SDK server prod

.github/workflows/build-test-deploy.yml

@@ -925,7 +925,7 @@ jobs:
     env:
       AWS_PROFILE: aws-deployer-connext
       TF_VAR_cartographer_image_tag: ${{ github.sha }}
-      TF_VAR_sdk_server_image_tag: ${{ github.sha }}
+      TF_VAR_full_image_name_sdk_server: ${{ fromJSON(needs.smoke-tests.outputs.sdk-server-tags).tags[0] }}
 
     runs-on: ubuntu-latest
     permissions:
@@ -1122,13 +1122,15 @@ jobs:
       router-executor-tags: ${{ needs.e2e-tests.outputs.router-executor-tags }}
       relayer-tags: ${{ needs.e2e-tests.outputs.relayer-tags }}
       watcher-tags: ${{ needs.e2e-tests.outputs.watcher-tags }}
+      sdk-server-tags: ${{ needs.e2e-tests.outputs.sdk-server-tags }}
 
   terraform-services-backend-prod-testnet:
     if: github.ref == 'refs/heads/testnet-prod' || github.ref == 'refs/heads/prod'
     needs: [e2e-tests]
     env:
       AWS_PROFILE: aws-deployer-connext
       TF_VAR_cartographer_image_tag: ${{ github.sha }}
+      TF_VAR_full_image_name_sdk_server: ${{ fromJSON(needs.e2e-tests.outputs.sdk-server-tags).tags[0] }}
 
     runs-on: ubuntu-latest
     permissions:
@@ -1188,13 +1190,15 @@ jobs:
       router-executor-tags: ${{ needs.e2e-tests.outputs.router-executor-tags }}
       relayer-tags: ${{ needs.e2e-tests.outputs.relayer-tags }}
       watcher-tags: ${{ needs.e2e-tests.outputs.watcher-tags }}
+      sdk-server-tags: ${{ needs.e2e-tests.outputs.sdk-server-tags }}
 
   terraform-services-backend-prod-mainnet:
     if: github.ref == 'refs/heads/prod'
     needs: [terraform-services-core-prod-testnet, terraform-services-backend-prod-testnet]
     env:
       AWS_PROFILE: aws-deployer-connext
       TF_VAR_cartographer_image_tag: ${{ github.sha }}
+      TF_VAR_full_image_name_sdk_server: ${{ fromJSON(needs.terraform-services-backend-prod-testnet.outputs.sdk-server-tags).tags[0] }}
 
     runs-on: ubuntu-latest
     permissions:

docker/README.md

@@ -4,15 +4,20 @@ Config.json located in the root of each project's docker ie (repo-root)/docker/<
 
 - router (pretty much just runs the router as configured).
 
-- integration 
+- integration
 
+Run everything from the root directory.
 
-Building:
-(from the repo root directory): 
-ex.
-```[sudo] bash docker/router/build-router.sh```
+Build:
 
-Running:
-ex.
-```<ENV variables> [sudo] docker run -d --name router nxtp-router```
+```
+docker build -f docker/sdk-server/Dockerfile -t sdk-server .
+```
 
+> NOTE: If you are using an M1/M2 mac, you must provide the flag "`--platform linux/amd64`"! Also if you are using Docker Desktop, make sure you turn ON the "Use Virtualization framework" option in Settings > General and turn OFF the "Use Rosetta for x86/amd64 emulation on Apple Silicon" option in Settings > Features in development.
+
+Run:
+
+```
+docker run -it sdk-server
+```

docker/sdk-server/Dockerfile

@@ -1,7 +1,7 @@
 FROM node:18-alpine as node
 
 # ----------------------------------------
-# Builds Docker container for nxtp-watcher package.
+# Builds Docker container for sdk-server package.
 
