Add basic Cloudwatch dashboard for master-slave receipe
Feature: Issue 13218
Change-Id: Ie4dbd25de9f279180b570b0da99c96602d123f6e
diff --git a/Configuration.md b/Configuration.md
index bfd73e1..3e5f89b 100644
--- a/Configuration.md
+++ b/Configuration.md
@@ -59,7 +59,7 @@
plugin.
* `METRICS_CLOUDWATCH_ENABLED`: Optional - Boolean.
-Whether to publish metrics to CloudWatch. Default: false
+Whether to publish metrics to CloudWatch and create CloudWatch dashboard. Default: false
* `METRICS_CLOUDWATCH_NAMESPACE`: Optional - String.
The CloudWatch namespace for Gerrit metrics. Default: _gerrit_
* `METRICS_CLOUDWATCH_RATE`: Optional - String.
diff --git a/master-slave/Makefile b/master-slave/Makefile
index 03e06c7..a32fd37 100644
--- a/master-slave/Makefile
+++ b/master-slave/Makefile
@@ -6,20 +6,26 @@
SERVICE_MASTER_TEMPLATE:=cf-service-master.yml
SERVICE_SLAVE_TEMPLATE:=cf-service-slave.yml
DNS_ROUTING_TEMPLATE:=cf-dns-route.yml
+CLOUDWATCH_DASHBOARD_TEMPLATE:=cf-dashboard.yml
AWS_FC_COMMAND=export AWS_PAGER=;aws cloudformation
.PHONY: create-all delete-all \
- cluster cluster-keys service-master service-slave dns-routing \
- delete-cluster delete-service-master delete-service-slave delete-dns-routing \
- wait-for-cluster-creation wait-for-service-master-creation wait-for-dns-routing-creation \
- wait-for-cluster-deletion wait-for-service-master-deletion wait-for-dns-routing-deletion \
+ cluster cluster-keys service-master service-slave dns-routing dashboard \
+ delete-cluster delete-service-master delete-service-slave delete-dns-routing delete-dashboard \
+ wait-for-cluster-creation wait-for-service-master-creation wait-for-dns-routing-creation wait-for-dashboard \
+ wait-for-cluster-deletion wait-for-service-master-deletion wait-for-dns-routing-deletion wait-for-dashboard-deletion \
gerrit-build gerrit-publish
+ifeq ($(METRICS_CLOUDWATCH_ENABLED),true)
+optional_dashboard_targets=dashboard wait-for-dashboard-creation
+endif
+
create-all: upload-common-templates \
gerrit-publish git-daemon-publish git-ssh-publish \
cluster wait-for-cluster-creation \
service-slave service-master \
wait-for-service-master-creation wait-for-service-slave-creation \
+ $(optional_dashboard_targets) \
dns-routing wait-for-dns-routing-creation
cluster: cluster-keys
@@ -129,6 +135,31 @@
ParameterKey=MasterServiceStackName,ParameterValue=$(SERVICE_MASTER_STACK_NAME) \
ParameterKey=SlaveServiceStackName,ParameterValue=$(SERVICE_SLAVE_STACK_NAME)
+dashboard:
+ifeq ($(METRICS_CLOUDWATCH_ENABLED),true)
+ifdef GERRIT_MASTER_INSTANCE_ID
+ $(eval DASHBOARD_OPTIONAL_PARAMS := $(DASHBOARD_OPTIONAL_PARAMS) ParameterKey=MasterInstanceId,ParameterValue=$(GERRIT_MASTER_INSTANCE_ID))
+endif
+ifdef GERRIT_SLAVE_INSTANCE_ID
+ $(eval DASHBOARD_OPTIONAL_PARAMS := $(DASHBOARD_OPTIONAL_PARAMS) ParameterKey=SlaveInstanceId,ParameterValue=$(GERRIT_SLAVE_INSTANCE_ID))
+endif
+ifdef METRICS_CLOUDWATCH_NAMESPACE
+ $(eval DASHBOARD_OPTIONAL_PARAMS := $(DASHBOARD_OPTIONAL_PARAMS) ParameterKey=MetricsCloudwatchNamespace,ParameterValue=$(METRICS_CLOUDWATCH_NAMESPACE))
+endif
+
+ $(AWS_FC_COMMAND) create-stack \
+ --stack-name $(DASHBOARD_STACK_NAME) \
+ --capabilities CAPABILITY_IAM \
+ --template-body file://`pwd`/$(CLOUDWATCH_DASHBOARD_TEMPLATE) \
+ --region $(AWS_REGION) \
+ --parameters \
+ $(DASHBOARD_OPTIONAL_PARAMS)
+
+else
+ @echo "METRICS_CLOUDWATCH_ENABLED is set to false. Dashboard creation skipped".
