Skip to content

Commit 445baaa

Browse files
agithomasagithomas
authored
Amazon Bedrock agentcore - add alert templates (elastic#16074)
* Amazon Bedrock agentcore - add alert templates * Updated the PR link * Updated stack dependency * Updated version
1 parent 33f764e commit 445baaa

File tree

5 files changed

+94
-3
lines changed

5 files changed

+94
-3
lines changed

‎packages/aws_bedrock_agentcore/changelog.yml‎

Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,9 @@
11
# newer versions go on top
2+
- version: "0.0.3"
3+
changes:
4+
- description: Add alerting rule template for the agent runtime
5+
type: enhancement
6+
link: https://github.com/elastic/integrations/pull/16074
27
- version: "0.0.2"
38
changes:
49
- description: Dashboard improvements for AgentCore runtime, gateway, identity and memory metrics.

‎packages/aws_bedrock_agentcore/kibana/alerting_rule_template/aws-bedrock-agentcore-agent-runtime-high-latency.json‎

Lines changed: 29 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,29 @@
1+
{
2+
"id": "aws-bedrock-agentcore-agent-runtime-high-latency",
3+
"type": "alerting_rule_template",
4+
"attributes": {
5+
"name": "[AWS Bedrock AgentCore] Agent runtime high latency",
6+
"tags": ["AWS", "Amazon Bedrock AgentCore", "Agent runtime"],
7+
"ruleTypeId": ".es-query",
8+
"schedule": {
9+
"interval": "5m"
10+
},
11+
"params": {
12+
"searchType": "esqlQuery",
13+
"timeWindowSize": 5,
14+
"timeWindowUnit": "m",
15+
"esqlQuery": {
16+
"esql": "// Alert triggers when high latency is detected during AWS Bedrock AgentCore runtime invocations.\n//\n// Latency measures the average time elapsed between receiving the request and\n// sending the final response token, measured in milliseconds.\n//\n// The alert is grouped by cloud account, region, and agent endpoint name (aws.dimensions.Name)\n// to pinpoint the specific agent endpoint experiencing high latency.\n//\n// To adjust sensitivity, change the threshold in the WHERE clause (default: 10000ms = 10 seconds).\n\nFROM metrics-aws_bedrock_agentcore.metrics-*\n| WHERE aws.dimensions.Operation == \"InvokeAgentRuntime\"\n| STATS avg_latency_ms = avg(aws.bedrock_agentcore.metrics.Latency.avg) BY cloud.account.id, cloud.region, aws.dimensions.Name\n| WHERE avg_latency_ms > 10000"
17+
},
18+
"groupBy": "row",
19+
"timeField": "event.ingested"
20+
},
21+
"alertDelay": {
22+
"active": 1
23+
}
24+
},
25+
"managed": true,
26+
"coreMigrationVersion": "8.8.0",
27+
"typeMigrationVersion": "10.1.0"
28+
}
29+

‎packages/aws_bedrock_agentcore/kibana/alerting_rule_template/aws-bedrock-agentcore-agent-runtime-system-errors.json‎

Lines changed: 29 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,29 @@
1+
{
2+
"id": "aws-bedrock-agentcore-agent-runtime-system-errors",
3+
"type": "alerting_rule_template",
4+
"attributes": {
5+
"name": "[AWS Bedrock AgentCore] Agent runtime system errors",
6+
"tags": ["AWS", "Amazon Bedrock AgentCore", "Agent runtime"],
7+
"ruleTypeId": ".es-query",
8+
"schedule": {
9+
"interval": "5m"
10+
},
11+
"params": {
12+
"searchType": "esqlQuery",
13+
"timeWindowSize": 5,
14+
"timeWindowUnit": "m",
15+
"esqlQuery": {
16+
"esql": "// Alert triggers when system errors occur during AWS Bedrock AgentCore runtime invocations.\n//\n// System errors correspond to InvocationError.Internal - Internal server error (500).\n// These are server-side errors indicating infrastructure or service issues.\n//\n// The alert is grouped by cloud account, region, and agent endpoint name (aws.dimensions.Name)\n// to pinpoint the specific agent endpoint experiencing issues.\n//\n// To reduce alert noise, increase the threshold (e.g., `total_system_errors > 5`) to only\n// alert on sustained error patterns rather than transient service disruptions.\n\nFROM metrics-aws_bedrock_agentcore.metrics-*\n| WHERE aws.dimensions.Operation == \"InvokeAgentRuntime\"\n| STATS total_system_errors = sum(aws.bedrock_agentcore.metrics.SystemErrors.sum) BY cloud.account.id, cloud.region, aws.dimensions.Name\n| WHERE total_system_errors > 0"
17+
},
18+
"groupBy": "row",
19+
"timeField": "event.ingested"
20+
},
21+
"alertDelay": {
22+
"active": 1
23+
}
24+
},
25+
"managed": true,
26+
"coreMigrationVersion": "8.8.0",
27+
"typeMigrationVersion": "10.1.0"
28+
}
29+

