January 3, 201115 yr I have smart data for 2 WD20EARS in mymain plugin / smart tab showing >30000 LOAD_CYCLE_COUNT. It is increasing every few seconds. I have had these drives for ~6 months now and am sure I am sure this has only been going on for the last few days. What can be the reason? Just to add - I have another WD10EADS drive that is much older and is at a constant 185 load cycle count
January 3, 201115 yr Most likely issue is power savings features in the drive's firmware have gone stray. There is a utility you can run called wdidle3 that can tune the settings to limit the excessive load cycle counts. I have 3 EADS drives and none have the LCC issue. There have been reports by others where some of their drives have it and some do not. It's a very odd issue since it doesn't seem to be consistent. In your case, did you install any addons that do periodic writes or content reads to the drives, causing them to unload and then load once idle, only to become active again a few seconds later? Another possible cause could be borderline PSU, however that should also increase the power cycle counts by the same rate, so I doubt that's the case here.
January 4, 201115 yr Take a look at this post I made a while back: http://lime-technology.com/forum/index.php?topic=9073.msg86601#msg86601 I had two WD20EARS drives with very high LCC's, and they weren't old at all. While it's probably harmless (that seems to be the majority consensus), as these drives can reach the millions, I preferred to run wdidle3 on mine, increasing the time from the stock 8 seconds to (IIRC) 5 minutes. The increase in those LCC numbers have decreased significantly. In fact, I ended up running this on all of my EARS drives (I have . Perhaps my power consumption has increased, but I seriously doubt it is material.
January 4, 201115 yr I had a netcat molex to 4 way sata power connector I had to move when shifting drives around to make a test server ... when I put it back it ran well for a few days ... and then the load cycle count ran up about 100,000 in under two hours on five drives ... I powered down and checked all the cables ... I powered it back up and was good for another two days ... and then they started spiking up at about the same rate ... I'm guessing it was loosing power on the 5V or 12V or ?V lines intermittently ... I replaced the netcat molex sata power 4 way with some regular splitters and it has been stable since. or it may have been a stray alpha particle that flipped the wrong bit and rebooting helped?
January 4, 201115 yr Author To continue my OP - I am averaging about 1000 LCCs a day. This gives me a reason to suspect the problem exists for about 30. Checking my unraid flash drive for changed files I found that the "mymain" major update was on December 5 - exactly 30 days ago. Can someone else check if they have this problem arise in the last month. I will try to uninstall unMenu (and myMain) and see what happens in the next day or 2.... Do I do this by simply removing the /boot/unmenu/uu from the go script? - tried uninstalling - too many things not working - shares, etc... Looking for another option...
January 4, 201115 yr use the button on the user-scripts page to kill unmenu. Or, log in and type killall awk Joe L.
January 5, 201115 yr UnMenu checks the drive ststus by using hdparm -C before checking for temp by using smartctl. by default, it checks once per minute when on the main page. Other than that it does nothing to spin up drives unless you press the spin-up button All that in mind, it could trigger a spinup if the firmware on the disk did not act correctly when it got the hdparm command.
January 5, 201115 yr UnMenu checks the drive ststus by using hdparm -C before checking for temp by using smartctl. by default, it checks once per minute when on the main page. Other than that lit does nothing to spin up drives unless you press the spin-up button All that in mind.it could trigger a spinup if the firmware on the disk d d not act correctly when it got the hdparm command. I beleive there is an issue with older versions of hdparm and the SuperMicro SAS controller where drives would spin up if inquired about their spin status. Don't remember exactly the issue.
Archived
This topic is now archived and is closed to further replies.