muster check

Check if resources are available and properly configured.

Synopsis

muster check [RESOURCE_TYPE] [NAME] [OPTIONS]

Description

The check command verifies if resources are available and properly configured in Muster. It validates that all dependencies are met and that resources can be used effectively.

Prerequisites: The aggregator server must be running (muster serve) before using this command.

Resource Types

Resource Type	Description	Example
`mcpserver`	Check MCP server status and connectivity	`muster check mcpserver kubernetes`
`workflow`	Check if a workflow is available (all required tools present)	`muster check workflow deploy-app`

Options

Output Control

--output, -o (string): Output format (table|json|yaml)
- Default: table
--quiet, -q: Suppress non-essential output
- Default: false

Configuration

--config-path (string): Custom configuration directory path
- Default: ~/.config/muster

Examples

Checking MCP Server Status

# Check MCP server health
muster check mcpserver kubernetes

# Example output:
# NAME         STATUS    RESPONSIVE   TOOLS   LAST_CHECK
# kubernetes   Healthy   Yes          15      30s ago

# If server has issues:
# NAME         STATUS    RESPONSIVE   TOOLS   LAST_CHECK
# kubernetes   Error     No           0       2m ago

Checking Workflow Availability

# Check if workflow can be executed
muster check workflow deploy-application

# Example output:
# NAME                 STATUS      TOOLS_AVAILABLE   DEPENDENCIES
# deploy-application   Available   5/5               All met

# If dependencies are missing:
# NAME                 STATUS        TOOLS_AVAILABLE   DEPENDENCIES
# deploy-application   Unavailable   4/5               x_helm_install missing

Check Results

MCP Server Health

When checking an MCP server, the command verifies:

Process Status: Server process is running
Connectivity: Server responds to health checks
Tool Inventory: All expected tools are exposed
Performance: Response times are acceptable

muster check mcpserver prometheus
# Checks:
# ✓ Process is running (PID: 1234)
# ✓ Server responds to ping (250ms)
# ✓ Tools are available (8 tools)
# ✓ Recent activity detected
# → Status: Healthy

Workflow Dependencies

When checking a workflow, the command verifies:

Step Validation: All workflow steps are defined
Tool Requirements: Required tools are available
Parameter Schema: Workflow parameters are valid
Execution Environment: Environment is ready

muster check workflow backup-database
# Checks:
# ✓ Workflow definition is valid (3 steps)
# ✓ Tool 'x_database_backup' is available
# ✓ Tool 'x_storage_upload' is available
# ✓ Required parameters are defined
# → Status: Available

Output Formats

Table Format (Default)

Human-readable status with clear indicators:

muster check workflow deploy-app
# NAME         STATUS      TOOLS   ISSUES
# deploy-app   Available   5/5     None

JSON Format

Detailed status information for programmatic use:

muster check workflow deploy-app --output json
# {
#   "name": "deploy-app",
#   "type": "workflow",
#   "status": "Available",
#   "checks": {
#     "toolsAvailable": {
#       "required": 5,
#       "available": 5,
#       "missing": []
#     },
#     "mcpServers": [
#       {
#         "name": "kubernetes",
#         "status": "Healthy",
#         "tools": ["x_kubernetes_apply", "x_kubernetes_get_status"]
#       }
#     ],
#     "issues": []
#   },
#   "lastChecked": "2024-01-07T10:00:00Z"
# }

YAML Format

Configuration-friendly output:

muster check workflow deploy-app --output yaml
# name: deploy-app
# type: workflow
# status: Available
# checks:
#   stepsValid: true
#   toolsRequired: 5
#   toolsAvailable: 5
#   missingTools: []
#   parametersValid: true
# dependencies:
#   - name: kubernetes
#     status: Available
#   - name: helm
#     status: Available
# lastChecked: "2024-01-07T10:00:00Z"

Status Indicators

Availability Status

Resources can have these availability statuses:

Available: Resource is ready to use
Unavailable: Resource has missing dependencies
Degraded: Resource is partially available
Unknown: Resource status cannot be determined
Error: Resource has configuration errors

Health Status (MCP Servers)

MCP servers have specific health indicators:

Healthy: Server is running and responsive
Unhealthy: Server is running but not responding correctly
Error: Server has errors or is not running
Starting: Server is in startup phase
Unknown: Server status cannot be determined

Use Cases

Pre-Deployment Validation

# Validate environment before deployment
echo "Validating deployment environment..."

# Check required workflows
muster check workflow deploy-application
muster check workflow rollback-deployment

# Check MCP servers
muster check mcpserver kubernetes
muster check mcpserver helm

echo "Environment validation complete"

