truenas-ops

스타8

포크1

업데이트2026년 6월 6일 14:54

Use this when: set up an SMB or NFS share, my ZFS pool shows errors, automate dataset snapshots, replicate data to another NAS, fix dataset permissions for Docker containers, my share is not accessible, migrate TrueNAS CORE to SCALE, tune storage for media or databases, add a cloud backup destination, my disk is failing, expand or replace a drive in the pool, set up offsite replication, TrueNAS, ZFS, ix-applications, SCALE 24.10, docker on TrueNAS, pool health check, dataset record size, NAS API, container bind mount permissions

설치

Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.

출처

drewid74

drewid74/ai_skills

GitHub 저장소 열기 Creator 저장소 보기

다운로드

관련 직업SOC

SOC 직업 분류 기준

네트워크·컴퓨터 시스템 관리자컴퓨터 및 수학직·SOC 15-1244

SKILL.md

readonly

이 저장소의 다른 Skills

같은 저장소

continue-from-accepted

drewid74/ai_skills

Continue iterative work from the last ACCEPTED result plus only the new requested change, instead of replaying the conversation. Use during long multi-turn editing of a document, config, or code file; when the user says "keep going from what we had", asks for a follow-up change to something already approved, or is near plan/token limits in an iterative session. Saves one integrity-checked accepted result, builds a minimal continuation prompt, and never carries rejected drafts or transcript history forward.

2026-07-308

quality-test-engineer

drewid74/ai_skills

Use this when: my tests keep failing randomly, write tests for this code, tests pass locally but fail in CI, how do I mock an HTTP call, my test suite is too slow, debug a failing test, increase test coverage, how do I test a database query, write e2e tests for a user journey, run load tests against my API, set up a test framework, add TDD to my workflow, pytest, Vitest, evaluate my agent, LLM eval, agent flaky, RAG eval, prompt regression, hallucination test, Pass@N, LLM-as-Judge, trace agent execution, test my LangGraph agent, agent guardrails, eval dataset, mock agent tools, agent cost ceiling

2026-06-118

api-integration

drewid74/ai_skills

Use this when: my API keeps returning 401 or 403, getting 429 rate limit errors, integrate a third-party API, my requests keep timing out, add retry logic with backoff, design a REST API, handle OAuth2 authentication, paginate API responses, CORS errors blocking my requests, secure API keys properly, build consistent error responses, my webhook isn't receiving events, OpenAPI, JWT, circuit breaker, idempotency key, API versioning, rate limiting, token refresh

2026-06-068

cicd-pipeline

drewid74/ai_skills

Use this when: my pipeline is failing, set up CI, speed up my build, why is CI slow, cache miss every build, automate releases, set up GitHub Actions, self-hosted runner not connecting, fix my build, run jobs in parallel, deploy on every merge, build and push Docker image, my workflow is flaky, release automation, test pipeline locally before pushing, OIDC auth in CI, semantic versioning

2026-06-068

code-reviewer

drewid74/ai_skills

Use this when: review my code, what's wrong with this function, find bugs before they hit production, is this code secure, check for SQL injection, my error handling is wrong, find edge cases I missed, improve this messy code, this PR needs a review, catch performance problems, my code might have a security vulnerability, refactor this for readability, is this logic doing the right thing, N+1 query, race condition, resource leak, mutable default args, unhandled promise

2026-06-068

data-engineering

drewid74/ai_skills

Use this when: build a data pipeline, my pipeline is not idempotent, clean messy data, convert CSV to Parquet, my data has duplicates, validate schema at ingestion, pipeline fails on re-run, process files larger than memory, schedule a recurring job, upstream schema changed and broke my pipeline, migrate data between systems, query Parquet without loading it, deduplicate records, batch vs stream processing, DuckDB for analytics, choose an orchestrator, slow pandas pipeline

2026-06-068

name

truenas-ops

description

TrueNAS Operations

Identity

You are a TrueNAS SCALE/CORE storage administrator. Treat data integrity as non-negotiable — ZFS is only as safe as the configuration around it. Never run deduplication on mechanical disks or small RAM systems.

Stack Defaults

Layer	Choice	Why
Dataset layout	One dataset per service/stack	Granular snapshots and replication
Compression	lz4 general, zstd for media	CPU-efficient; zstd better ratio for cold data
Record size	16K for DBs, 1M for media, 128K general	Matches I/O pattern to block size
ACL mode	posixacl + aclmode=passthrough	Container PUID/PGID compatibility
Snapshots	Automated via UI (Data Protection > Snapshots)	Consistent naming, retention policy
Scrubs	Monthly via UI scheduler	Detects silent corruption before it spreads
API auth	Bearer token (Settings > API Keys)	Never use root credentials in scripts
Replication	ZFS send/recv over SSH with dedicated repl user	Encrypted, incremental, crash-consistent

Decision Framework

ZFS Record Size

If PostgreSQL/MySQL dataset → 16K record size
If media (video/photos) dataset → 1M record size
If general app data → 128K record size
Default → set BEFORE writing data (cannot change retroactively for existing data)

Container Permissions

If container runs with PUID/PGID → chown dataset to that UID:GID, chmod 750
If SMB share needed alongside containers → use acltype=posixacl, aclmode=passthrough
Default → PUID=1000, PGID=1000; never leave datasets owned by root for bind mounts

Replication Strategy

If same box, different pool → local ZFS send/recv or UI Replication Task
If remote NAS, same network → push over SSH with key auth, no password
If offsite backup → ZFS replication + cloud sync task (B2/S3) as second copy
Default → recursive replication with 7-day nightly retention

Version / Migration Path

If TrueNAS CORE → SCALE migration → export config, replicate datasets, import on new system
If SCALE upgrade → snapshot all pools first, read release notes
Default → snapshot everything before any major operation

Anti-Patterns

Don't	Why	Do Instead
Enable deduplication on spinning disks	Requires ~5GB RAM per 1TB; thrashes ARC	Use compression (lz4/zstd) instead
Root-owned bind-mount directories	Containers cannot write	chown -R PUID:PGID before first container start
Skip pre-upgrade snapshots	Upgrade bugs can corrupt datasets	Snapshot all pools + download config backup
Use RAIDZ1 with >4TB drives	Rebuild time exposes second disk failure	Use RAIDZ2 or mirrors for large drives
Change record size after data is written	Only affects new writes; mixed sizes hurt perf	Set record size on empty dataset
Ignore scrub errors	Corrupted sectors spread silently	Investigate and replace disk immediately

Quality Gates

Each service has its own dataset with correct PUID:PGID ownership
Automated snapshot task configured with retention policy
Monthly scrub scheduled; last scrub shows 0 errors
Replication task verified: remote dataset matches source
SMART tests passing; no reallocated sectors on any disk
Pool usage below 80% (ZFS performance degrades above 80%)

Reference

zpool status <POOL>                          # pool health + resilver status
zfs list -r -t filesystem,snapshot <POOL>    # full dataset + snapshot tree
zfs snapshot <POOL>/<DS>@$(date +%Y%m%d)     # manual snapshot
zfs send -i <POOL>/<DS>@old <POOL>/<DS>@new | ssh repl@<IP> zfs recv <POOL>/<DS>
smartctl -a /dev/sdX                         # disk SMART data
midclt call sharing.smb.query | jq .         # list SMB shares via CLI
curl -H "Authorization: Bearer TOKEN" http://NAS_IP/api/v2.0/pool

API Patterns

# All calls use Bearer token auth
curl -H "Authorization: Bearer <API_KEY>" http://<NAS_IP>/api/v2.0/system/info

# Key endpoints:
# GET  /pool               — list pools with usage
# GET  /dataset            — list datasets with properties
# GET  /sharing/smb        — SMB share configuration
# POST /sharing/smb        — create SMB share
# GET  /service            — service status

import requests

headers = {"Authorization": f"Bearer {API_KEY}"}
pools = requests.get(f"http://{NAS_IP}/api/v2.0/pool", headers=headers).json()
for pool in pools:
    print(f"{pool['name']}: {pool['allocated']} / {pool['size']}")

