[tmpnet] Enure node config is saved on restart for both runtimes #4015
Conversation
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
The addition of the kube node runtime added a `Restart` method to the `NodeRuntime` interface to support runtime-specific restart behavior. This introduced a disparity in restart behavior between the process and kube runtimes. The kube runtime was saving the node configuration before restarting, but the process runtime was not. When starting a network configured with one or more subnets using the process runtime, this resulted in `track-subnets` for the bootstrap node not being serialized to disk. Only the bootstrap node was affected because only it was being restarted after having `track-subnets` set. The remaining nodes were started with `Network.StartNode` - which ensures serialization - after `track-subnets` was set. This broke the xsvm and subnet-evm antithesis testing because both depend on the process runtime to initialize a target network and the bootstrap node would end up without a `track-subnets` value. This resulted in 404 responses when trying to query a subnet's rest endpoint. The fix is to ensure that `Node.Restart` serializes node configuration regardless of runtime.
There was a problem hiding this comment.
Pull Request Overview
This PR unifies restart behavior across runtimes by ensuring node configuration is always persisted before a restart and removes duplicate persistence in the Kubernetes runtime.
- Adds a call to
n.Write()in the genericNode.Restartto save config for all runtimes - Removes the redundant
p.node.Write()inKubeRuntime.Restart - Guarantees that
track-subnetsand other settings are serialized when restarting a process runtime node
Reviewed Changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.
| File | Description |
|---|---|
| tests/fixture/tmpnet/node.go | Added n.Write() call in Node.Restart to persist config prior to restart |
| tests/fixture/tmpnet/kube_runtime.go | Removed redundant p.node.Write() block in KubeRuntime.Restart |
Comments suppressed due to low confidence (1)
tests/fixture/tmpnet/node.go:191
- Add a test to verify that calling Restart on a process runtime actually persists the node configuration and retains
track-subnetsafter restart.
func (n *Node) Restart(ctx context.Context) error {
maru-ava
commented
Jun 15, 2025
| @@ -545,7 +545,7 @@ func (n *Network) Restart(ctx context.Context) error { | |||
| if err := restartNodes(ctx, nodes); err != nil { | |||
| return err | |||
| } | |||
| return WaitForHealthyNodes(ctx, n.log, n.Nodes) | |||
| return WaitForHealthyNodes(ctx, n.log, nodes) | |||
There was a problem hiding this comment.
This was failing when n.Nodes included a stopped bootstrap test node
maru-ava
force-pushed
the
tmpnet-save-on-restart
branch
from
7e2d0be to
80a7c57
Compare
maru-ava
force-pushed
the
tmpnet-save-on-restart
branch
from
80a7c57 to
889c198
Compare
StephenButtolph
approved these changes
Jun 15, 2025
RodrigoVillar
approved these changes
Jun 16, 2025
Elvis339
approved these changes
Jun 16, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Why this should be merged
The addition of the kube node runtime added a
Restartmethod to theNodeRuntimeinterface to support runtime-specific restart behavior. This introduced a disparity in restart behavior between the process and kube runtimes. The kube runtime was saving the node configuration before restarting, but the process runtime was not.When starting a network configured with one or more subnets using the process runtime, this resulted in
track-subnetsfor the bootstrap node not being serialized to disk. Only the bootstrap node was affected because only it was being restarted after havingtrack-subnetsset. The remaining nodes were started withNetwork.StartNode- which ensures serialization - aftertrack-subnetswas set.This broke the xsvm and subnet-evm antithesis testing because both depend on the serialized state resulting from the process runtime initializing a target network, and the serialized bootstrap node config would end up without a
track-subnetsvalue. This resulted in 404 responses when trying to query a subnet's rest endpoint.e2e subnet tests for xsvm were unaffected because those tests only target running nodes for which
track-subnetsis set correctly without regard to whether node configuration is serialized. The issue would only have surfaced in the e2e suite if someone had tried to use the--restart-networkflag (e.g. to restart the nodes to use an updated avalanchego and/or xsvm plugin binary).The fix
Node.Restartserializes node configuration regardless of runtime.Replaces #4013
How this was tested
Need to be documented in RELEASES.md?
N/A