fix: send SIGHUP to MCP container after initial deploy by tsivaprasad · Pull Request #378 · pgEdge/control-plane

tsivaprasad · 2026-05-05T12:25:29Z

Summary

This PR ensures that once the MCP container is confirmed running during initial provisioning, a SIGHUP signal is sent to trigger a config reload.

Changes

service_instance.go — call spec.PostDeploy() after WaitForService()
service_instance_spec.go — add PostDeploy() with a service-type switch
mcp_config_resource.go — add public PostDeploy() that calls the existing signalConfigReload()

Testing

Verification:

Created DB and MCP service using below config

curl -s -X POST http://localhost:3000/v1/databases \
  -H "Content-Type: application/json" -d '{
  "id": "test-mcp-embed",
  "spec": {
    "database_name": "test_mcp_embed",
    "database_users": [
      {
        "username": "admin",
        "password": "password",
        "db_owner": true,
        "attributes": ["LOGIN", "SUPERUSER"]
      }
    ],
    "port": 0,
    "nodes": [
      { "name": "n1", "host_ids": ["host-1"] }
    ],
    "services": [
      {
        "service_id": "mcp-server",
        "service_type": "mcp",
        "version": "latest",
        "host_ids": ["host-1"],
        "port": 0,
        "connect_as": "admin",
        "config": {
          "embedding_provider": "openai",
          "embedding_model": "text-embedding-3-small",
          "embedding_api_key": "sk-proj-1"
        }
      }
    ]
  }
}'

Confirmed from config.yaml that embedding.enabled: true (Enabled)

embedding:
  enabled: true
  provider: openai
  model: text-embedding-3-small

From logs confirmed that config reloaded:

Database configured: admin@:0/test_mcp_embed (per-session connections)
Received SIGHUP, reloading configuration...
Updated database configurations: 1 database(s)
Configuration reloaded successfully from /app/data/config.yaml

Checklist

Tests no change and no existing tests failure

PLAT-588

coderabbitai · 2026-05-05T12:25:42Z

Warning

Rate limit exceeded

@tsivaprasad has exceeded the limit for the number of commits that can be reviewed per hour. Please wait 40 minutes and 54 seconds before requesting another review.

To keep reviews running without waiting, you can enable usage-based add-on for your organization. This allows additional reviews beyond the hourly cap. Account admins can enable it under billing.

⌛ How to resolve this issue?

After the wait time has elapsed, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

We recommend that you space out your commits to avoid hitting the rate limit.

🚦 How do rate limits work?

CodeRabbit enforces hourly rate limits for each developer per organization.

Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout.

Please see our FAQ for further information.

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: 53978953-17f9-4923-8917-c2030f433d12

📥 Commits

Reviewing files that changed from the base of the PR and between 2f9be52 and 44c8871.

📒 Files selected for processing (2)

server/internal/orchestrator/swarm/mcp_config_resource.go
server/internal/orchestrator/swarm/service_instance_spec.go

📝 Walkthrough

Walkthrough

Added a post-deployment hook mechanism that allows service resources to execute custom logic after their containers are confirmed running. The MCP service uses this hook to trigger a best-effort configuration reload via SIGHUP after deployment completes.

Changes

Post-Deployment Hook System

Layer / File(s)	Summary
Hook Interface `server/internal/orchestrator/swarm/service_instance_spec.go`	`ServiceInstanceSpecResource.PostDeploy` method defines the post-deployment hook, dispatching to service-type-specific implementations (currently MCP).
MCP Implementation `server/internal/orchestrator/swarm/mcp_config_resource.go`	`MCPConfigResource.PostDeploy` invokes `signalConfigReload` as a best-effort operation, discarding errors to prevent deployment failure.
Deployment Integration `server/internal/orchestrator/swarm/service_instance.go`	`ServiceInstanceResource.deploy` calls `spec.PostDeploy` after the Docker service confirms tasks have started but before setting the instance state to Running in etcd.

Poem

🐰 A hook springs forth, so clever and neat,
When containers wake, their config to greet!
SIGHUP whispers gently through the MCP air,
Best-effort reloads handled with care.
Deployment flows on, no failures to fear!

