Skip to content

feat: add configurable pprof bind address#464

Merged
Ronkahn21 merged 6 commits into
ai-dynamo:mainfrom
Ronkahn21:feat/pprof-bind-address
Mar 9, 2026
Merged

feat: add configurable pprof bind address#464
Ronkahn21 merged 6 commits into
ai-dynamo:mainfrom
Ronkahn21:feat/pprof-bind-address

Conversation

@Ronkahn21

@Ronkahn21 Ronkahn21 commented Mar 2, 2026

Copy link
Copy Markdown
Contributor

What type of PR is this?

/kind feature

What this PR does / why we need it:

Adds a configurable pprofBindAddress to the operator config and Helm chart, enabling pprof profiling for scale testing. Also adds a generic annotations field to the Helm chart pod template to support auto-scrape by profiling systems, and a pprof service port with a Helm helper for port extraction.

Which issue(s) this PR fixes:

Fixes #402

Special notes for your reviewer:

This is PR 1 of 3 for the scale testing infrastructure. PR 2 (#465) adds the infra-manager CLI with Pyroscope annotation injection, and a follow-up PR will migrate the e2e workflow to use the new CLI.

Does this PR introduce a API change?

NONE

Additional documentation e.g., enhancement proposals, usage docs, etc.:

NONE

Add pprofBindAddress field to operator config to allow
customizing the pprof server's bind address. Update Helm
charts to expose pprof as a service and add API docs.

Part of ai-dynamo#402

Signed-off-by: Ron Kahn <rkahn@nvidia.com>
@Ronkahn21 Ronkahn21 force-pushed the feat/pprof-bind-address branch from 46b7a4f to 7c0e9db Compare March 2, 2026 10:18

@renormalize renormalize left a comment

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.

Thanks for the PR @Ronkahn21! Just a couple comments~

Comment thread operator/internal/controller/manager.go Outdated
Comment thread operator/charts/templates/deployment.yaml Outdated
Split the single pprofBindAddress string field into separate
pprofBindHost and pprofBindPort fields to properly support IPv6
addresses via net.JoinHostPort. Use ptr.Deref for cleaner nil checks.

Signed-off-by: Ron Kahn <rkahn@nvidia.com>
Update API documentation to reflect the separation of pprofBindAddress into pprofBindHost and pprofBindPort fields, improving support for IPv6 and external scraping configurations.

Signed-off-by: Ron Kahn <rkahn@nvidia.com>
renormalize
renormalize previously approved these changes Mar 6, 2026

@renormalize renormalize left a comment

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.

Added a few nits. We can merge this after they're addressed. Thanks.

Comment thread operator/internal/controller/manager.go Outdated
Comment thread operator/api/config/v1alpha1/defaults.go Outdated
Comment thread operator/api/config/v1alpha1/defaults.go Outdated

@gflarity gflarity left a comment

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.

One small thing, otherwise LGTM.

Comment thread operator/charts/values.yaml Outdated
Signed-off-by: Ron Kahn <rkahn@nvidia.com>
gflarity
gflarity previously approved these changes Mar 6, 2026
Defaulters guarantee PprofBindHost and PprofBindPort are always set,
so remove redundant nil checks and ptr.Deref calls. Make
DefaultPprofBindPort private since it's no longer used externally.

Signed-off-by: Ron Kahn <rkahn@nvidia.com>
shayasoolin
shayasoolin previously approved these changes Mar 8, 2026
Comment thread operator/charts/templates/deployment.yaml Outdated
Comment thread operator/api/config/v1alpha1/types.go
Signed-off-by: Ron Kahn <rkahn@nvidia.com>

@shmuel-runai shmuel-runai left a comment

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.

LGTM

@Ronkahn21 Ronkahn21 merged commit 66e51ee into ai-dynamo:main Mar 9, 2026
19 of 21 checks passed
Ronkahn21 added a commit to Ronkahn21/grove that referenced this pull request Mar 10, 2026
enoodle pushed a commit to enoodle/grove that referenced this pull request Mar 24, 2026
Signed-off-by: Erez Freiberger <enoodle@gmail.com>
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.

GREP: E2E Scale Test Infrastructure

5 participants