datadog-error-tracking

name	datadog-error-tracking
description	Search and manage Datadog Error Tracking issues via API - find grouped errors by service, resolve/ignore issues. Use when investigating production errors or managing error tracking issue states.

Datadog Error Tracking

CLI Commands

# Search issues by service (last 7 days)
dd-cli search-et-issues 'service:my-service-*' --from now-7d

# Search with specific error type
dd-cli search-et-issues 'service:my-service AND @error.type:NullPointerException' --from now-1d

# Search logs-based error tracking (instead of APM traces)
dd-cli search-et-issues 'service:my-service' --track logs

# Sort by most frequent
dd-cli search-et-issues 'service:my-service' --from now-7d --order-by TOTAL_COUNT

# Get a specific issue with details
dd-cli get-et-issue aaaaaaaa-bbbb-cccc-dddd-eeeeeeeeeeee

# Resolve an issue
dd-cli update-et-issue-state aaaaaaaa-... RESOLVED

# Ignore an issue (suppresses monitor notifications)
dd-cli update-et-issue-state aaaaaaaa-... IGNORED

# Reopen a resolved/ignored issue
dd-cli update-et-issue-state aaaaaaaa-... OPEN

search-et-issues Options

Option	Default	Description
QUERY	(required)	Search query (Datadog search syntax)
--from	now-1d	Start time (now-1h, now-7d, or epoch ms)
--to	now	End time
--track	trace	Source: trace (APM), logs, rum
--order-by	-	TOTAL_COUNT, FIRST_SEEN, IMPACTED_SESSIONS, PRIORITY

Issue States

State	Meaning
OPEN	New / for review
RESOLVED	Fixed
IGNORED	Suppressed from monitor alerts

Tip: Ignoring an issue automatically mutes it from Error Tracking monitor notifications.

Error Tracking Monitors

Error Tracking has its own monitor type (error-tracking alert) separate from metric/log monitors. These use a special query DSL:

# New issue monitor (fires on first occurrence or regression)
error-tracking("env:prod service:my-service-*").source("all").new().rollup("count").by("issue.id").last("1d") > 0

# High impact monitor (fires on high error count)
error-tracking("env:prod service:my-service").source("all").impact().rollup("count").by("issue.id").last("5m") > 1

Wildcard tip: Use service:my-prefix-* to automatically cover all current and future deployables without updating the monitor each time a new service is added.

API Details

• Endpoint: POST /api/v2/error-tracking/issues/search
• Get issue: GET /api/v2/error-tracking/issues/{issue_id}
• Update state: PUT /api/v2/error-tracking/issues/{issue_id}/state
• Permission: error_tracking_read / error_tracking_write

Common Patterns

# Triage: find all open issues for a service, sorted by frequency
dd-cli search-et-issues 'service:my-service state:OPEN' --from now-7d --order-by TOTAL_COUNT

# Investigate: get full details on a specific issue
dd-cli get-et-issue <issue-id>

# Bulk resolve: pipe issue IDs through update
dd-cli search-et-issues 'service:my-service' --from now-30d | \
  jq -r '.included[]? | select(.attributes.state == "OPEN") | .id' | \
  xargs -I{} dd-cli update-et-issue-state {} RESOLVED