 # ----------------------------------------
 
@@ -23,6 +23,7 @@ ARG TEMP_DEPS_DIR
 COPY .yarn /tmp/build/.yarn/
 COPY .yarnrc.yml /tmp/build/
 COPY package.json /tmp/build/
+COPY packages/examples/sdk-server/package.json /tmp/build/packages/examples/sdk-server/
 COPY packages/agents/sdk/package.json /tmp/build/packages/agents/sdk/
 COPY packages/adapters/cache/package.json /tmp/build/packages/adapters/cache/
 COPY packages/adapters/subgraph/package.json /tmp/build/packages/adapters/subgraph/
@@ -38,6 +39,7 @@ RUN yarn install
 
 # ----- Copy source and all other files that affect lint, test, build -----
 COPY config config/
+COPY packages/examples/sdk-server /tmp/build/packages/examples/sdk-server
 COPY packages/agents/sdk /tmp/build/packages/agents/sdk
 COPY packages/adapters/cache /tmp/build/packages/adapters/cache
 COPY packages/adapters/subgraph /tmp/build/packages/adapters/subgraph
@@ -68,7 +70,7 @@ COPY --from=build --chown=node:node /tmp/build /home/node
 # This user is created in the base image with uid and gid = 1000.
 USER node
 
-WORKDIR /home/node/packages/agents/sdk
+WORKDIR /home/node/packages/examples/sdk-server
 
 EXPOSE 8080
 

ops/modules/db-alarms/main.tf

@@ -0,0 +1,64 @@
+locals {
+  thresholds = {
+    CPUUtilizationThreshold   = var.cpu_utilization_threshold
+    FreeStorageSpaceThreshold = var.free_storage_space_threshold
+  }
+
+  alarm_names = toset([
+    "cpu_utilization_too_high",
+    "free_storage_space_threshold",
+  ])
+}
+
+
+resource "aws_sns_topic_subscription" "target" {
+  count     = var.enable_cpu_utilization_alarm || var.enable_free_storage_space_too_low_alarm ? length(var.sns_topic_subscription_emails) : 0
+  topic_arn = aws_sns_topic.topic[0].arn
+  protocol  = "email"
+  endpoint  = var.sns_topic_subscription_emails[count.index]
+}
+
+resource "aws_sns_topic" "topic" {
+  count = var.enable_cpu_utilization_alarm || var.enable_free_storage_space_too_low_alarm ? 1 : 0
+  name  = "${var.environment}-${var.stage}-${var.is_replica ? "replica-" : ""}db-sns-topic"
+}
+
+
+resource "aws_cloudwatch_metric_alarm" "cpu_utilization_too_high" {
+  count               = var.enable_cpu_utilization_alarm ? 1 : 0
+  alarm_name          = "${var.db_instance_name}-cpu-utilization-too-high"
+  comparison_operator = "GreaterThanThreshold"
+  evaluation_periods  = "1"
+  metric_name         = "CPUUtilization"
+  namespace           = "AWS/RDS"
+  period              = "600"
+  statistic           = "Average"
+  threshold           = local.thresholds["CPUUtilizationThreshold"]
+  alarm_description   = "Average database CPU utilization over last 10 minutes too high"
+  alarm_actions       = aws_sns_topic.topic.*.arn
+  ok_actions          = aws_sns_topic.topic.*.arn
+
+  dimensions = {
+    DBInstanceIdentifier = var.db_instance_id
+  }
+}
+
+
+resource "aws_cloudwatch_metric_alarm" "free_storage_space_too_low" {
+  count               = var.enable_free_storage_space_too_low_alarm ? 1 : 0
+  alarm_name          = "${var.db_instance_name}-free-storage-space-too-low"
+  comparison_operator = "LessThanThreshold"
+  evaluation_periods  = "1"
+  metric_name         = "FreeStorageSpace"
+  namespace           = "AWS/RDS"
+  period              = "600"
+  statistic           = "Average"
+  threshold           = local.thresholds["FreeStorageSpaceThreshold"]
+  alarm_description   = "Average database free storage space over last 10 minutes too low"
+  alarm_actions       = aws_sns_topic.topic.*.arn
+  ok_actions          = aws_sns_topic.topic.*.arn
+
+  dimensions = {
+    DBInstanceIdentifier = var.db_instance_id
+  }
+}

