![]() ![]() I see a difference in power consumption as the drives draw less power and the chassis is cooler so the PWM fans drop to around 500-600 RPM, drawing less power, instead of screaming at 1600RPM. I just need to workout how to get the power state of the drives. (hint FOR LOOP)īonus points: Reboot and verify changes to power condition survives power cycle. ICT 3000 Idle_a condition timer (100 ms)Repeat all of the above for all SAS drives in your storage pool. # sdparm -flexible -6 -p po -l /dev/sda |grep -e "ICT\|IDLE_C " Disks rotating at full speed (7200 RPM).Disables most of the servo system, reduces processor and channel power consumption.Let's enable and reduce it down to 5 minutes as this is a disk used in a zpool for archives and backups and occasional media file access SYCT 0 Standby_y condition timer (100 ms)ĬCF_IDLE 0 check condition on transition from idleĬCF_STAND 0 check condition on transition from standbyĬCF_STOPP 0 check condition on transition from stoppedNOTE: IBCT (Idle_b condition timer) has has a condition of 6000 (10 minutes) but is not enable (IDLE_B 0) it. IBCT 6000 Idle_b condition timer (100 ms) SCT 9000 Standby_z condition timer (100 ms) PM_BG 0 Power management, background functions, precedence SMART Health Status: OKAnd a bit more information including the temperatureĭirect access device specific parameters: WP=0 DPOFUA=1 SMART support is: Available - device has SMART capability.Ĭopyright (C) 2002-16, Bruce Allen, Christian Franke, = START OF READ SMART DATA SECTION = Smartctl 6.5 r4318 (local build)Ĭopyright (C) 2002-16, Bruce Allen, Christian Franke, = START OF INFORMATION SECTION = These are SCSI drives, the usual hdparm tool command don't work. They weren't spinning down, but why and how to change that behavior. After doing my defaults and setting up a pool (RAID2z) and having the server run for awhile, I realised that the PWM fans wouldn't throttle.Īfter writing a little script to monitor temperature and fan sensors via IPMI and the HDD temps using smartmontools, it became clear that the unaccounted heat was coming from the HDD's. My latest acquisitions were 8 x 4Tb SAS drives. My defaults are to always integrate and update and wipe with a bad blocks check, adjust block size if needed and stress test prior to adding them to a storage pool. Very few have issues but all need some amount attention beyond that when purchasing consumer grade new. Why do I want power control over my spinning rust.spin down to save noise and heat = save energyįor my home lab I buy enterprise HDD's from ebay et al at amazing price points per Gb. My hard drives were spinning at 7600 RPM 24/7.Arrgg! ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |