You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Deprecate PromQL endpoint, promote OpenMetrics to GA (#4337)
- Add deprecation notices to PromQL-related pages (reference, general-setup,
prometheus-grafana, promql) with April 2 2026 deprecation and October 5 2026
disable dates
- Rewrite cloud/metrics landing page: title to "Temporal Cloud Metrics",
lead with OpenMetrics, move PromQL to deprecated section at bottom
- Remove Public Preview designations from all OpenMetrics pages
- Update migration guide timeline with concrete deprecation dates
- Change OpenMetrics data lookback from 2 minutes to 3 minutes
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Brian MacDonald <brian.macdonald@temporal.io>
Co-authored-by: Nish Krishnan <nishkrishnan@users.noreply.github.com>
description: Get detailed insights using the Temporal SDK and Cloud Metrics.
5
+
description: Monitor Temporal Cloud workloads with Cloud metrics and SDK metrics.
6
6
slug: /cloud/metrics/
7
7
toc_max_heading_level: 4
8
8
keywords:
@@ -21,29 +21,42 @@ tags:
21
21
- Temporal Cloud
22
22
---
23
23
24
-
Temporal offers two distinct sources of metrics: [Cloud/Server Metrics](/cloud/metrics/reference) and [SDK Metrics](/references/sdk-metrics).
25
-
Each source provides options for levels of granularity and filtering, monitoring-tool integrations, and configuration.
26
-
Before implementing Temporal Cloud observability, decide what you need to measure for your use case. There are two primary use cases for metrics:
24
+
Temporal offers two distinct sources of metrics: [Cloud Metrics](/cloud/metrics/openmetrics/metrics-reference) and [SDK Metrics](/references/sdk-metrics).
25
+
Each source provides different levels of granularity, filtering options, monitoring-tool integrations, and configuration.
27
26
28
-
-To measure the health and performance of Temporal-backed applications and key business processes.
29
-
-To measure the health and performance of Temporal infrastructure and user provided infrastructure in the form of Temporal Workers and Temporal Clients.
27
+
-**SDK metrics** monitor individual Workers and your code's behavior from the perspective of your application.
28
+
-**Cloud metrics** monitor Temporal Cloud's behavior from the perspective of the Temporal Service.
30
29
31
-
When measuring the performance of Temporal-backed applications and key business processes, you should rely on Temporal SDK metrics as a source of truth.
32
-
This is because Temporal SDKs provide visibility from the perspective of your application, not from the perspective of the Temporal Service.
30
+
When used together, Cloud and SDK metrics measure the health and performance of your full Temporal infrastructure, including the Temporal Cloud Service and user-supplied Temporal Workers.
33
31
34
-
SDK metrics monitor individual workers and your code's behavior.
35
-
Cloud metrics monitor Temporal behavior.
36
-
When used together, Temporal Cloud and SDK metrics measure the health and performance of your full Temporal infrastructure, including the Temporal Cloud Service and user-supplied Temporal Workers.
32
+
## Cloud Metrics
37
33
38
-
Cloud Metrics for all Namespaces in your account are available from two sources:
34
+
Cloud metrics for all Namespaces in your account are available from the [OpenMetrics endpoint](/cloud/metrics/openmetrics), a Prometheus-compatible scrapable endpoint at `metrics.temporal.io`.
39
35
40
-
-[OpenMetrics Endpoint](/cloud/metrics/openmetrics) - A Prometheus-compatible scrapable endpoint.
41
-
-[PromQL Endpoint](/cloud/metrics/promql) - A Prometheus query endpoint.
36
+
-[OpenMetrics overview](/cloud/metrics/openmetrics) - Getting started and key concepts
37
+
-[Metrics integrations](/cloud/metrics/openmetrics/metrics-integrations) - Datadog, Grafana Cloud, New Relic, ClickStack, and more
38
+
-[API reference](/cloud/metrics/openmetrics/api-reference) - Endpoint specification and advanced configuration
39
+
-[Metrics reference](/cloud/metrics/openmetrics/metrics-reference) - Complete catalog of all `temporal_cloud_v1_*` metrics
42
40
43
-
:::note
41
+
## SDK Metrics
44
42
45
-
OpenMetrics is the recommended option for most users.
43
+
SDK metrics are emitted by your Workers and Clients.
44
+
For setup instructions, see [SDK metrics setup](/cloud/metrics/sdk-metrics-setup).
Copy file name to clipboardExpand all lines: docs/cloud/metrics/openmetrics/api-reference.mdx
+2-7Lines changed: 2 additions & 7 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -25,11 +25,6 @@ import { CaptionedImage } from '@site/src/components';
25
25
26
26
The Temporal Cloud OpenMetrics API provides actionable operational metrics about your Temporal Cloud deployment. This is a scrapable HTTP API that returns metrics in OpenMetrics format, suitable for ingestion by Prometheus-compatible monitoring systems.
