Postgres` to support disaster recovery
+3. How does load balancing work/ingress work?
+    - Currently, our application spins up one load balancer using a k8s service of type `LoadBalancer` for our frontend. If you do not want to setup a load balancer you can simply port-forward the frontend and use that as your external ip for the application.
+    - We also have an option for the chart to provision an ingress resource for the application.
+4. How can we authenticate to the application?
+    - Currently, we do not support auth.
+5. How can I use External `Redis`?
+    - You can configure external redis using the external sections in the `values.yaml` file. You will need to provide the connection url/params for the redis instance. Look at the configuration above example for more information.
+6. What networking configuration is needed  for the application?
+    - Our deployment only needs egress for a few things:
+        - Fetching images (If mirroring your images, this may not be needed)
+        - Talking to any LLMs
+    - Your VPC can set up rules to limit any other access.
+7. What resources should we allocate to the application?
+    - We recommend at least 4 vCPUs and 16GB of memory for our application.
+    - We have some default resources set in our `values.yaml` file. You can override these values to tune resource usage for your organization.
+    - If the metrics server is enabled in your cluster, we also recommend enabling autoscaling on all deployments.
+
+## General parameters
+
+| Key | Type | Default | Description |
+|-----|------|---------|-------------|
+| commonAnnotations | object | `{}` | Annotations that will be applied to all resources created by the chart |
+| commonLabels | object | `{}` | Labels that will be applied to all resources created by the chart |
+| fullnameOverride | string | `""` | String to fully override `"openGPTs.fullname"` |
+| images.backendImage.pullPolicy | string | `"Always"` |  |
+| images.backendImage.repository | string | `"docker.io/langchain/open-gpts"` |  |
+| images.backendImage.tag | string | `"latest"` |  |
+| images.imagePullSecrets | list | `[]` | Secrets with credentials to pull images from a private registry. Specified as name: value. |
+| images.redisImage.pullPolicy | string | `"IfNotPresent"` |  |
+| images.redisImage.repository | string | `"docker.io/redis/redis-stack-server"` |  |
+| images.redisImage.tag | string | `"latest"` |  |
+| ingress.annotations | object | `{}` |  |
+| ingress.enabled | bool | `false` |  |
+| ingress.hostname | string | `""` |  |
+| ingress.ingressClassName | string | `""` |  |
+| ingress.labels | object | `{}` |  |
+| ingress.tls | list | `[]` |  |
+| nameOverride | string | `""` | Provide a name in place of `open-gpts` |
+
+## Configs
+
+| Key | Type | Default | Description |
+|-----|------|---------|-------------|
+| config.existingSecretName | string | `""` |  |
+| config.kayApiKey | string | `""` |  |
+| config.openaiApiKey | string | `""` |  |
+| config.tavilyApiKey | string | `""` |  |
+| config.ydcApiKey | string | `""` |  |
+
+## Backend
+
+| Key | Type | Default | Description |
+|-----|------|---------|-------------|
+| backend.autoscaling.enabled | bool | `false` |  |
+| backend.autoscaling.maxReplicas | int | `5` |  |
+| backend.autoscaling.minReplicas | int | `1` |  |
+| backend.autoscaling.targetCPUUtilizationPercentage | int | `80` |  |
+| backend.containerPort | int | `8000` |  |
+| backend.deployment.affinity | object | `{}` |  |
+| backend.deployment.annotations | object | `{}` |  |
+| backend.deployment.extraEnv | list | `[]` |  |
+| backend.deployment.labels | object | `{}` |  |
+| backend.deployment.nodeSelector | object | `{}` |  |
+| backend.deployment.podSecurityContext | object | `{}` |  |