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely summarizes the main change: sending SIGHUP to the MCP container after initial deployment to enable config reload.
Description check	✅ Passed	The PR description is well-structured following the template with Summary, Changes, Testing sections, and includes a checked Checklist. However, the Changelog entry requirement from the template checklist was not addressed.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests
Commit unit tests in branch PLAT-588-mcp-server-starts-without-embedding-config-on-initial-provisioning

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

codacy-production · 2026-05-05T12:27:04Z

Up to standards ✅

🟢 Issues 0 issues

Results:
0 new issues

View in Codacy

🟢 Metrics 0 duplication

Metric Results

Duplication 0

View in Codacy

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer}
_{TIP This summary will be updated as you push new changes.}

coderabbitai

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@server/internal/orchestrator/swarm/mcp_config_resource.go`:
- Around line 260-262: In PostDeploy, do not ignore the return from
r.signalConfigReload(ctx, rc); instead check the error and emit a structured
zerolog error log including at least error (err), operation ("PostDeploy" or
"signalConfigReload"), and relevant resource/context identifiers (e.g., rc or
resource name); use the existing logger (e.g., r.logger or rc.Logger) and
include domain error mapping to HTTP codes where appropriate (via Goa) when
handling the error downstream so reload failures are visible in logs and can be
translated to proper status codes.

In `@server/internal/orchestrator/swarm/service_instance_spec.go`:
- Around line 98-100: The current code silently ignores errors from
resource.FromContext; change it to log failures using structured zerolog and
retain the successful path: call resource.FromContext[*MCPConfigResource](rc,
MCPConfigResourceIdentifier(s.ServiceInstanceID)); if err != nil, emit a
structured error log (include fields like "service_instance_id"
s.ServiceInstanceID, "resource" "MCPConfigResource", and the error value) using
the existing zerolog logger available in the request context or rc, and only
call mcpCfg.PostDeploy(ctx, rc) when err == nil; keep the same identifiers
(resource.FromContext, MCPConfigResource, MCPConfigResourceIdentifier,
PostDeploy, mcpCfg) so reviewers can locate the change.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

Push a commit to this branch (recommended)
Create a new PR with the fixes

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro

Run ID: f2e8d827-e5f3-41fd-ab8f-50136e4a876d

📥 Commits

Reviewing files that changed from the base of the PR and between ea89e93 and 2f9be52.

📒 Files selected for processing (3)

server/internal/orchestrator/swarm/mcp_config_resource.go
server/internal/orchestrator/swarm/service_instance.go
server/internal/orchestrator/swarm/service_instance_spec.go

rshoemaker

LGTM.

One minor comment: cluster update will cause a double SIGHUP on the container ( once in MCPConfigResource.Update() and again in ServiceInstanceResource.deploy() ), while create will only cause a single SIGHUP. The double signal is benign though - not worth changing IMO.

tsivaprasad · 2026-05-05T15:58:46Z

Will address complete solution as part PLAT-589

fix: send SIGHUP to MCP container after initial deploy

2f9be52

coderabbitai Bot reviewed May 5, 2026

View reviewed changes

Comment thread server/internal/orchestrator/swarm/mcp_config_resource.go

Comment thread server/internal/orchestrator/swarm/service_instance_spec.go Outdated

addressing AI review comments

44c8871

rshoemaker approved these changes May 5, 2026

View reviewed changes

tsivaprasad merged commit 21f23cb into main May 5, 2026
3 checks passed

tsivaprasad deleted the PLAT-588-mcp-server-starts-without-embedding-config-on-initial-provisioning branch May 5, 2026 15:59

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix: send SIGHUP to MCP container after initial deploy#378

fix: send SIGHUP to MCP container after initial deploy#378
tsivaprasad merged 2 commits into
mainfrom
PLAT-588-mcp-server-starts-without-embedding-config-on-initial-provisioning

tsivaprasad commented May 5, 2026 •

edited by atlassian Bot

Loading

Uh oh!

coderabbitai Bot commented May 5, 2026 •

edited

Loading

Rate limit exceeded

Walkthrough

Changes

Poem

Uh oh!

codacy-production Bot commented May 5, 2026 •

edited

Loading

Uh oh!

coderabbitai Bot left a comment

Uh oh!

Uh oh!

Uh oh!

rshoemaker left a comment

Uh oh!

tsivaprasad commented May 5, 2026 •

edited by atlassian Bot

Loading

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

tsivaprasad commented May 5, 2026 • edited by atlassian Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Changes

Testing

Checklist

Uh oh!

coderabbitai Bot commented May 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Rate limit exceeded

Walkthrough

Changes

Poem

Uh oh!

codacy-production Bot commented May 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Up to standards ✅

Uh oh!

coderabbitai Bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

rshoemaker left a comment

Choose a reason for hiding this comment

Uh oh!

tsivaprasad commented May 5, 2026 • edited by atlassian Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

tsivaprasad commented May 5, 2026 •

edited by atlassian Bot

Loading

coderabbitai Bot commented May 5, 2026 •

edited

Loading

codacy-production Bot commented May 5, 2026 •

edited

Loading

tsivaprasad commented May 5, 2026 •

edited by atlassian Bot

Loading