fix(ci): push latest tag and reduce monitoring costs

- Add latest tag push alongside SHA tag for manual rollback/debugging
- Reduce log retention from 3 to 1 day
- Comment out bot_error_metrics and alerts to save ~$0.47/month
- Minor whitespace fix in cd.yml

.github/workflows/cd.yml

@@ -187,7 +187,7 @@ jobs:
           git fetch origin "${{ github.event.workflow_run.head_branch }}"
           latest_sha=$(git rev-parse "origin/${{ github.event.workflow_run.head_branch }}")
           deploy_sha="${{ steps.images.outputs.deploy_sha }}"
-          
+
           if [[ "$latest_sha" != "$deploy_sha" ]]; then
              echo "::notice::Newer commit ($latest_sha) found on branch. Skipping deployment of $deploy_sha to avoid race conditions."
              echo "skip=true" >> "$GITHUB_OUTPUT"

.github/workflows/ci.yml

@@ -160,6 +160,7 @@ jobs:
           repo="${GCP_REGION}-docker.pkg.dev/${{ vars.GCP_PROJECT_ID }}/${ARTIFACT_REPOSITORY}"
           echo "name=${repo}/${{ matrix.service }}:${GITHUB_SHA}" >> "$GITHUB_OUTPUT"
           echo "cache_ref=${repo}/${{ matrix.service }}:cache" >> "$GITHUB_OUTPUT"
+          echo "latest=${repo}/${{ matrix.service }}:latest" >> "$GITHUB_OUTPUT"
 
       - name: Build and push
         uses: docker/build-push-action@v6
@@ -167,7 +168,9 @@ jobs:
           context: .
           file: apps/${{ matrix.service }}/Dockerfile
           push: true
-          tags: ${{ steps.image.outputs.name }}
+          tags: |
+            ${{ steps.image.outputs.name }}
+            ${{ steps.image.outputs.latest }}
           platforms: linux/amd64
           provenance: false
           cache-from: type=registry,ref=${{ steps.image.outputs.cache_ref }}

infra/terraform/monitoring.tf

@@ -51,7 +51,7 @@ resource "google_logging_project_bucket_config" "default" {
   project        = var.project_id
   location       = "global"
   bucket_id      = "_Default"
-  retention_days = 3
+  retention_days = 1
 }
 
 resource "google_monitoring_notification_channel" "email" {
@@ -68,26 +68,28 @@ resource "google_monitoring_notification_channel" "email" {
   depends_on = [google_project_service.enabled]
 }
 
-resource "google_logging_metric" "bot_error_events" {
+# DEV-187: Commented out to save ~$0.47/month on Cloud Monitoring costs
-  for_each = local.bot_error_metrics
+# TODO: Re-enable if alerting on specific bot error events becomes necessary
-
+# resource "google_logging_metric" "bot_error_events" {
-  project     = var.project_id
+#   for_each = local.bot_error_metrics
-  name        = each.value.metric_name
+#
-  description = "Counts `${each.value.event}` log events for ${module.bot_api_service.name}."
+#   project     = var.project_id
-  filter      = <<-EOT
+#   name        = each.value.metric_name
-resource.type="cloud_run_revision"
+#   description = "Counts `${each.value.event}` log events for ${module.bot_api_service.name}."
-resource.labels.service_name="${module.bot_api_service.name}"
+#   filter      = <<-EOT
-jsonPayload.event="${each.value.event}"
+# resource.type="cloud_run_revision"
-  EOT
+# resource.labels.service_name="${module.bot_api_service.name}"
-
+# jsonPayload.event="${each.value.event}"
-  metric_descriptor {
+#   EOT
-    metric_kind = "DELTA"
+#
-    value_type  = "INT64"
+#   metric_descriptor {
-    unit        = "1"
+#     metric_kind = "DELTA"
-  }
+#     value_type  = "INT64"
-
+#     unit        = "1"
-  depends_on = [google_project_service.enabled]
+#   }
-}
+#
+#   depends_on = [google_project_service.enabled]
+# }
 
 resource "google_monitoring_alert_policy" "bot_api_5xx" {
   project      = var.project_id
@@ -138,50 +140,52 @@ metric.labels.response_code_class="5xx"
   depends_on = [google_project_service.enabled]
 }
 
-resource "google_monitoring_alert_policy" "bot_error_events" {
+# DEV-187: Commented out to save ~$0.47/month on Cloud Monitoring costs
-  for_each = local.bot_error_metrics
+# TODO: Re-enable if alerting on specific bot error events becomes necessary
-
+# resource "google_monitoring_alert_policy" "bot_error_events" {
-  project      = var.project_id
+#   for_each = local.bot_error_metrics
-  display_name = each.value.display_name
+#
-  combiner     = "OR"
+#   project      = var.project_id
-
+#   display_name = each.value.display_name
-  notification_channels = [
+#   combiner     = "OR"
-    for channel in google_monitoring_notification_channel.email : channel.name
+#
-  ]
+#   notification_channels = [
-
+#     for channel in google_monitoring_notification_channel.email : channel.name
-  documentation {
+#   ]
-    content   = "Structured bot failure event `${each.value.event}` was logged by `${module.bot_api_service.name}` in `${var.environment}`."
+#
-    mime_type = "text/markdown"
+#   documentation {
-  }
+#     content   = "Structured bot failure event `${each.value.event}` was logged by `${module.bot_api_service.name}` in `${var.environment}`."
-
+#     mime_type = "text/markdown"
-  conditions {
+#   }
-    display_name = each.value.display_name
+#
-
+#   conditions {
-    condition_threshold {
+#     display_name = each.value.display_name
-      filter = <<-EOT
+#
-resource.type="cloud_run_revision"
+#     condition_threshold {
-resource.labels.service_name="${module.bot_api_service.name}"
+#       filter = <<-EOT
-metric.type="logging.googleapis.com/user/${google_logging_metric.bot_error_events[each.key].name}"
+# resource.type="cloud_run_revision"
-      EOT
+# resource.labels.service_name="${module.bot_api_service.name}"
-
+# metric.type="logging.googleapis.com/user/${google_logging_metric.bot_error_events[each.key].name}"
-      comparison      = "COMPARISON_GT"
+#       EOT
-      threshold_value = 0
+#
-      duration        = "0s"
+#       comparison      = "COMPARISON_GT"
-
+#       threshold_value = 0
-      aggregations {
+#       duration        = "0s"
-        alignment_period   = "300s"
+#
-        per_series_aligner = "ALIGN_RATE"
+#       aggregations {
-      }
+#         alignment_period   = "300s"
-
+#         per_series_aligner = "ALIGN_RATE"
-      trigger {
+#       }
-        count = 1
+#
-      }
+#       trigger {
-    }
+#         count = 1
-  }
+#       }
-
+#     }
-  alert_strategy {
+#   }
-    auto_close = "1800s"
+#
-  }
+#   alert_strategy {
-
+#     auto_close = "1800s"
-  depends_on = [google_logging_metric.bot_error_events]
+#   }
-}
+#
+#   depends_on = [google_logging_metric.bot_error_events]
+# }