Skip to content

add manifests and documentation for observability#57

Merged
cooktheryan merged 5 commits intoogx-ai:mainfrom
sallyom:observability-hub
Apr 10, 2025
Merged

add manifests and documentation for observability#57
cooktheryan merged 5 commits intoogx-ai:mainfrom
sallyom:observability-hub

Conversation

@sallyom
Copy link
Copy Markdown
Collaborator

@sallyom sallyom commented Apr 4, 2025

Description

Provides documentation for setting up a minimal observability stack on OpenShift for collecting LLamastack and vLLM telemetry

How Has This Been Tested?

Follow the kubernetes/observability/README.md to set up the observability stack.

sallyom added 2 commits April 7, 2025 13:17
Signed-off-by: sallyom <somalley@redhat.com>
Signed-off-by: sallyom <somalley@redhat.com>
@sallyom sallyom force-pushed the observability-hub branch 2 times, most recently from c2e3a79 to 2b11b5c Compare April 7, 2025 19:41
Signed-off-by: sallyom <somalley@redhat.com>
@sallyom sallyom force-pushed the observability-hub branch from 2b11b5c to 5e45f76 Compare April 7, 2025 20:40
editable: true
type: tempo
# This is specific to "observability-hub" namespace. If running tempostack elsewhere, need to update
url: "https://tempo-tempostack-gateway-observability-hub.apps.ocp-beta-test.nerc.mghpcc.org/api/traces/v1/dev/tempo"
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@sallyom can we flip this to a svc address?

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

hopefully, yes - I think I had an issue with that - I'll ask Pavol when we meet today what's up


#### metrics

For vLLM, metrics are generated by default and are exposed at `vllm-endpoint:port/metrics`. For a list of metrics,
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@sallyom can we follow up on that one jira ticket to see if there is an open data hub vllm image we could use that includes your fix?

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

the metrics are available with the opendatahub image but not tracing - I'll check the jira - if there are no plans on including the tracing pkgs, I'll remove the vllm-tracing docs from here.


#### metrics

For vLLM, metrics are generated by default and are exposed at `vllm-endpoint:port/metrics`. For a list of metrics,
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@sallyom can we follow up on that one jira ticket to see if there is an open data hub vllm image we could use that includes your fix?

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

this shouldn't pause the merging but something we maybe make a note of to patch later

sallyom added 2 commits April 10, 2025 13:18
Signed-off-by: sallyom <somalley@redhat.com>
Signed-off-by: sallyom <somalley@redhat.com>
@sallyom sallyom force-pushed the observability-hub branch from d0c8469 to 4cca9b0 Compare April 10, 2025 17:39
@cooktheryan cooktheryan merged commit 2bc9b22 into ogx-ai:main Apr 10, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants