1 contributor
# autoSMART Cluster Configuration # Location: /etc/pve/autoSMART/cluster.conf # This file is shared across all Proxmox cluster nodes [cluster] # Cluster identification cluster_name = proxmox-cluster-main cluster_id = pve-cluster-001 nodes = node91,node92,node93 # Database configuration (shared cluster database) [database] host = 192.168.2.91 port = 5432 database = autosmart_cluster username = autosmart_cluster password = cluster_secure_password_here connection_timeout = 30 pool_size = 10 # OpenAI configuration (shared API key) [openai] api_key = your_cluster_openai_api_key_here model = gpt-4 max_tokens = 1500 temperature = 0.3 rate_limit_delay = 2 # Madagascar inventory integration [madagascar] inventory_path = /etc/pve/autoSMART/madagascar_inventory.json update_interval = 3600 sync_across_nodes = true # Cluster-wide SMART monitoring parameters [smart_parameters] # Critical parameters (high weight for AI analysis) Reallocated_Sector_Ct = 1,10.0,true,Critical reallocated sectors Reallocated_Event_Count = 1,9.0,true,Reallocation events Current_Pending_Sector = 1,9.5,true,Pending sector reallocation Offline_Uncorrectable = 1,10.0,true,Uncorrectable sectors UDMA_CRC_Error_Count = 10,5.0,true,Communication errors Spin_Retry_Count = 1,8.0,true,Spindle motor retries # Important parameters (medium weight) Raw_Read_Error_Rate = 100000,3.0,true,Raw read errors Seek_Error_Rate = 100000,4.0,true,Seek operation errors Load_Cycle_Count = 100000,2.0,true,Head load cycles Power_On_Hours = 35000,2.0,true,Power-on time Temperature_Celsius = 50,3.0,true,Operating temperature # Monitoring parameters (low weight) Start_Stop_Count = 10000,1.0,true,Start/stop cycles Power_Cycle_Count = 10000,1.0,true,Power cycles Command_Timeout = 100,2.0,true,Command timeouts High_Fly_Writes = 1,4.0,true,Head fly height issues Airflow_Temperature_Cel = 45,1.5,true,Airflow temperature # Cluster-wide alert settings [alerts] email_enabled = true email_smtp_server = mail.domain.com email_smtp_port = 587 email_username = autosmart@domain.com email_password = email_password_here email_recipients = admin@domain.com,ops@domain.com email_critical_only = false # Risk level alert thresholds alert_critical_immediate = true alert_high_delay_minutes = 30 alert_moderate_delay_hours = 4 alert_low_daily_summary = true # Data retention (cluster-wide policy) [retention] smart_readings_days = 365 predictions_days = 180 alerts_days = 90 cleanup_interval_hours = 24 # Cluster synchronization [synchronization] node_discovery_interval = 300 health_check_interval = 60 failover_enabled = true backup_nodes = node92,node93 primary_node = node91