Health Monitoring

# Regular health check script
echo "Performing system health check..."

CRITICAL_SERVERS=("kubernetes" "prometheus" "github")
CRITICAL_WORKFLOWS=("deploy-app" "backup-db" "scale-service")

echo "Checking critical MCP servers..."
for server in "${CRITICAL_SERVERS[@]}"; do
  STATUS=$(muster check mcpserver "$server" --output json | jq -r '.status')
  echo "  $server: $STATUS"

  if [ "$STATUS" != "Healthy" ]; then
    echo "    WARNING: $server is not healthy!"
  fi
done

echo "Checking critical workflows..."
for workflow in "${CRITICAL_WORKFLOWS[@]}"; do
  STATUS=$(muster check workflow "$workflow" --output json | jq -r '.status')
  echo "  $workflow: $STATUS"

  if [ "$STATUS" != "Available" ]; then
    echo "    WARNING: $workflow is not available!"
  fi
done

CI/CD Integration

#!/bin/bash
# Pre-deployment environment check for CI/CD

REQUIRED_WORKFLOWS=("deploy-app" "rollback" "health-check")

echo "Running pre-deployment checks..."

ALL_CHECKS_PASSED=true

# Check Workflows
for wf in "${REQUIRED_WORKFLOWS[@]}"; do
  echo "Checking Workflow: $wf"
  if ! muster check workflow "$wf" --quiet; then
    echo "FAIL: Workflow $wf is not available"
    ALL_CHECKS_PASSED=false
  else
    echo "PASS: Workflow $wf is available"
  fi
done

if [ "$ALL_CHECKS_PASSED" = true ]; then
  echo "All checks passed - deployment can proceed"
  exit 0
else
  echo "Some checks failed - deployment blocked"
  exit 1
fi

Error Handling

Resource Not Found

muster check workflow non-existent
# Error: workflow 'non-existent' not found

# Solution: List available resources
muster list workflow

Missing Dependencies

muster check workflow deploy-app
# NAME         STATUS        TOOLS   ISSUES
# deploy-app   Unavailable   4/5     x_kubernetes_scale missing

# Solution: Check MCP server providing missing tool
muster check mcpserver kubernetes
muster list mcpserver  # Find which server should provide the tool

MCP Server Issues

muster check mcpserver kubernetes
# NAME         STATUS   RESPONSIVE   TOOLS   ISSUES
# kubernetes   Error    No           0       Connection timeout

# Solution: Check server status and restart if needed
muster get mcpserver kubernetes
# Restart the MCP server if necessary

Connectivity Issues

muster check workflow deploy-app
# Error: failed to connect to aggregator

# Solution: Ensure aggregator is running
muster serve  # In another terminal

Exit Codes

Code	Meaning
0	Resource is available and healthy
1	General error or invalid arguments
2	Resource not found
3	Resource is unavailable or unhealthy
4	Partial availability (degraded)
5	Connection error (server not running)

Auto-Completion

The check command supports tab completion:

# Resource types
muster check [TAB]
# Suggestions: mcpserver, workflow

# Resource names (context-aware)
muster check mcpserver [TAB]
# Suggestions: kubernetes, prometheus, github, ...

Performance Considerations

Batch Checking

# Check multiple resources efficiently
RESOURCES=("deploy-app" "rollback" "health-check")

echo "Batch checking workflows..."
for resource in "${RESOURCES[@]}"; do
  muster check workflow "$resource" --quiet || echo "ISSUE: $resource"
done

Caching Results

# Cache check results for scripting
CHECK_CACHE="/tmp/muster_checks.json"

# Perform checks and cache results
{
  echo "{"
  echo "  \"workflows\": {"
  for wf in $(muster list workflow --output json | jq -r '.workflows[].name'); do
    STATUS=$(muster check workflow "$wf" --output json | jq -r '.status')
    echo "    \"$wf\": \"$STATUS\","
  done | sed '$ s/,$//'
  echo "  },"
  echo "  \"timestamp\": \"$(date -Iseconds)\""
  echo "}"
} > "$CHECK_CACHE"

# Use cached results
cat "$CHECK_CACHE" | jq '.workflows'

Related Commands

list - List resources to find what to check
get - Get detailed information after check fails
create - Create resources after verifying dependencies
start - Start resources after confirming availability

Advanced Usage

Deep Health Check

#!/bin/bash
# Comprehensive system health check

echo "Performing deep health check..."

# Check all MCP servers
echo "=== MCP Server Health ==="
muster list mcpserver --output json | jq -r '.mcpServers[].name' | while read server; do
  echo "Checking $server..."
  muster check mcpserver "$server" --output json | \
    jq -r 'if .status == "Healthy" then "✓ \(.name): OK" else "✗ \(.name): \(.status)" end'
done

# Check all workflows
echo "=== Workflow Readiness ==="
muster list workflow --output json | jq -r '.workflows[].name' | while read wf; do
  echo "Checking $wf..."
  muster check workflow "$wf" --output json | \
    jq -r 'if .status == "Available" then "✓ \(.name): OK" else "✗ \(.name): \(.status)" end'
done

echo "Health check complete"

Uh oh!

FilesExpand file tree

check.md

Latest commit

History