kindofblue42

Late to the party here, but I wanted to share a script I tweaked to monitor NVME TBW/GBW/health. Requires nvme-cli from Nerd Pack. #!/bin/bash #arrayStarted=true # Ref: # https://forums.unraid.net/bug-reports/stable-releases/683-docker-image-huge-amount-of-unnecessary-writes-on-cache-r733/page/16/?tab=comments#comment-9945 # SSDs drives=( "nvme0n1" "nvme1n1" ) for drive in ${drives[*]} do device=/dev/$drive TBWSDB_TB=$(nvme smart-log $device | awk '/data_units_written/{ gsub(",", "") ; print $3 * 512000 / 1e+12 }') TBWSDB_GB=$(nvme smart-log $device | awk '/data_units_written/{ gsub(",", "") ; print $3 * 512000 / 1e+9 }') PWRON_DAYS=$(nvme smart-log $device | awk '/power_on_hours/{ print $3 / 24}') HEALTH=$(nvme smart-log $device | awk '/percentage_used/{ print $3 }') # Set permissions and clear previous average touch /mnt/user/logs/"$drive"_health.log chmod a+rw /mnt/user/logs/"$drive"_health.log sed -i '/Average writes/d' /mnt/user/logs/"$drive"_health.log echo "TBW on $(date +"%d-%m-%Y %H:%M:%S") --> $TBWSDB_TB TB, which is $TBWSDB_GB GB -- Health (% used, higher is worse): $HEALTH" >> /mnt/user/logs/"$drive"_health.log echo "$TBWSDB_GB $PWRON_DAYS" | awk '{printf "Average writes/day: %.2fGB\n", $1/$2}' >> /mnt/user/logs/"$drive"_health.log done