whekin/household-bot
1
0
Fork 0
mirror of https://github.com/whekin/household-bot.git synced 2026-03-31 08:54:04 +00:00
Code Issues Packages Projects Releases Wiki Activity

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

Browse Source 
- 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
This commit is contained in:
whekin
2026-03-16 05:17:47 +04:00
parent 224cdc2d45
commit 3d15754936
3 changed files with 77 additions and 70 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
.github/workflows/cd.yml vendored
View File

@@ -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"

5
.github/workflows/ci.yml vendored
View File

@@ -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 }}

140
infra/terraform/monitoring.tf
View File

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