+| backend.deployment.replicaCount | int | `1` |  |
+| backend.deployment.resources | object | `{}` |  |
+| backend.deployment.securityContext | object | `{}` |  |
+| backend.deployment.sidecars | list | `[]` |  |
+| backend.deployment.tolerations | list | `[]` |  |
+| backend.deployment.volumeMounts | list | `[]` |  |
+| backend.deployment.volumes | list | `[]` |  |
+| backend.migrations.affinity | object | `{}` |  |
+| backend.migrations.annotations | object | `{}` |  |
+| backend.migrations.enabled | bool | `true` |  |
+| backend.migrations.extraEnv | list | `[]` |  |
+| backend.migrations.labels | object | `{}` |  |
+| backend.migrations.nodeSelector | object | `{}` |  |
+| backend.migrations.podSecurityContext | object | `{}` |  |
+| backend.migrations.replicaCount | int | `1` |  |
+| backend.migrations.resources | object | `{}` |  |
+| backend.migrations.securityContext | object | `{}` |  |
+| backend.migrations.sidecars | list | `[]` |  |
+| backend.migrations.tolerations | list | `[]` |  |
+| backend.migrations.volumeMounts | list | `[]` |  |
+| backend.migrations.volumes | list | `[]` |  |
+| backend.name | string | `"backend"` |  |
+| backend.service.annotations | object | `{}` |  |
+| backend.service.httpPort | int | `80` |  |
+| backend.service.httpsPort | int | `443` |  |
+| backend.service.labels | object | `{}` |  |
+| backend.service.loadBalancerIP | string | `""` |  |
+| backend.service.loadBalancerSourceRanges | list | `[]` |  |
+| backend.service.type | string | `"LoadBalancer"` |  |
+| backend.serviceAccount.annotations | object | `{}` |  |
+| backend.serviceAccount.create | bool | `true` |  |
+| backend.serviceAccount.labels | object | `{}` |  |
+| backend.serviceAccount.name | string | `""` |  |
+
+## Redis
+
+| Key | Type | Default | Description |
+|-----|------|---------|-------------|
+| redis.containerPort | int | `6379` |  |
+| redis.external.connectionUrl | string | `""` |  |
+| redis.external.enabled | bool | `false` |  |
+| redis.external.existingSecretName | string | `""` |  |
+| redis.name | string | `"redis"` |  |
+| redis.service.annotations | object | `{}` |  |
+| redis.service.labels | object | `{}` |  |
+| redis.service.loadBalancerIP | string | `""` |  |
+| redis.service.loadBalancerSourceRanges | list | `[]` |  |
+| redis.service.port | int | `6379` |  |
+| redis.service.type | string | `"ClusterIP"` |  |
+| redis.serviceAccount.annotations | object | `{}` |  |
+| redis.serviceAccount.create | bool | `true` |  |
+| redis.serviceAccount.labels | object | `{}` |  |
+| redis.serviceAccount.name | string | `""` |  |
+| redis.statefulSet.affinity | object | `{}` |  |
+| redis.statefulSet.annotations | object | `{}` |  |
+| redis.statefulSet.extraEnv | list | `[]` |  |
+| redis.statefulSet.labels | object | `{}` |  |
+| redis.statefulSet.nodeSelector | object | `{}` |  |
+| redis.statefulSet.persistence.enabled | bool | `false` |  |
+| redis.statefulSet.persistence.size | string | `"8Gi"` |  |
+| redis.statefulSet.persistence.storageClassName | string | `""` |  |
+| redis.statefulSet.podSecurityContext | object | `{}` |  |
+| redis.statefulSet.resources | object | `{}` |  |
+| redis.statefulSet.securityContext | object | `{}` |  |
+| redis.statefulSet.sidecars | list | `[]` |  |
+| redis.statefulSet.tolerations | list | `[]` |  |
+| redis.statefulSet.volumeMounts | list | `[]` |  |
+| redis.statefulSet.volumes | list | `[]` |  |
+
+## Maintainers
+
+| Name | Email | Url |
+| ---- | ------ | --- |
+| Ankush | <ankush@langchain.dev> |  |
+
+----------------------------------------------
+Autogenerated from chart metadata using [helm-docs v1.11.3](https://github.com/norwoodj/helm-docs/releases/v1.11.3)
+## Docs Generated by [helm-docs](https://github.com/norwoodj/helm-docs)
+`helm-docs -t ./charts/opengpts/README.md.gotmpl`