+endif
+
+
wait-for-cluster-creation:
@echo "*** Wait for cluster stack '$(CLUSTER_STACK_NAME)' creation"
$(AWS_FC_COMMAND) wait stack-create-complete \
@@ -157,6 +188,13 @@
--region $(AWS_REGION)
@echo "*** DNS Routing stack '$(DNS_ROUTING_STACK_NAME)' created"
+wait-for-dashboard-creation:
+ @echo "*** Wait for dashboard stack '$(DASHBOARD_STACK_NAME)' creation"
+ $(AWS_FC_COMMAND) wait stack-create-complete \
+ --stack-name $(DASHBOARD_STACK_NAME) \
+ --region $(AWS_REGION)
+ @echo "*** Dashboard stack '$(DASHBOARD_STACK_NAME)' created"
+
wait-for-cluster-deletion:
@echo "*** Wait for cluster stack '$(CLUSTER_STACK_NAME)' deletion"
$(AWS_FC_COMMAND) wait stack-delete-complete \
@@ -185,6 +223,13 @@
--region $(AWS_REGION)
@echo "*** DNS routing stack '$(DNS_ROUTING_STACK_NAME)' deleted"
+wait-for-dashboard-deletion:
+ @echo "*** Wait for dashboard stack '$(DASHBOARD_STACK_NAME)' deletion"
+ $(AWS_FC_COMMAND) wait stack-delete-complete \
+ --stack-name $(DASHBOARD_STACK_NAME) \
+ --region $(AWS_REGION)
+ @echo "*** Dashboard stack '$(DASHBOARD_STACK_NAME)' deleted"
+
delete-cluster:
$(AWS_FC_COMMAND) delete-stack \
--stack-name $(CLUSTER_STACK_NAME) \
@@ -205,10 +250,16 @@
--stack-name $(DNS_ROUTING_STACK_NAME) \
--region $(AWS_REGION)
+delete-dashboard:
+ $(AWS_FC_COMMAND) delete-stack \
+ --stack-name $(DASHBOARD_STACK_NAME) \
+ --region $(AWS_REGION)
+
delete-all: delete-dns-routing wait-for-dns-routing-deletion \
delete-service-slave wait-for-service-slave-deletion \
delete-service-master wait-for-service-master-deletion \
- delete-cluster wait-for-cluster-deletion
+ delete-cluster wait-for-cluster-deletion \
+ delete-dashboard wait-for-dashboard-deletion
gerrit-publish:
$(MAKE) -C ../gerrit gerrit-publish RECIPE=master-slave
diff --git a/master-slave/README.md b/master-slave/README.md
index 4338ac8..2581e05 100644
--- a/master-slave/README.md
+++ b/master-slave/README.md
@@ -5,11 +5,12 @@
## Architecture
-Four templates are provided in this example:
+Five templates are provided in this example:
* `cf-cluster`: define the ECS cluster and the networking stack
* `cf-service-master`: define the service stack running Gerrit master
* `cf-service-slave`: define the service stack running Gerrit slave
* `cf-dns-route`: define the DNS routing for the service
+* `cf-dashboard`: define the CloudWatch dashboard for the services
### Networking
@@ -72,6 +73,7 @@
* `SERVICE_MASTER_STACK_NAME`: Optional. Name of the master service stack. `gerrit-service-master` by default.
* `SERVICE_SLAVE_STACK_NAME`: Optional. Name of the slave service stack. `gerrit-service-slave` by default.
+* `DASHBOARD_STACK_NAME` : Optional. Name of the dashboard stack. `gerrit-dashboard` by default.
* `MASTER_SUBDOMAIN`: Optional. Name of the master sub domain. `gerrit-master-demo` by default.
* `SLAVE_SUBDOMAIN`: Optional. Name of the slave sub domain. `gerrit-slave-demo` by default.
* `CLUSTER_DESIRED_CAPACITY`: Optional. Number of EC2 instances composing the cluster. `1` by default.
diff --git a/master-slave/cf-dashboard.yml b/master-slave/cf-dashboard.yml
new file mode 100644
index 0000000..66120ea
--- /dev/null
+++ b/master-slave/cf-dashboard.yml
@@ -0,0 +1,392 @@
+AWSTemplateFormatVersion: '2010-09-09'
+Description: Deploy a Master-Slave CloudWatch dashboard.
