managing-foundationdb

name	managing-foundationdb
description	Use when working with Foundationdb — foundationDB cluster management, layer inspection, transaction rate monitoring, and storage engine analysis. Covers cluster health via fdbcli status, process roles, coordination state, backup/restore status, and data distribution metrics. Read this skill before any FoundationDB operations.
connection_type	foundationdb
preload	false

FoundationDB Management Skill

Monitor, analyze, and optimize FoundationDB clusters safely.

MANDATORY: Discovery-First Pattern

Always run fdbcli status before any operations. Never assume cluster configuration or layer availability.

Phase 1: Discovery

#!/bin/bash

FDB_CLUSTER="${FDB_CLUSTER_FILE:-/etc/foundationdb/fdb.cluster}"

echo "=== Cluster Status ==="
fdbcli --exec "status details" -C "$FDB_CLUSTER" 2>/dev/null

echo ""
echo "=== Cluster Configuration ==="
fdbcli --exec "configure get" -C "$FDB_CLUSTER" 2>/dev/null || \
    fdbcli --exec "status json" -C "$FDB_CLUSTER" 2>/dev/null | python3 -c "
import sys, json
data = json.load(sys.stdin)
cluster = data.get('cluster', {})
config = cluster.get('configuration', {})
print(f\"Redundancy: {config.get('redundancy_mode','?')}\")
print(f\"Storage engine: {config.get('storage_engine','?')}\")
print(f\"Log spill: {config.get('log_spill','?')}\")
" 2>/dev/null

Phase 1 outputs: Cluster health, process count, redundancy mode, storage engine, transaction rate.

Phase 2: Analysis

#!/bin/bash

FDB_CLUSTER="${FDB_CLUSTER_FILE:-/etc/foundationdb/fdb.cluster}"

echo "=== Process Health ==="
fdbcli --exec "status json" -C "$FDB_CLUSTER" 2>/dev/null | python3 -c "
import sys, json
data = json.load(sys.stdin)
cluster = data.get('cluster', {})
procs = cluster.get('processes', {})
for pid, p in procs.items():
    roles = [r['role'] for r in p.get('roles', [])]
    print(f\"  {p.get('address','?')}: roles={roles} cpu={p.get('cpu',{}).get('usage_cores','?')} mem={p.get('memory',{}).get('used_bytes',0)//1048576}MB\")
" 2>/dev/null | head -15

echo ""
echo "=== Workload Metrics ==="
fdbcli --exec "status json" -C "$FDB_CLUSTER" 2>/dev/null | python3 -c "
import sys, json
data = json.load(sys.stdin)
wl = data.get('cluster', {}).get('workload', {})
ops = wl.get('operations', {})
for op_type, stats in ops.items():
    print(f\"  {op_type}: {stats.get('hz', 0):.0f} ops/sec\")
txns = wl.get('transactions', {})
for t_type, stats in txns.items():
    print(f\"  txn_{t_type}: {stats.get('hz', 0):.0f}/sec\")
" 2>/dev/null

echo ""
echo "=== Data Distribution ==="
fdbcli --exec "status json" -C "$FDB_CLUSTER" 2>/dev/null | python3 -c "
import sys, json
data = json.load(sys.stdin)
dd = data.get('cluster', {}).get('data', {})
print(f\"Total KV size: {dd.get('total_kv_size_bytes', 0)//1048576}MB\")
print(f\"Total disk used: {dd.get('total_disk_used_bytes', 0)//1048576}MB\")
print(f\"Moving data: {dd.get('moving_data', {}).get('in_flight_bytes', 0)//1048576}MB\")
print(f\"Partitions: {dd.get('partitions_count', '?')}\")
" 2>/dev/null

echo ""
echo "=== Backup Status ==="
fdbcli --  -C  2>/dev/null | python3 -c  2>/dev/null

Output Format

FOUNDATIONDB ANALYSIS
=====================
Cluster: [healthy/degraded/unavailable]
Processes: [count] | Redundancy: [mode] | Engine: [type]

ISSUES FOUND:
- [issue with affected process/role]

RECOMMENDATIONS:
- [actionable recommendation]

Anti-Hallucination Rules

1. NEVER assume resource names — always discover via CLI/API in Phase 1 before referencing in Phase 2.
2. NEVER fabricate metric names or dimensions — verify against the service documentation or --help output.
3. NEVER mix CLI commands between service versions — confirm which version/API you are targeting.
4. ALWAYS use the discovery → verify → analyze chain — every resource referenced must have been discovered first.
5. ALWAYS handle empty results gracefully — an empty response is valid data, not an error to retry.

Counter-Rationalizations

Shortcut	Counter	Why
"I'll skip discovery and check known resources"	Always run Phase 1 discovery first	Resource names change, new resources appear — assumed names cause errors
"The user only asked for a quick check"	Follow the full discovery → analysis flow	Quick checks miss critical issues; structured analysis catches silent failures
"Default configuration is probably fine"	Audit configuration explicitly	Defaults often leave logging, security, and optimization features disabled
"Metrics aren't needed for this"	Always check relevant metrics when available	API/CLI responses show current state; metrics reveal trends and intermittent issues
"I don't have access to that"	Try the command and report the actual error	Assumed permission failures prevent useful investigation; actual errors are informative