- Archive: 85 agent .txt files → docs/archive/agents/legacy_txt/ - Scripts: Move 110 shell scripts → scripts/ (keep deploy.sh in root) - Models: Move 18 .safetensors → ml/models/checkpoints/training_artifacts/ - Delete: 34 directories (~33GB freed) - target/, coverage_*, test artifacts - Build: Clean 14 build artifacts (.rlib, .o, .pid, binaries) - Tests: Move 14 .rs files → tests/standalone/ - SQL: Move 5 files → sql/ (keep init-db*.sql for Docker) - Wave 153: Archive to docs/archive/historical/wave153/ - Docs: Archive 9 markdown files to wave_d/reports/ and historical/ Total impact: ~34GB freed (both waves), root directory cleaned from 583 to ~40 essential files Directory count reduced from 65 to 31 (52% reduction) All historical data preserved in organized archive structure
361 lines
12 KiB
Bash
Executable File
361 lines
12 KiB
Bash
Executable File
#!/bin/bash
|
|
|
|
# Graceful Degradation Test Suite
|
|
# Tests system behavior under various failure conditions
|
|
|
|
set -e
|
|
|
|
echo "╔════════════════════════════════════════════════════════════════╗"
|
|
echo "║ GRACEFUL DEGRADATION TEST SUITE ║"
|
|
echo "║ Testing System Resilience Under Failures ║"
|
|
echo "╚════════════════════════════════════════════════════════════════╝"
|
|
echo ""
|
|
|
|
# Colors for output
|
|
RED='\033[0;31m'
|
|
GREEN='\033[0;32m'
|
|
YELLOW='\033[1;33m'
|
|
BLUE='\033[0;34m'
|
|
NC='\033[0m' # No Color
|
|
|
|
# Test results tracking
|
|
TESTS_PASSED=0
|
|
TESTS_FAILED=0
|
|
TESTS_TOTAL=0
|
|
|
|
# Helper functions
|
|
log_test() {
|
|
echo -e "${BLUE}[TEST]${NC} $1"
|
|
TESTS_TOTAL=$((TESTS_TOTAL + 1))
|
|
}
|
|
|
|
log_pass() {
|
|
echo -e "${GREEN}[PASS]${NC} $1"
|
|
TESTS_PASSED=$((TESTS_PASSED + 1))
|
|
}
|
|
|
|
log_fail() {
|
|
echo -e "${RED}[FAIL]${NC} $1"
|
|
TESTS_FAILED=$((TESTS_FAILED + 1))
|
|
}
|
|
|
|
log_info() {
|
|
echo -e "${YELLOW}[INFO]${NC} $1"
|
|
}
|
|
|
|
# Function to check Docker container health
|
|
check_container_health() {
|
|
local container=$1
|
|
local status=$(docker inspect --format='{{.State.Health.Status}}' $container 2>/dev/null)
|
|
if [ "$status" = "healthy" ]; then
|
|
return 0
|
|
else
|
|
return 1
|
|
fi
|
|
}
|
|
|
|
# Function to check service responds (not auth failure)
|
|
check_service_responsive() {
|
|
local port=$1
|
|
if nc -zv localhost $port 2>&1 | grep -q "succeeded"; then
|
|
return 0
|
|
else
|
|
return 1
|
|
fi
|
|
}
|
|
|
|
# Capture baseline metrics
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "BASELINE: Capturing Normal Operation Metrics"
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
|
|
log_test "Baseline: All Docker containers healthy"
|
|
BASELINE_HEALTHY=true
|
|
|
|
containers=("foxhunt-api-gateway" "foxhunt-trading-service" "foxhunt-backtesting-service" "foxhunt-ml-training-service" "foxhunt-postgres" "foxhunt-redis")
|
|
for container in "${containers[@]}"; do
|
|
if check_container_health "$container"; then
|
|
log_pass "$container is healthy"
|
|
else
|
|
log_fail "$container is unhealthy"
|
|
BASELINE_HEALTHY=false
|
|
fi
|
|
done
|
|
|
|
if [ "$BASELINE_HEALTHY" = false ]; then
|
|
echo ""
|
|
echo -e "${RED}ERROR: Baseline health check failed. Cannot proceed with degradation tests.${NC}"
|
|
exit 1
|
|
fi
|
|
|
|
log_test "Baseline: Services respond to connections"
|
|
# Check TCP connectivity (proves services are listening)
|
|
services=("api_gateway:50051" "trading_service:50052" "backtesting_service:50053" "ml_training_service:50054")
|
|
for svc in "${services[@]}"; do
|
|
name="${svc%%:*}"
|
|
port="${svc##*:}"
|
|
if check_service_responsive "$port"; then
|
|
log_pass "$name responds on port $port"
|
|
else
|
|
