Newer
■
■
■
■
■
■
■
■
■
■
Older
88 lines
|
2.636kb
| 1 |
# autoSMART Cluster Configuration |
|
| 2 |
# Location: /etc/pve/autoSMART/cluster.conf |
|
| 3 |
# This file is shared across all Proxmox cluster nodes |
|
| 4 | ||
| 5 |
[cluster] |
|
| 6 |
# Cluster identification |
|
| 7 |
cluster_name = proxmox-cluster-main |
|
| 8 |
cluster_id = pve-cluster-001 |
|
| 9 |
nodes = node91,node92,node93 |
|
| 10 | ||
| 11 |
# Database configuration (shared cluster database) |
|
| 12 |
[database] |
|
| 13 |
host = 192.168.2.91 |
|
| 14 |
port = 5432 |
|
| 15 |
database = autosmart_cluster |
|
| 16 |
username = autosmart_cluster |
|
| 17 |
password = cluster_secure_password_here |
|
| 18 |
connection_timeout = 30 |
|
| 19 |
pool_size = 10 |
|
| 20 | ||
| 21 |
# OpenAI configuration (shared API key) |
|
| 22 |
[openai] |
|
| 23 |
api_key = your_cluster_openai_api_key_here |
|
| 24 |
model = gpt-4 |
|
| 25 |
max_tokens = 1500 |
|
| 26 |
temperature = 0.3 |
|
| 27 |
rate_limit_delay = 2 |
|
| 28 | ||
| 29 |
# Madagascar inventory integration |
|
| 30 |
[madagascar] |
|
| 31 |
inventory_path = /etc/pve/autoSMART/madagascar_inventory.json |
|
| 32 |
update_interval = 3600 |
|
| 33 |
sync_across_nodes = true |
|
| 34 | ||
| 35 |
# Cluster-wide SMART monitoring parameters |
|
| 36 |
[smart_parameters] |
|
| 37 |
# Critical parameters (high weight for AI analysis) |
|
| 38 |
Reallocated_Sector_Ct = 1,10.0,true,Critical reallocated sectors |
|
| 39 |
Reallocated_Event_Count = 1,9.0,true,Reallocation events |
|
| 40 |
Current_Pending_Sector = 1,9.5,true,Pending sector reallocation |
|
| 41 |
Offline_Uncorrectable = 1,10.0,true,Uncorrectable sectors |
|
| 42 |
UDMA_CRC_Error_Count = 10,5.0,true,Communication errors |
|
| 43 |
Spin_Retry_Count = 1,8.0,true,Spindle motor retries |
|
| 44 | ||
| 45 |
# Important parameters (medium weight) |
|
| 46 |
Raw_Read_Error_Rate = 100000,3.0,true,Raw read errors |
|
| 47 |
Seek_Error_Rate = 100000,4.0,true,Seek operation errors |
|
| 48 |
Load_Cycle_Count = 100000,2.0,true,Head load cycles |
|
| 49 |
Power_On_Hours = 35000,2.0,true,Power-on time |
|
| 50 |
Temperature_Celsius = 50,3.0,true,Operating temperature |
|
| 51 | ||
| 52 |
# Monitoring parameters (low weight) |
|
| 53 |
Start_Stop_Count = 10000,1.0,true,Start/stop cycles |
|
| 54 |
Power_Cycle_Count = 10000,1.0,true,Power cycles |
|
| 55 |
Command_Timeout = 100,2.0,true,Command timeouts |
|
| 56 |
High_Fly_Writes = 1,4.0,true,Head fly height issues |
|
| 57 |
Airflow_Temperature_Cel = 45,1.5,true,Airflow temperature |
|
| 58 | ||
| 59 |
# Cluster-wide alert settings |
|
| 60 |
[alerts] |
|
| 61 |
email_enabled = true |
|
| 62 |
email_smtp_server = mail.domain.com |
|
| 63 |
email_smtp_port = 587 |
|
| 64 |
email_username = autosmart@domain.com |
|
| 65 |
email_password = email_password_here |
|
| 66 |
email_recipients = admin@domain.com,ops@domain.com |
|
| 67 |
email_critical_only = false |
|
| 68 | ||
| 69 |
# Risk level alert thresholds |
|
| 70 |
alert_critical_immediate = true |
|
| 71 |
alert_high_delay_minutes = 30 |
|
| 72 |
alert_moderate_delay_hours = 4 |
|
| 73 |
alert_low_daily_summary = true |
|
| 74 | ||
| 75 |
# Data retention (cluster-wide policy) |
|
| 76 |
[retention] |
|
| 77 |
smart_readings_days = 365 |
|
| 78 |
predictions_days = 180 |
|
| 79 |
alerts_days = 90 |
|
| 80 |
cleanup_interval_hours = 24 |
|
| 81 | ||
| 82 |
# Cluster synchronization |
|
| 83 |
[synchronization] |
|
| 84 |
node_discovery_interval = 300 |
|
| 85 |
health_check_interval = 60 |
|
| 86 |
failover_enabled = true |
|
| 87 |
backup_nodes = node92,node93 |
|
| 88 |
primary_node = node91 |