NETGEAR is aware of a growing number of phone and online scams. To learn how to stay safe click here.
Forum Discussion
[Solved] Seagate ST3000DM001-9YN166: Load Cycle Count
Hi,
sorry if this has been discussed already, all i could find are topics about upgrading the firmware on these drives.
I've got a NV+v2 with 4 3TB ST3000DM001-9YN166 disks in it, firmware CCH4 (the latest).
System has been running for a couple of months now, and all was working smoothly until i got a warning 2 weeks ago:
Looking into the SMART status, it indeed gives some ridiculously high number (4295032833) on command time out for disk 1. On the other 3 disks it's 0.
Another thing i noticed in the SMART status for all disks though, is that the load count cycle is quite high: around 43000 after 1800 power on hours.
After a bit of googling however, all i could find was this problem cropping up on WD green drives. The solution seems to be to set the idle time a bit higher.
I did find this old topic (http://www.readynas.com/forum/viewtopic.php?f=36&t=51536) explaining how to do it on some other seagate model, but i wanted to get some feedback before i attempt to change things via ssh. Don't wanna corrupt anything.
So basically my questions are:
1) Is the command time out problem in anyway related to the lcc problem, or is this drive just about to die no matter what?
2) Does anybody else have these high lcc numbers with the ST3000DM001 drives?
3) Is it worthwhile changing the idle time setting with hdparm?
3b) If yes, are the instructions in that old topic still valid?
On a related note, i did install the ssh add-on and looked around a little, however when i try to get drive status with hdparm i get this:
Same for sdb, sdc, and sdd
Am i using the right identifiers here? Sorry for the ignorance :oops:
Hope this isn't all too confusing, i tried to cram a lot of information in this post :-)
Thanks in advance for any help!
EDIT: apparently i do not have permission to use the url tag..
sorry if this has been discussed already, all i could find are topics about upgrading the firmware on these drives.
I've got a NV+v2 with 4 3TB ST3000DM001-9YN166 disks in it, firmware CCH4 (the latest).
System has been running for a couple of months now, and all was working smoothly until i got a warning 2 weeks ago:
Detected increasing command timeouts[65537] on disk 1 [ST3000DM001-9YN166, W1F083HJ]. This often indicates an impending failure. Please be prepared to replace this disk to maintain data redundancy.
Looking into the SMART status, it indeed gives some ridiculously high number (4295032833) on command time out for disk 1. On the other 3 disks it's 0.
Another thing i noticed in the SMART status for all disks though, is that the load count cycle is quite high: around 43000 after 1800 power on hours.
After a bit of googling however, all i could find was this problem cropping up on WD green drives. The solution seems to be to set the idle time a bit higher.
I did find this old topic (http://www.readynas.com/forum/viewtopic.php?f=36&t=51536) explaining how to do it on some other seagate model, but i wanted to get some feedback before i attempt to change things via ssh. Don't wanna corrupt anything.
So basically my questions are:
1) Is the command time out problem in anyway related to the lcc problem, or is this drive just about to die no matter what?
2) Does anybody else have these high lcc numbers with the ST3000DM001 drives?
3) Is it worthwhile changing the idle time setting with hdparm?
3b) If yes, are the instructions in that old topic still valid?
On a related note, i did install the ssh add-on and looked around a little, however when i try to get drive status with hdparm i get this:
root@nas:~# hdparm -i /dev/sda
/dev/sda:
HDIO_GET_IDENTITY failed: Inappropriate ioctl for device
Same for sdb, sdc, and sdd
Am i using the right identifiers here? Sorry for the ignorance :oops:
Hope this isn't all too confusing, i tried to cram a lot of information in this post :-)
Thanks in advance for any help!
EDIT: apparently i do not have permission to use the url tag..
50 Replies
Replies have been turned off for this discussion
I believe he means the "full disk test" function on the boot menu, which is described in the V2 hardware manual. If so, that test is only done upon request.intothevoid wrote: The problem appears isolated to the disk test. So that would seem to rule out the supply and the contacts - they wouldn't mysteriously heal when the test completes.
It is conceivable that the disk test includes some commands that provoke the drive to fail. For instance, malformed/illegal commands, or perhaps legal but unusual commands that the drive firmware doesn't handle properly.
I'd suggest a support case.
In the meantime I would simply not run that test! And record/track the current counts and confirm that they never increase in normal operation.
Is that test run automatically at boot, or on a schedule? I've never manually triggered it.
Both times my drive 'failed' (and had a higher command timeout value afterwards) the NAS was in normal operation.
EDIT: Included quote for top of page clarity.
It looks to me like there is more than one issue here related to command timeouts for this disk model. For some, the command timeout errors in the alerts/log appear to be false (at least the count is reported as 0 when looking directly at the SMART stats). Others have some real command timeout errors, but the count in the alert is wrong. In spett's case, the timeout counts (though huge) are confirmed when looking directly at the SMART stats, and these counts only go up when he manually runs a disk test on the ReadyNAS.
I'm not sure how to untangle this now, since we have several folks seeing different things posting on the same thread (whose title is not about command timeouts at all!).intothevoid wrote: The problem appears isolated to the disk test. So that would seem to rule out the supply and the contacts - they wouldn't mysteriously heal when the test completes.