ops/modules/db-alarms/outputs.tf

@@ -0,0 +1,7 @@
+output "free_storage_space_threshold_alarm_id" {
+  value = aws_cloudwatch_metric_alarm.free_storage_space_too_low[0].alarm_name
+}
+
+output "cpu_utilization_threshold_id" {
+  value = aws_cloudwatch_metric_alarm.free_storage_space_too_low[0].alarm_name
+}

ops/modules/db-alarms/variables.tf

@@ -0,0 +1,52 @@
+variable "db_instance_name" {
+  type        = string
+  description = "The name of the database instance"
+}
+
+variable "db_instance_id" {
+  type        = string
+  description = "The id of the database instance"
+}
+
+variable "is_replica" {
+  type        = bool
+  description = "Whether the database instance is a replica"
+}
+
+variable "enable_cpu_utilization_alarm" {
+  type        = bool
+  description = "Whether to enable the CPU utilization alarm"
+}
+
+variable "enable_free_storage_space_too_low_alarm" {
+  type        = bool
+  description = "Whether to enable the free storage space too low alarm"
+}
+
+variable "environment" {
+  type        = string
+  description = "The environment of the database instance"
+}
+
+variable "stage" {
+  type        = string
+  description = "The stage of the database instance"
+}
+
+variable "sns_topic_subscription_emails" {
+  type        = list(string)
+  description = "The emails to subscribe to the SNS topic"
+}
+
+variable "free_storage_space_threshold" {
+  type        = number
+  description = "The free storage space threshold"
+  default     = 5
+}
+
+
+variable "cpu_utilization_threshold" {
+  type        = number
+  description = "The free storage space threshold"
+  default     = 90
+}

ops/testnet/prod/backend/config.tf

@@ -11,7 +11,13 @@ locals {
     DD_LAMBDA_HANDLER   = "packages/agents/cartographer/poller/dist/index.handler"
   }
 
-
+  sdk_server_env_vars = [
+    { name = "NXTP_CONFIG", value = local.local_sdk_server_config },
+    { name = "ENVIRONMENT", value = var.environment },
+    { name = "STAGE", value = var.stage },
+    { name = "DD_PROFILING_ENABLED", value = "true" },
+    { name = "DD_ENV", value = "${var.environment}-${var.stage}" },
+  ]
   postgrest_env_vars = [
     { name = "PGRST_ADMIN_SERVER_PORT", value = "3001" },
     { name = "PGRST_DB_URI", value = "postgres://${var.postgres_user}:${var.postgres_password}@${module.cartographer_db_replica.db_instance_endpoint}/connext" },
@@ -20,7 +26,21 @@ locals {
     { name = "ENVIRONMENT", value = var.environment },
     { name = "STAGE", value = var.stage }
   ]
-
+  local_sdk_server_config = jsonencode({
+    logLevel = "debug"
+    chains = {
+      "1735356532" = {
+        providers = ["https://goerli.optimism.io/"]
+      }
+      "1735353714" = {
+        providers = ["https://goerli.infura.io/v3/9aa3d95b3bc440fa88ea12eaa4456161"]
+      }
+      "9991" = {
+        providers = ["https://rpc.ankr.com/polygon_mumbai"]
+      }
+    }
+    environment = var.stage
+  })
   local_cartographer_config = jsonencode({
     logLevel = "debug"
     chains = {

ops/testnet/prod/backend/main.tf

@@ -22,6 +22,10 @@ data "aws_route53_zone" "primary" {
   zone_id = "Z03634792TWUEHHQ5L0YX"
 }
 
+locals {
+  db_alarm_emails = ["carlo@connext.network", "rahul@connext.network", "preetham@connext.network", "sanchay@connext.network"]
+}
+
 module "cartographer_db" {
   domain                = "cartographer"
   source                = "../../../modules/db"
@@ -52,6 +56,18 @@ module "cartographer_db" {
   publicly_accessible        = true
 }
 
