github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #8247] Enhanced System Observability for Multi-Server Environments (Unified Endpoints?) #51779

New Issue
Closed
opened 2026-04-28 20:56:03 -05:00 by GiteaMirror · 5 comments
GiteaMirror commented 2026-04-28 20:56:03 -05:00
Owner
Copy Link

Originally created by @dezoito on GitHub (Dec 26, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/8247

As Ollama adoption grows, the lack of comprehensive system metrics makes it challenging to meet standard operational requirements - monitoring, alerting, and planning across development, staging, and production environments.

This can also prevent a wider adoption in commercial and production applications.

While the current endpoints (/api/version, /api/tags, /api/ps) provide basic information, consolidating and expanding these into a single observability endpoint would significantly improve monitoring and management capabilities,

Proposed: New /api/info endpoint returning unified system metrics, like:

{
  "version": "0.1.16",
  "system": {
    "cpu": {
      "cores": 16,
      "threads": 32,
      "usage_percent": 45.2,
      ...
    },
    "memory": {
      "total_bytes": 34359738368,
      "used_bytes": 28859738368,
      ...
    },
    "gpus": [
      {
        "name": "NVIDIA GeForce RTX 4090",
        "memory": {
          "total_bytes": 25769803776,
          "used_bytes": 16106127360,
          ...
        },
      }
    ]
  },
  "models": {
    "loaded": [...],
    "available": [...],
  },
}

This enhancement would:

  • Enable proper monitoring and alerting in production environments
  • Simplify capacity planning across deployments
  • Allow integration with standard observability tools

To the community:

  • What other metrics or stats would be useful to have?
  • How do you currently monitor or observe your running Ollama instances?
Originally created by @dezoito on GitHub (Dec 26, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/8247 As Ollama adoption grows, the lack of comprehensive system metrics makes it challenging to meet standard operational requirements - monitoring, alerting, and planning across development, staging, and production environments. This can also prevent a wider adoption in commercial and production applications. While the current endpoints (`/api/version`, `/api/tags`, `/api/ps`) provide basic information, consolidating and expanding these into a single observability endpoint would significantly improve monitoring and management capabilities, Proposed: New `/api/info` endpoint returning unified system metrics, like: ```json { "version": "0.1.16", "system": { "cpu": { "cores": 16, "threads": 32, "usage_percent": 45.2, ... }, "memory": { "total_bytes": 34359738368, "used_bytes": 28859738368, ... }, "gpus": [ { "name": "NVIDIA GeForce RTX 4090", "memory": { "total_bytes": 25769803776, "used_bytes": 16106127360, ... }, } ] }, "models": { "loaded": [...], "available": [...], }, } ``` This enhancement would: - Enable proper monitoring and alerting in production environments - Simplify capacity planning across deployments - Allow integration with standard observability tools To the community: - What other metrics or stats would be useful to have? - How do you currently monitor or observe your running Ollama instances?
GiteaMirror added the feature request label 2026-04-28 20:56:03 -05:00
GiteaMirror commented 2026-04-28 20:56:05 -05:00
Author
Owner
Copy Link

@maglore9900 commented on GitHub (Dec 26, 2024):

Agree. This would be very helpful for enterprise type deployments, but also helpful for power users, and potentially troubleshooting.

<!-- gh-comment-id:2562816396 --> @maglore9900 commented on GitHub (Dec 26, 2024): Agree. This would be very helpful for enterprise type deployments, but also helpful for power users, and potentially troubleshooting.
GiteaMirror commented 2026-04-28 20:56:05 -05:00
Author
Owner
Copy Link

@kth8 commented on GitHub (Dec 26, 2024):

Prometheus and Grafana is widely used in commercial and production environments and the standard endpoint for that is /metrics. In order for Prometheus to scrape that endpoint, the format should be like this instead of JSON.

<!-- gh-comment-id:2562818401 --> @kth8 commented on GitHub (Dec 26, 2024): Prometheus and Grafana is widely used in commercial and production environments and the standard endpoint for that is `/metrics`. In order for Prometheus to scrape that endpoint, the format should be like [this](https://demo.promlabs.com/metrics) instead of JSON.
GiteaMirror commented 2026-04-28 20:56:05 -05:00
Author
Owner
Copy Link

@dezoito commented on GitHub (Dec 26, 2024):

In order for Prometheus to scrape that endpoint, the format should be like this instead of json.

Thank you for the feedback!

That looks way more detailed, of course.

I still think a simple JSON response could still add a ton of value (and would be easier to implement by combining the existing endpoints).

Someone cue in the "why not both" meme, please...

<!-- gh-comment-id:2562831601 --> @dezoito commented on GitHub (Dec 26, 2024): > In order for Prometheus to scrape that endpoint, the format should be like [this](https://demo.promlabs.com/metrics) instead of json. Thank you for the feedback! That looks way more detailed, of course. I still think a simple JSON response could still add a ton of value (and would be easier to implement by combining the existing endpoints). Someone cue in the "why not both" meme, please...
GiteaMirror commented 2026-04-28 20:56:05 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Dec 26, 2024):

https://github.com/ollama/ollama/issues/3144

<!-- gh-comment-id:2562875176 --> @rick-github commented on GitHub (Dec 26, 2024): https://github.com/ollama/ollama/issues/3144
GiteaMirror commented 2026-04-28 20:56:06 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jan 13, 2025):

closing as dupe #3144

<!-- gh-comment-id:2586036739 --> @rick-github commented on GitHub (Jan 13, 2025): closing as dupe #3144
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label feature request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#51779
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?