Implement compute domain DRA plugin with state management#153
Merged
Conversation
enoodle
force-pushed
the
erez/compute-domain-dra-plugin-implementation
branch
from
65a5900 to
d9908e0
Compare
enoodle
force-pushed
the
erez/compute-domain-dra-plugin-implementation
branch
2 times, most recently
from
55baafd to
e2fee94
Compare
Add full implementation for the compute domain DRA plugin: - State management with checkpoint persistence for crash recovery - CDI (Container Device Interface) spec generation for device exposure - Health check server for liveness/readiness probes - Device node simulation via symlinks to /dev/null - Driver tests for prepare/unprepare resource claims - OpenShift SCC annotation in helm templates
…lugin - Add computeDomainDevicePluginLabelKey to status-updater node labeling - Add compute-domain-dra-plugin to integration test setup - Enable computeDomainDraPlugin in integration test values
enoodle
force-pushed
the
erez/compute-domain-dra-plugin-implementation
branch
from
e2fee94 to
b8e3022
Compare
gshaibi
approved these changes
Jan 15, 2026
Contributor
gshaibi
left a comment
gshaibi
left a comment
There was a problem hiding this comment.
Amazing PR, WELL DONE!!!
Approved with some minor nit comments
| Path: path.Join(config.flags.kubeletRegistrarDirectoryPath, consts.ComputeDomainDriverName+"-reg.sock"), | ||
| }).String() | ||
| log.Info("connecting to registration socket", "path", regSockPath) | ||
| regConn, err := grpc.NewClient( |
Contributor
There was a problem hiding this comment.
not that relevant, but do we want to close it somewhere?
| if readErr == nil && current == target { | ||
| return nil | ||
| } | ||
| if err := os.Remove(path); err != nil { |
Contributor
There was a problem hiding this comment.
Suggested change
| if err := os.Remove(path); err != nil { | |
| if err = os.Remove(path); err != nil { |
| return err | ||
| } | ||
| } else { | ||
| if err := os.Remove(path); err != nil { |
Contributor
There was a problem hiding this comment.
Suggested change
| if err := os.Remove(path); err != nil { | |
| if err = os.Remove(path); err != nil { |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
This PR adds the full implementation for the compute domain DRA plugin, building on the skeleton from #154
Changes
state.go): Manages prepared claims and domain info with checkpoint persistence for crash recoverycdi.go): Generates Container Device Interface specs for device exposure to containerscheckpoint.go): Saves/restores plugin state across restartsnvcdi_device.go): Creates fake device nodes via symlinks to /dev/nullhealth.go): gRPC health check for liveness/readiness probesdriver.go): Wires up state management, implements Prepare/Unprepare methodsTesting
Dependencies
k8s.io/kubernetesfor checkpoint manager