log_fail "$name not responding on port $port"
|
|
BASELINE_HEALTHY=false
|
|
fi
|
|
done
|
|
|
|
log_test "Baseline: Prometheus monitoring operational"
|
|
if curl -sf http://localhost:9090/-/healthy > /dev/null 2>&1; then
|
|
log_pass "Prometheus is healthy"
|
|
else
|
|
log_fail "Prometheus is unhealthy"
|
|
fi
|
|
|
|
echo ""
|
|
|
|
# TEST 1: Redis Failure
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "TEST 1: Redis Failure Degradation"
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "Expected: Services continue with degraded caching, no catastrophic failure"
|
|
echo ""
|
|
|
|
log_test "Stopping Redis container"
|
|
docker stop foxhunt-redis > /dev/null 2>&1
|
|
sleep 3
|
|
|
|
log_test "Checking services continue to operate without Redis"
|
|
|
|
# API Gateway should continue (rate limiting degraded)
|
|
if check_container_health "foxhunt-api-gateway"; then
|
|
log_pass "API Gateway operational without Redis (degraded rate limiting)"
|
|
else
|
|
if docker ps | grep -q foxhunt-api-gateway; then
|
|
log_pass "API Gateway running without Redis (may show unhealthy but not crashed)"
|
|
else
|
|
log_fail "API Gateway crashed when Redis failed"
|
|
fi
|
|
fi
|
|
|
|
# Trading Service should continue (caching degraded)
|
|
if check_service_responsive "50052"; then
|
|
log_pass "Trading Service responds without Redis (degraded caching)"
|
|
else
|
|
log_fail "Trading Service not responding without Redis"
|
|
fi
|
|
|
|
# Check service logs for graceful degradation messages
|
|
log_info "Checking for degradation warnings in logs..."
|
|
if docker logs foxhunt-api-gateway --tail 20 2>&1 | grep -qi "redis"; then
|
|
log_pass "API Gateway logs Redis connection issues (expected)"
|
|
fi
|
|
|
|
log_test "Restoring Redis"
|
|
docker start foxhunt-redis > /dev/null 2>&1
|
|
sleep 5
|
|
|
|
# Wait for Redis to be healthy
|
|
log_info "Waiting for Redis recovery..."
|
|
for i in {1..10}; do
|
|
if docker exec foxhunt-redis redis-cli ping 2>/dev/null | grep -q PONG; then
|
|
log_pass "Redis recovered successfully"
|
|
break
|
|
fi
|
|
sleep 1
|
|
done
|
|
|
|
# Check services recover
|
|
sleep 3
|
|
if check_container_health "foxhunt-api-gateway"; then
|
|
log_pass "API Gateway recovered after Redis restore"
|
|
else
|
|
log_fail "API Gateway did not fully recover"
|
|
fi
|
|
|
|
echo ""
|
|
|
|
# TEST 2: ML Service Down
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "TEST 2: ML Service Down Degradation"
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "Expected: Trading continues without ML predictions"
|
|
echo ""
|
|
|
|
log_test "Stopping ML Training Service"
|
|
docker stop foxhunt-ml-training-service > /dev/null 2>&1
|
|
sleep 2
|
|
|
|
log_test "Verifying trading continues without ML service"
|
|
|
|
# API Gateway should continue (ML methods will fail but gateway operational)
|
|
if check_service_responsive "50051"; then
|
|
log_pass "API Gateway operational without ML service"
|
|
else
|
|
log_fail "API Gateway affected by ML service failure"
|
|
fi
|
|
|
|
# Trading Service should continue (no ML dependency)
|
|
if check_service_responsive "50052"; then
|
|
log_pass "Trading Service operational without ML predictions"
|
|
else
|
|
log_fail "Trading Service depends on ML service (CRITICAL)"
|
|
fi
|
|
|
|
# Backtesting should continue
|
|
if check_service_responsive "50053"; then
|
|
log_pass "Backtesting Service operational without ML"
|
|
else
|
|
log_fail "Backtesting Service affected by ML failure"
|
|
fi
|
|
|
|
log_test "Restoring ML Training Service"
|
|
docker start foxhunt-ml-training-service > /dev/null 2>&1
|
|
sleep 10
|
|
|
|
log_info "Waiting for ML service recovery..."