‎packages/aws_bedrock_agentcore/kibana/alerting_rule_template/aws-bedrock-agentcore-agent-runtime-user-errors.json‎

Lines changed: 28 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,28 @@
1+
{
2+
"id": "aws-bedrock-agentcore-agent-runtime-user-errors",
3+
"type": "alerting_rule_template",
4+
"attributes": {
5+
"name": "[AWS Bedrock AgentCore] Agent runtime user errors",
6+
"tags": ["AWS", "Amazon Bedrock AgentCore", "Agent runtime"],
7+
"ruleTypeId": ".es-query",
8+
"schedule": {
9+
"interval": "5m"
10+
},
11+
"params": {
12+
"searchType": "esqlQuery",
13+
"timeWindowSize": 5,
14+
"timeWindowUnit": "m",
15+
"esqlQuery": {
16+
"esql": "// Alert triggers when user errors occur during AWS Bedrock AgentCore runtime invocations.\n//\n// User errors correspond to client-side errors (4xx) including:\n// - InvocationError.Validation - Client provided invalid input (400)\n// - InvocationError.ResourceNotFound - Requested resource doesn't exist (404)\n// - InvocationError.AccessDenied - Client lacks permissions (403)\n// - InvocationError.Conflict - Resource conflict (409)\n//\n// The alert is grouped by cloud account, region, and agent endpoint name (aws.dimensions.Name)\n// to pinpoint the specific agent endpoint experiencing issues.\n//\n// To reduce alert noise, increase the threshold (e.g., `total_user_errors > 10`) to only\n// alert on sustained error patterns rather than isolated incidents.\n\nFROM metrics-aws_bedrock_agentcore.metrics-*\n| WHERE aws.dimensions.Operation == \"InvokeAgentRuntime\"\n| STATS total_user_errors = sum(aws.bedrock_agentcore.metrics.UserErrors.sum) BY cloud.account.id, cloud.region, aws.dimensions.Name\n| WHERE total_user_errors > 0"
17+
},
18+
"groupBy": "row",
19+
"timeField": "event.ingested"
20+
},
21+
"alertDelay": {
22+
"active": 1
23+
}
24+
},
25+
"managed": true,
26+
"coreMigrationVersion": "8.8.0",
27+
"typeMigrationVersion": "10.1.0"
28+
}

‎packages/aws_bedrock_agentcore/manifest.yml‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,7 +1,7 @@
1-
format_version: 3.3.5
1+
format_version: 3.4.0
22
name: aws_bedrock_agentcore
33
title: "Amazon Bedrock AgentCore"
4-
version: 0.0.2
4+
version: 0.0.3
55
source:
66
license: "Elastic-2.0"
77
description: "Collect Amazon Bedrock AgentCore's Agent runtime, Gateway, Identity and Memory metrics using Elastic Agent"
@@ -12,7 +12,7 @@ categories:
1212
- observability
1313
conditions:
1414
kibana:
15-
version: "^8.19.0 || ^9.1.0"
15+
version: "^8.19.0 || ^9.2.1"
1616
elastic:
1717
subscription: "basic"
1818
screenshots:

0 commit comments

Comments
 (0)