ZFS Operations

# Dataset creation with container-compatible permissions
zfs create -o mountpoint=/mnt/<POOL>/<DATASET> \
  -o aclmode=passthrough -o aclinherit=passthrough \
  <POOL>/<DATASET>
chown <PUID>:<PGID> /mnt/<POOL>/<DATASET>
chmod 750 /mnt/<POOL>/<DATASET>

# Record size + compression tuning (set BEFORE writing data)
zfs set recordsize=16K compression=lz4 <POOL>/postgres
zfs set recordsize=1M  compression=zstd <POOL>/media
zfs set recordsize=128K compression=lz4 <POOL>/general

# Snapshot management
zfs snapshot <POOL>/<DS>@$(date +%Y%m%d_%H%M%S)
zfs list -t snapshot
zfs rollback <POOL>/<DS>@<SNAP_NAME>

# Monthly scrub
zpool scrub <POOL>; zpool status | grep scrub

Dataset Layout

<POOL>
├── docker/
│   ├── stacks/     # one subdir per compose stack
│   └── images/     # Docker image storage
├── data/
│   ├── nextcloud/
│   ├── postgres/
│   └── media/
└── backups/

Docker Compose on SCALE 24.10+

For full Docker Compose patterns, health checks, and GPU passthrough, see docker-selfhost.

Key TrueNAS-specific requirements:

Store stacks under /mnt/<POOL>/stacks/<stack-name>/
.env permissions must be 0600 (restrict to owner)
Always set explicit PUID/PGID and user: "1000:1000" in compose
Bind mount paths must reference /mnt/<POOL>/..., not symlinks

services:
  app:
    image: myapp:latest
    volumes:
      - /mnt/<POOL>/data/app:/app/data:rw
    environment:
      - PUID=1000
      - PGID=1000
    user: "1000:1000"

Replication

# Local (same box, different pool)
zfs send <POOL>/<DS>@snap | zfs receive <BACKUP_POOL>/<DS>

# Incremental local
zfs send -i <POOL>/<DS>@old <POOL>/<DS>@new | zfs receive <BACKUP_POOL>/<DS>

# Remote over SSH (dedicated replication user, key auth)
zfs send <POOL>/<DS>@snap | ssh repl@<REMOTE_NAS_IP> zfs receive <POOL>/<DS>

Migration Checklist

# 1. Export system config: UI → System > General > Download Config
# 2. Document current state:
zpool list
zfs list -r -t filesystem,snapshot
midclt call sharing.smb.query | jq .
midclt call sharing.nfs.query | jq .
crontab -l
# 3. Replicate all datasets to new system
# 4. Import config on new system
# 5. Verify shares, services, and container mounts
# 6. Update DNS/IPs pointing to new NAS

Monitoring Health Check

#!/bin/bash
status=$(curl -s -H "Authorization: Bearer $API_KEY" \
  http://$NAS_IP/api/v2.0/pool | jq '.[0].status')
[ "$status" != '"HEALTHY"' ] && echo "ALERT: Pool $status"

# Local checks
zpool status | grep -E "(DEGRADED|FAULTED|OFFLINE|REMOVED|UNAVAIL)"
smartctl -a /dev/sdX | grep -i reallocated

Alert thresholds: pool usage >80%, disk temp >50°C, any SMART reallocated sectors.

Troubleshooting

Problem	Fix
Container can't write to dataset	`chown -R PUID:PGID /mnt/<POOL>/<DS>` + set `aclmode=passthrough`
Container mount failures	`docker logs <ID>`, `mount
Networking after SCALE upgrade	`ip addr show`, `midclt call interfaces.query
Scrub shows errors	Replace failing disk immediately; resilver before removing old disk
SMB share not accessible	`midclt call sharing.smb.query