27
27
28
-
:::tip SUPPORT, STABILITY, and DEPENDENCY INFO
29
-
30
-
Temporal Cloud OpenMetrics support is available in [Public Preview](/evaluate/development-production-features/release-stages#public-preview).
31
-
32
-
:::
33
28
34
29
## Available Metrics Reference
35
30
@@ -134,7 +129,7 @@ Implement retry logic in your client to gracefully handle transient API failures
134
129
135
130
### Data Latency
136
131
137
-
Metric data points are available for query within 2 minutes of their origination. This is in line with the freshest metrics [available from any major service provider](https://docs.datadoghq.com/integrations/guide/cloud-metric-delay/). This latency should be accounted for when setting up monitoring alerts.
132
+
Metric data points are available for query within 3 minutes of their origination. This is in line with the freshest metrics [available from any major service provider](https://docs.datadoghq.com/integrations/guide/cloud-metric-delay/). This latency should be accounted for when setting up monitoring alerts.
138
133
139
134
## Endpoints
140
135
@@ -152,7 +147,7 @@ Returns metrics in OpenMetrics format suitable for scraping by Prometheus-compat
152
147
153
148
#### Timestamp Offset
154
149
155
-
To account for metric data latency, this endpoint returns metrics from the current timestamp minus a fixed offset. The current offset is 2 minutes rounded down to the start of the minute. To accommodate this offset, the timestamps in the response should be honored when importing the metrics. For example, in Prometheus this can be controlled using the `honor\_timestamps` flag.
150
+
To account for metric data latency, this endpoint returns metrics from the current timestamp minus a fixed offset. The current offset is 3 minutes rounded down to the start of the minute. To accommodate this offset, the timestamps in the response should be honored when importing the metrics. For example, in Prometheus this can be controlled using the `honor\_timestamps` flag.
*[Migration Guide](/cloud/metrics/openmetrics/migration-guide) - Guide on how to transition from the Prometheus query endpoint
44
39
45
40
## Overview
46
-
Temporal Cloud OpenMetrics exposes 50+ metrics covering workflow lifecycles, task queue operations, service performance, and system limits. All metrics are aggregated over one-minute windows and available for scraping within two minutes.
41
+
Temporal Cloud OpenMetrics exposes 50+ metrics covering workflow lifecycles, task queue operations, service performance, and system limits. All metrics are aggregated over one-minute windows and available for scraping within three minutes.
47
42
48
43
*[Set up authentication and scraping](/cloud/metrics/openmetrics/api-reference#authentication) with the API documentation.
49
44
* Browse the [complete metrics catalog](/cloud/metrics/openmetrics/metrics-reference) for descriptions and labels.
The PromQL endpoint and its `temporal_cloud_v0_*` metrics were deprecated on April 2, 2026 and are no longer accepting new users.
31
+
The PromQL endpoint will be disabled for all users on **October 5, 2026**.
32
+
33
+
For Grafana setup with the OpenMetrics endpoint, see the [OpenMetrics integrations page](/cloud/metrics/openmetrics/metrics-integrations).
34
+
35
+
:::
36
+
28
37
**How to set up Grafana with Temporal Cloud PromQL endpoint to view Cloud metrics.**
29
38
30
39
Temporal Cloud emits metrics through a [Prometheus HTTP API endpoint](https://prometheus.io/docs/prometheus/latest/querying/api/), which can be directly used as a Prometheus data source in Grafana or to query and export Cloud metrics to any observability platform.
Copy file name to clipboardExpand all lines: docs/cloud/metrics/reference.mdx
+12-3Lines changed: 12 additions & 3 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -26,10 +26,19 @@ keywords:
26
26
- gRPC error count
27
27
---
28
28
29
+
:::danger PromQL endpoint deprecated
30
+
31
+
The PromQL endpoint and its `temporal_cloud_v0_*` metrics were deprecated on April 2, 2026 and are no longer accepting new users.
32
+
The PromQL endpoint will be disabled for all users on **October 5, 2026**.
33
+
34
+
Migrate to the [OpenMetrics endpoint](/cloud/metrics/openmetrics) which provides `temporal_cloud_v1_*` metrics with higher cardinality, accurate percentiles, and simplified authentication.
35
+
See the [migration guide](/cloud/metrics/openmetrics/migration-guide) for a complete v0-to-v1 metric mapping.
36
+
37
+
:::
38
+
29
39
A metric is a measurement or data point that provides insights into the performance and health of a system.
30
-
This document describes the metrics available on the Temporal Cloud platform.
31
-
Temporal Cloud metrics help you monitor performance and troubleshoot errors.
32
-
They provide insights into different aspects of the Service.
40
+
This document describes the `temporal_cloud_v0_*` metrics available from the deprecated Temporal Cloud PromQL endpoint.
41
+
For the current metrics reference, see the [OpenMetrics metrics reference](/cloud/metrics/openmetrics/metrics-reference).
0 commit comments