+module "cartographer-db-alarms" {
+  source                                  = "../../../modules/db-alarms"
+  db_instance_name                        = module.cartographer_db.db_instance_name
+  db_instance_id                          = module.cartographer_db.db_instance_id
+  is_replica                              = false
+  enable_cpu_utilization_alarm            = true
+  enable_free_storage_space_too_low_alarm = true
+  stage                                   = var.stage
+  environment                             = var.environment
+  sns_topic_subscription_emails           = local.db_alarm_emails
+}
+
 module "cartographer_db_replica" {
   domain              = "cartographer"
   source              = "../../../modules/db-replica"
@@ -87,6 +103,17 @@ module "cartographer_db_replica" {
   publicly_accessible   = module.cartographer_db.db_publicly_accessible
 }
 
+module "cartographer-db-replica-alarms" {
+  source                                  = "../../../modules/db-alarms"
+  db_instance_name                        = module.cartographer_db.db_instance_name
+  db_instance_id                          = module.cartographer_db.db_instance_id
+  is_replica                              = true
+  enable_cpu_utilization_alarm            = true
+  enable_free_storage_space_too_low_alarm = true
+  stage                                   = var.stage
+  environment                             = var.environment
+  sns_topic_subscription_emails           = local.db_alarm_emails
+}
 
 module "postgrest" {
   source                   = "../../../modules/service"
@@ -117,6 +144,36 @@ module "postgrest" {
   domain                   = var.domain
 }
 
+module "sdk-server" {
+  source                   = "../../../modules/service"
+  region                   = var.region
+  dd_api_key               = var.dd_api_key
+  zone_id                  = data.aws_route53_zone.primary.zone_id
+  execution_role_arn       = data.aws_iam_role.ecr_admin_role.arn
+  cluster_id               = module.ecs.ecs_cluster_id
+  vpc_id                   = module.network.vpc_id
+  private_subnets          = module.network.private_subnets
+  lb_subnets               = module.network.public_subnets
+  internal_lb              = false
+  docker_image             = var.full_image_name_sdk_server
+  container_family         = "sdk-server"
+  container_port           = 8080
+  loadbalancer_port        = 80
+  cpu                      = 256
+  memory                   = 512
+  instance_count           = 2
+  timeout                  = 180
+  environment              = var.environment
+  stage                    = var.stage
+  ingress_cdir_blocks      = ["0.0.0.0/0"]
+  ingress_ipv6_cdir_blocks = []
+  service_security_groups  = flatten([module.network.allow_all_sg, module.network.ecs_task_sg])
+  cert_arn                 = var.certificate_arn_testnet
+  container_env_vars       = local.sdk_server_env_vars
+  domain                   = var.domain
+}
+
+
 module "cartographer-routers-lambda-cron" {
   source              = "../../../modules/lambda"
   ecr_repository_name = "nxtp-cartographer"

ops/testnet/prod/backend/outputs.tf

@@ -30,6 +30,11 @@ output "postgrest-service-endpoint" {
   value = module.postgrest.service_endpoint
 }
 
+output "sdk-server-service-endpoint" {
+  value = module.sdk-server.service_endpoint
+}
+
+
 output "db-instance-endpoint" {
   value = module.cartographer_db.db_instance_endpoint
 }

ops/testnet/prod/backend/variables.tf

@@ -31,6 +31,12 @@ variable "cartographer_image_tag" {
   default     = "latest"
 }
 
+variable "full_image_name_sdk_server" {
+  type        = string
+  description = "router image name"
+  default     = "ghcr.io/connext/sdk-server:sha-e33b4fd"
+}
+
 variable "certificate_arn_testnet" {
   default = "arn:aws:acm:us-west-1:679752396206:certificate/0ebbf095-681a-4a0a-9dc9-fa70cb80166a"
 }