It is conceivable that the disk test includes some commands that provoke the drive to fail. For instance, malformed/illegal commands, or perhaps legal but unusual commands that the drive firmware doesn't handle properly.
I'd suggest a support case.
In the meantime I would simply not run that test! And record/track the current counts and confirm that they never increase in normal operation.
Is that test run automatically at boot, or on a schedule? I've never manually triggered it.
Both times my drive 'failed' (and had a higher command timeout value afterwards) the NAS was in normal operation.
EDIT: Included quote for top of page clarity.
No, I have run this test manually trough the boot menu. That is a offline hard drive test.
Haven't noticed anything unusual in normal operation without the nas completely freezing when using sabnzbd from readynas extras. Could be a problem with sabnzbd, just strange that no one is mention any problems with it..- I am running a different seagate drive st32000641as, but again in Drive 1 and have just rcvd error with high command timeouts of 4295032833.
I took out drive and ran sea tools and the drive passed every test.
The other 2 drives (I'm only running 3 drives at present in the NV+ v2) are fine at zero command timeouts.
I did have disk spindown set to 15 minutes and have since turned that off. - Hmm, seems to me like we're looking at two different paths leading to the same problem.
1) running the disk test leads to an extremely high command timeout on disk1.
2) the NAS drops disk1 saying it has failed, after reboot it resyncs and the command time out has increased (to a ridiculous value).
The high load cycle count form the thread title was my original guess for the cause of the problem, but after that was fixed with the HDPARM tweak, the command timeout problem (via scenario 2) persisted.
So either there's something wrong with all our drives, and it's a coincidence they all happen to be installed in the first slot on the NAS, or there is some weird compatibility issue with these seagate ST's and the NV+ v2, which causes the drive installed as disk1 to fail.
It's still strange why for some the command time out value seems to be real and for some not though. To complicate matters, I cannot test my drives in a normal computer as i only have a laptop...
Seatools will find and test them even if they are in a USB enclosure or USB->Sata adapter. My laptop has eSata, and that of course works as well (though you need an eSata->Sata cable, and a power supply).intothevoid wrote: To complicate matters, I cannot test my drives in a normal computer as i only have a laptop...
Though you can test it (and verify the SMART stats), upgrading the firmware when you are using USB is another matter.intothevoid wrote:
So either there's something wrong with all our drives, and it's a coincidence they all happen to be installed in the first slot on the NAS, or there is some weird compatibility issue with these seagate ST's and the NV+ v2, which causes the drive installed as disk1 to fail.
I think I have ruled out faulty disks by resyncing a new drive in slot 1 and then sync the disc that where in slot 1 in slot 2.
Same result after a disk test, command timeout only increased on disk 1.StephenB wrote: Seatools will find and test them even if they are in a USB enclosure or USB->Sata adapter. My laptop has eSata, and that of course works as well (though you need an eSata->Sata cable, and a power supply).
Though you can test it (and verify the SMART stats), upgrading the firmware when you are using USB is another matter.
Ah, good to know. Will test it this weekend when i have some more time.Spett wrote: I think I have ruled out faulty disks by resyncing a new drive in slot 1 and then sync the disc that where in slot 1 in slot 2.
Same result after a disk test, command timeout only increased on disk 1.
Yeah seems like it's not the drives. What do you guys think, time to open a support ticket? And if yes, can I refer to this thread if I make one?- My 3 ST drives were donated from a friend who had them running in a DROBO enclosure error free for 6 months.
Now I know a drive can fail at any time - but coincidence that after a week in my brand new readynas nv+ v2, whilst sitting idle overnight, I suddenly get 4295032833 errors??
I dont think so.
I agree - sounds very much like Firmware issues on Readynas to me. - Just thought I'd add that I'm experiencing the same issue with my NV+ v2 and Seagate ST2000DM001. I've had the NV+ v2 and 3x ST2000DM001 working fine for just over 3 months now. All drives were updated by me the CC4H firmware when new, prior to being installed in the NAS. Also, prior to installing, I checked each disk with all Seatools tests (boot disk) and all were fine.
Suddenly, yesterday, I got an email from the NAS telling me that all 3 disks had failed. I ended up having to unplug the NAS from the wall to get it to power off (it was unresponsive). When I rebooted, my volume was back to normal and everything seems fine.
However, looking at the SMART info I see a command timeout value for Disk 1 of 4295032833. Both Drives 2 and 3 have a command timeout of "0". I have no idea what this value was before yesterday's incident or whether the two are related...
Seems like there might be a compatibility issue between these Seagate drives and the NV+ V2. Strange that it's always Drive 1 that has these super high command timeouts and that the high numbers are always about the same, if not identical.
I have never run any full disk check in the NAS (that you access through the boot menu) only the shorter volume check on reboot. My spin-down time is set for 60 min and has been so for months. have been on 5.3.5 the whole time. - FYI, I was reading on the Seagate forums that a SMART Command Timeout reading of 4295032833 is actually a hexadecimal value.
In hexadecimal, 4295032833 = 0x0001 0001 0001, which the poster thought represented a value of "1"
Therefore, the reading of 4295032833 probably doesn't indicate that a whole bunch of Command Timeouts has occurred, but rather only a single one.
NETGEAR Academy
Boost your skills with the Netgear Academy - Get trained, certified and stay ahead with the latest Netgear technology!
Join Us!