+Parameters:
+ MasterInstanceId:
+ Description: Optional identifier for the Gerrit master instance
+ Default: gerrit-master-slave-MASTER
+ Type: String
+ SlaveInstanceId:
+ Description: Optional identifier for the Gerrit slave instance
+ Default: gerrit-master-slave-SLAVE
+ Type: String
+ MetricsCloudwatchNamespace:
+ Description: The CloudWatch namespace for Gerrit metrics
+ Type: String
+ Default: gerrit
+Resources:
+ CloudWatchDashboard:
+ Type: AWS::CloudWatch::Dashboard
+ Properties:
+ DashboardName: !Ref AWS::StackName
+ DashboardBody: !Sub |
+ {
+ "start": "-PT6H",
+ "periodOverride": "auto",
+ "widgets": [
+ {
+ "type": "metric",
+ "x": 0,
+ "y": 0,
+ "width": 24,
+ "height": 3,
+ "properties": {
+ "metrics": [
+ [ "${MetricsCloudwatchNamespace}", "sshd/sessions/connected", "InstanceId", "${MasterInstanceId}", "Type", "gauge" ],
+ [ "${MetricsCloudwatchNamespace}", "sshd/sessions/connected", "InstanceId", "${SlaveInstanceId}", "Type", "gauge" ]
+ ],
+ "view": "singleValue",
+ "region": "${AWS::Region}",
+ "stat": "Maximum",
+ "period": 60,
+ "title": "SSH Connection",
+ "stacked": false
+ }
+ },
+ {
+ "type": "metric",
+ "x": 6,
+ "y": 3,
+ "width": 6,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ { "expression": "RATE(METRICS())", "label": "Expression1", "id": "e1", "region": "${AWS::Region}" } ],
+ [ "${MetricsCloudwatchNamespace}", "proc/cpu/usage", "InstanceId", "${MasterInstanceId}", "Type", "gauge", { "id": "m1", "visible": false } ],
+ [ "${MetricsCloudwatchNamespace}", "proc/cpu/usage", "InstanceId", "${SlaveInstanceId}", "Type", "gauge", { "id": "m2", "visible": false } ]
+ ],
+ "view": "timeSeries",
+ "stacked": false,
+ "region": "${AWS::Region}",
+ "stat": "Average",
+ "period": 60,
+ "title": "Gerrit Cpu Load",
+ "yAxis": {
+ "right": {
+ "showUnits": true
+ },
+ "left": {
+ "showUnits": false,
+ "min": 0
+ }
+ }
+ }
+ },
+ {
+ "type": "metric",
+ "x": 12,
+ "y": 33,
+ "width": 12,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ "${MetricsCloudwatchNamespace}", "jgit/block_cache/cache_used", "InstanceId", "${MasterInstanceId}", "Type", "gauge", { "id": "m1" } ],
+ [ "${MetricsCloudwatchNamespace}", "jgit/block_cache/cache_used", "InstanceId", "${SlaveInstanceId}", "Type", "gauge", { "id": "m2" } ]
+ ],
+ "view": "timeSeries",
+ "stacked": false,
+ "region": "${AWS::Region}",
+ "yAxis": {
+ "left": {
+ "showUnits": false,
+ "label": ""
+ }
+ },
+ "title": "JGit Cache",
+ "period": 60,
+ "stat": "Average"
+ }
+ },
+ {
+ "type": "metric",
+ "x": 18,
+ "y": 3,
+ "width": 6,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ "${MetricsCloudwatchNamespace}", "jvm.memory.total.used", "InstanceId", "${MasterInstanceId}", "Type", "gauge" ],
+ [ "${MetricsCloudwatchNamespace}", "jvm.memory.total.used", "InstanceId", "${SlaveInstanceId}", "Type", "gauge" ]
+ ],
+ "view": "timeSeries",
+ "stacked": false,
+ "region": "${AWS::Region}",
+ "yAxis": {
+ "left": {
+ "showUnits": false
+ }
+ },
+ "stat": "Average",
+ "period": 60,
+ "title": "Gerrit Used memory"
+ }
+ },
+ {
+ "type": "metric",
+ "x": 12,
+ "y": 3,
+ "width": 6,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ "${MetricsCloudwatchNamespace}", "proc/cpu/system_load", "InstanceId", "${MasterInstanceId}", "Type", "gauge" ],
+ [ "${MetricsCloudwatchNamespace}", "proc/cpu/system_load", "InstanceId", "${SlaveInstanceId}", "Type", "gauge" ]