|
|
for i in {1..15}; do
|
|
if check_container_health "foxhunt-ml-training-service"; then
|
|
log_pass "ML Training Service recovered successfully"
|
|
break
|
|
fi
|
|
sleep 1
|
|
done
|
|
|
|
echo ""
|
|
|
|
# TEST 3: Backtesting Service Failure
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "TEST 3: Backtesting Service Failure"
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "Expected: Core trading and API Gateway unaffected"
|
|
echo ""
|
|
|
|
log_test "Stopping Backtesting Service"
|
|
docker stop foxhunt-backtesting-service > /dev/null 2>&1
|
|
sleep 2
|
|
|
|
log_test "Verifying core services continue"
|
|
|
|
if check_service_responsive "50051"; then
|
|
log_pass "API Gateway operational without Backtesting"
|
|
else
|
|
log_fail "API Gateway requires Backtesting service"
|
|
fi
|
|
|
|
if check_service_responsive "50052"; then
|
|
log_pass "Trading Service operational without Backtesting"
|
|
else
|
|
log_fail "Trading Service requires Backtesting service"
|
|
fi
|
|
|
|
log_test "Restoring Backtesting Service"
|
|
docker start foxhunt-backtesting-service > /dev/null 2>&1
|
|
sleep 10
|
|
|
|
for i in {1..10}; do
|
|
if check_container_health "foxhunt-backtesting-service"; then
|
|
log_pass "Backtesting Service recovered"
|
|
break
|
|
fi
|
|
sleep 1
|
|
done
|
|
|
|
echo ""
|
|
|
|
# TEST 4: Service Recovery
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "TEST 4: Automatic Service Recovery"
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "Expected: All services recover automatically"
|
|
echo ""
|
|
|
|
log_test "Testing automatic recovery"
|
|
log_info "All services should be healthy again"
|
|
|
|
# Give services time to stabilize
|
|
sleep 5
|
|
|
|
# Check all services healthy again
|
|
RECOVERY_SUCCESS=true
|
|
for container in "${containers[@]}"; do
|
|
if check_container_health "$container"; then
|
|
log_pass "$container recovered successfully"
|
|
else
|
|
log_fail "$container failed to recover"
|
|
RECOVERY_SUCCESS=false
|
|
fi
|
|
done
|
|
|
|
if [ "$RECOVERY_SUCCESS" = true ]; then
|
|
log_pass "All services recovered automatically"
|
|
else
|
|
log_fail "Some services failed to recover"
|
|
fi
|
|
|
|
echo ""
|
|
|
|
# TEST 5: Critical Function Availability
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "TEST 5: Critical Function Availability During Degradation"
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "Expected: Core functions survive all failure scenarios"
|
|
echo ""
|
|
|
|
log_test "Verifying critical functions always available"
|
|
|
|
# Authentication (JWT validation is stateless, no dependencies)
|
|
if check_service_responsive "50051"; then
|
|
log_pass "Authentication available (JWT validation stateless)"
|
|
else
|
|
log_fail "Authentication unavailable"
|
|
fi
|
|
|
|
# Trading operations (core function)
|
|
if check_service_responsive "50052"; then
|
|
log_pass "Trading operations available"
|
|
else
|
|
log_fail "Trading operations unavailable"
|
|
fi
|
|
|
|
# Monitoring (Prometheus should continue)
|
|
if curl -sf http://localhost:9090/-/healthy > /dev/null 2>&1; then
|
|
log_pass "Monitoring operational (Prometheus)"
|
|
else
|
|
log_fail "Monitoring unavailable"
|
|
fi
|
|
|
|
# Metrics endpoints (services should export metrics even when degraded)
|
|
if curl -sf http://localhost:9092/metrics > /dev/null 2>&1; then
|
|
log_pass "Metrics collection operational"
|
|
else
|
|
log_fail "Metrics collection unavailable"
|
|
fi
|
|
|
|
echo ""
|
|
|
|
# Final Summary
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "TEST SUMMARY"
|
|
echo "═══════════════════════════════════════════════════════════════"
|
|
echo "Total Tests: $TESTS_TOTAL"
|
|
echo "Passed: $TESTS_PASSED"
|
|
echo "Failed: $TESTS_FAILED"
|
|
echo ""
|
|
|
|
if [ $TESTS_FAILED -eq 0 ]; then
|
|
echo -e "${GREEN}✓ ALL TESTS PASSED${NC}"
|
|
echo "System demonstrates excellent graceful degradation"
|
|
exit 0
|
|
else
|
|
PASS_RATE=$((TESTS_PASSED * 100 / TESTS_TOTAL))
|
|
if [ $PASS_RATE -ge 80 ]; then
|
|
echo -e "${GREEN}✓ ACCEPTABLE PASS RATE${NC}"
|
|
echo "Pass Rate: ${PASS_RATE}%"
|
|
echo "System demonstrates good graceful degradation"
|
|
exit 0
|
|
else
|
|
echo -e "${YELLOW}⚠ SOME TESTS FAILED${NC}"
|
|
echo "Pass Rate: ${PASS_RATE}%"
|
|
echo "System shows degradation gaps"
|
|
exit 1
|
|
fi
|
|
fi
|