Skip to content

Commit

Permalink
Apply suggestions from code review
Browse files Browse the repository at this point in the history
Co-authored-by: Austin Lai <76412946+alai97@users.noreply.github.com>
  • Loading branch information
dkirov-dd and alai97 committed Sep 24, 2024
1 parent f9e154c commit e377c37
Show file tree
Hide file tree
Showing 3 changed files with 5 additions and 5 deletions.
4 changes: 2 additions & 2 deletions anyscale/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -6,11 +6,11 @@ This check monitors [Anyscale][1]. Anyscale is a compute platform that hosts [Ra

## Setup

To install and configure the Datadog Agent with the Ray integration on Anyscale, follow [this documentation][6].
To install and configure the Datadog Agent with the Ray integration on Anyscale, see the [official Anyscale documentation][6].

## Data Collected

See the [Data Collected by Ray][4].
See the [Ray integration's Data Collected section][4].

## Troubleshooting

Expand Down
4 changes: 2 additions & 2 deletions anyscale/assets/monitors/failed_task.json
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,11 @@
"version": 2,
"created_at": "2023-12-13",
"last_updated_at": "2023-12-13",
"title": "High Number of Failed Tasks on anyscale Node",
"title": "High Number of Failed Tasks on Anyscale Node",
"description": "Tasks are scheduled to workers by the Anyscale scheduler. This monitor alerts when there are too many scheduling failures within a specified time period due to a particular reason.",
"definition": {
"id": 136348417,
"name": "High Number of Failed Tasks on anyscale Node",
"name": "High Number of Failed Tasks on Anyscale Node",
"type": "query alert",
"query": "avg(last_5m):avg:ray.scheduler.failed_worker_startup{*} by {reason,nodeaddress} > 25",
"message": "{{#is_alert}}\nThere are {{value}} tasks that failed to be schedule because workers were not available. This occurred on node: {{nodeaddress.name}}. The stated reason was {{reason.name}}.\n{{/is_alert}}",
Expand Down
2 changes: 1 addition & 1 deletion anyscale/assets/monitors/gpu_utilization.json
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
"version": 2,
"created_at": "2023-12-13",
"last_updated_at": "2023-12-13",
"title": "Low GPU Utilization low on Anyscale Node",
"title": "GPU Utilization low on Anyscale Node",
"description": "Ray can leverage the computing power of GPUs to perform machine learning or data processing tasks. Underutilizing available GPU can be expensive. This monitor alerts when the GPU utilization is low on a Anyscale node indicating overprovisioning.",
"definition": {
"id": 136350024,
Expand Down

0 comments on commit e377c37

Please sign in to comment.