+ ],
+ "view": "timeSeries",
+ "stacked": false,
+ "title": "Gerrit System Load",
+ "region": "${AWS::Region}",
+ "yAxis": {
+ "left": {
+ "showUnits": false
+ }
+ },
+ "stat": "Average",
+ "period": 60
+ }
+ },
+ {
+ "type": "metric",
+ "x": 0,
+ "y": 9,
+ "width": 6,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ "${MetricsCloudwatchNamespace}", "jvm.gc.G1-Old-Generation.time", "InstanceId", "${MasterInstanceId}", "Type", "gauge" ],
+ [ "${MetricsCloudwatchNamespace}", "jvm.gc.G1-Old-Generation.time", "InstanceId", "${SlaveInstanceId}", "Type", "gauge" ]
+ ],
+ "view": "timeSeries",
+ "stacked": false,
+ "title": "GC Time",
+ "region": "${AWS::Region}",
+ "stat": "Average",
+ "period": 60,
+ "yAxis": {
+ "left": {
+ "showUnits": false
+ }
+ }
+ }
+ },
+ {
+ "type": "metric",
+ "x": 6,
+ "y": 9,
+ "width": 6,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ "${MetricsCloudwatchNamespace}", "proc/jvm/thread/num_live", "InstanceId", "${MasterInstanceId}", "Type", "gauge" ],
+ [ "${MetricsCloudwatchNamespace}", "proc/jvm/thread/num_live", "InstanceId", "${SlaveInstanceId}", "Type", "gauge" ]
+ ],
+ "view": "timeSeries",
+ "stacked": false,
+ "title": "Active Threads",
+ "region": "${AWS::Region}",
+ "period": 60,
+ "stat": "Average",
+ "yAxis": {
+ "left": {
+ "showUnits": false
+ }
+ }
+ }
+ },
+ {
+ "type": "metric",
+ "x": 0,
+ "y": 15,
+ "width": 6,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ "${MetricsCloudwatchNamespace}", "http/server/rest_api/server_latency_total", "InstanceId", "${MasterInstanceId}", "Type", "99.9%", { "id": "m2" } ]
+ ],
+ "view": "timeSeries",
+ "stacked": true,
+ "region": "${AWS::Region}",
+ "stat": "Average",
+ "period": 60,
+ "title": "HTTP requests latency",
+ "yAxis": {
+ "left": {
+ "showUnits": false,
+ "label": "ms"
+ }
+ }
+ }
+ },
+ {
+ "type": "metric",
+ "x": 6,
+ "y": 15,
+ "width": 6,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ { "expression": "m1+m2", "label": "HTTP hits per second", "id": "e2", "region": "${AWS::Region}" } ],
+ [ "${MetricsCloudwatchNamespace}", "http/server/success_count_total", "InstanceId", "${MasterInstanceId}", "Type", "count", { "id": "m2", "visible": false } ],
+ [ ".", "http/server/error_count_total", ".", ".", ".", ".", { "id": "m1", "visible": false } ]
+ ],
+ "view": "timeSeries",
+ "stacked": true,
+ "title": "HTTP hits per second",
+ "region": "${AWS::Region}",
+ "stat": "Sum",
+ "period": 60,
+ "yAxis": {
+ "left": {
+ "showUnits": false,
+ "label": "ops",
+ "min": 0
+ }
+ }
+ }
+ },
+ {
+ "type": "metric",
+ "x": 12,
+ "y": 15,
+ "width": 6,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ { "expression": "(m1/(m1+m2))*100", "label": "% of HTTP Errors", "id": "e1", "region": "${AWS::Region}" } ],
+ [ "${MetricsCloudwatchNamespace}", "http/server/error_count_total", "InstanceId", "${MasterInstanceId}", "Type", "count", { "id": "m1", "visible": false } ],
+ [ ".", "http/server/success_count_total", ".", ".", ".", ".", { "id": "m2", "visible": false } ]
+ ],
+ "view": "timeSeries",
+ "stacked": true,
+ "region": "${AWS::Region}",
+ "stat": "Sum",
+ "period": 60,
+ "title": "% of HTTP Errors",
+ "yAxis": {
+ "left": {
+ "showUnits": false,
+ "max": 100,
+ "min": 0
+ }
+ }
+ }
+ },
+ {
+ "type": "metric",
+ "x": 12,
+ "y": 9,
+ "width": 6,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ "${MetricsCloudwatchNamespace}", "git/upload-pack/request_count_total", "InstanceId", "${MasterInstanceId}", "Type", "count", { "id": "m1" } ],
+ [ "${MetricsCloudwatchNamespace}", "git/upload-pack/request_count_total", "InstanceId", "${SlaveInstanceId}", "Type", "count", { "id": "m1" } ]
+ ],
+ "view": "timeSeries",
+ "stacked": false,
+ "region": "${AWS::Region}",
+ "title": "Git upload pack - count",
+ "stat": "Sum",
+ "period": 60,
+ "yAxis": {
+ "left": {
+ "showUnits": false,
+ "min": 0
+ }
+ }
+ }
+ },
+ {
+ "type": "metric",
+ "x": 0,
+ "y": 21,
+ "width": 24,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ { "expression": "RATE(METRICS())*PERIOD(m1)", "label": "Expression1", "id": "e1" } ],
+ [ "${MetricsCloudwatchNamespace}", "queue/index_batch/total_scheduled_tasks_count", "InstanceId", "${MasterInstanceId}", "Type", "gauge", { "id": "m1", "visible": false } ],
+ [ ".", "queue/receive_commits/total_scheduled_tasks_count", ".", ".", ".", ".", { "id": "m2", "visible": false } ],
+ [ ".", "queue/work_queue/total_scheduled_tasks_count", ".", ".", ".", ".", { "id": "m3", "visible": false } ],
+ [ ".", "queue/ssh_command_start/total_scheduled_tasks_count", ".", ".", ".", ".", { "id": "m4", "visible": false } ]
+ ],
+ "view": "timeSeries",
+ "stacked": true,
+ "title": "Scheduled Tasks Queues",
+ "region": "${AWS::Region}",
+ "stat": "Sum",
+ "period": 60,
+ "yAxis": {
+ "left": {
+ "showUnits": false,
+ "min": 0
+ }
+ }
+ }
+ },
+ {
+ "type": "metric",
+ "x": 0,
+ "y": 27,
+ "width": 24,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ "${MetricsCloudwatchNamespace}", "queue/send_email/scheduled_tasks", "InstanceId", "${MasterInstanceId}", "Type", "gauge" ]
+ ],
+ "view": "timeSeries",
+ "stacked": true,
+ "region": "${AWS::Region}",
+ "title": "Scheduled email tasks in the queue",
+ "stat": "Sum",
+ "period": 300,
+ "yAxis": {
+ "left": {
+ "showUnits": false
+ }
+ }
+ }
+ },
+ {
+ "type": "metric",
+ "x": 0,
+ "y": 33,
+ "width": 12,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ "${MetricsCloudwatchNamespace}", "jgit/block_cache/open_files", "InstanceId", "${MasterInstanceId}", "Type", "gauge" ],
+ [ "${MetricsCloudwatchNamespace}", "jgit/block_cache/open_files", "InstanceId", "${SlaveInstanceId}", "Type", "gauge" ]
+ ],
+ "view": "timeSeries",
+ "stacked": false,
+ "title": "Pack files cached",
+ "region": "${AWS::Region}",
+ "stat": "Sum",
+ "period": 60,
+ "yAxis": {
+ "left": {
+ "showUnits": false
+ }
+ }
+ }
+ },
+ {
+ "type": "metric",
+ "x": 0,
+ "y": 3,
+ "width": 6,
+ "height": 6,
+ "properties": {
+ "metrics": [
+ [ "AWS/EC2", "CPUUtilization" ]
+ ],
+ "view": "timeSeries",
+ "stacked": false,
+ "region": "${AWS::Region}",
+ "title": "Cluster CPU Load",
+ "period": 60,
+ "stat": "Average"
+ }
+ }
+ ]
+ }
diff --git a/master-slave/setup.env.template b/master-slave/setup.env.template
index 4a295ff..98389ae 100644
--- a/master-slave/setup.env.template
+++ b/master-slave/setup.env.template
@@ -6,6 +6,7 @@
SERVICE_GRAFANA_STACK_NAME:=$(AWS_PREFIX)-grafana
DNS_ROUTING_STACK_NAME:=$(AWS_PREFIX)-dns-routing
DNS_ROUTING_MONITORING_STACK_NAME:=$(AWS_PREFIX)-monitoring-dns-routing
+DASHBOARD_STACK_NAME:=$(AWS_PREFIX)-dashboard
HOSTED_ZONE_NAME:=yourcompany.com
MASTER_SUBDOMAIN:=$(AWS_PREFIX)-master.gerrit-demo
SLAVE_SUBDOMAIN:=$(AWS_PREFIX)-slave.gerrit-demo