One hard drive is dying

    Hey guys. So, I did an update in OMV dashboard earlier today, and it required a restart. I did it, and after 5-10 minutes, I see that it is not online, still. So, I had to hook up a monitor to my NAS and got this hard drive error below. I have 3 3TB drives in a ZFS pool in RAID-Z1, I think, where 1 drive of 3 can fail. I went to bios and booted to my SSD drive with OMV. I never noticed any issues in everyday work of my NAS until this reboot.


    What's the procedure here now? How to replace this drive and recover my data on the new one?


    Also, since I wont be able to find the exact model, can I get maybe 4TB or 6TB? Also, mine is 7200rpm and 64MB cache. I see that 4TB are now 128MB cache but 5400rpm.

    Nabukodonosor You should have had warning of issues either through enabling S.M.A.R.T and/or via email notifications where ZFS ZED events are enabled.


    Please post output of "zpool status" as shown via WEB UI - select your pool name and then "details" and just cut and paste between code blocks what the current pool status is.


    What's the "power on hours" value for you Toshiba P300 drives? May be you need to plan on gradually replacing them all, or perhaps more quickly if other drives show SMART errors and/or you're getting checksum errors on your pool. When did you last scrub your pool?


    Aren't Toshiba N300 drives still 7200 rpm? Speed and cache size is not critical, if there's no exact match among possible replacements, but make sure it's a CMR drive. If all current HDDs are to replaced over next few moths it make sense to go for at least 4TB drives all of of same type as once all existing HDDs are replaced you will end up with a larger pool of usable space.

    Quote from Nabukodonosor

    https://pastebin.com/LGkb91rH


    And yes, 4TB is the plan. I will look for CMR drives. So, I can just add 4TB in place of this 3TB failed drive? I guess the useful space will still be 3TB until I replace of 3 drives?

    30k hours is well used, but not excessive. True, you will not see the enlarged pool size unit all disks in the pool are replaced. Are you sure this is not a transient error on that hdd? Are their corresponding errors in the SMART tests? I'd be inclined to clear the checksum errors and run a scrub on the pool. If the scrub generates more errors you know for sure the HDD should be replaced fairly promptly.

    OK I missed you had a "bad" SMART indication in your screen shot, but you've not shown what SMART "attributes" are giving this. Check the "attributes" for further info, e.g. read errors, relocated sectors etc. If there's no doubt the disk is bad, then replace it. Clearing checksum errors and zpool scrub should follow any replacement.

    Some people will say just one reallocated sector means you should swap the drive out, others would say up to the threshold number is acceptable and it's how quickly the number grows that's important, so schedule smart tests on hdd accordingly. I'll leave you to read around on the topic and make your own decision. In the meantime, I would clear the checksum errors and scrub the pool now, and then review the result.


    At the CLI : zpool clear MainPool


    Start the scrub from the WEB UI

    I did start the scrub, but didn't clear the checksum. If it's important, I can do it again once it's done.


    Once it's done, where can I see the results? I've never done this.


    Quote

    so schedule smart tests on hdd accordingly

    How often should it be done? Also, should it be short self test, or long?

    If you decide to keep the HDD with the bad sector for now, a daily long test should show if/how the reallocated sector count is increasing. Either way, if it were me, I'd get a spare HDD now, choosing a make/model/size that fits with a longer term plan to replace all the ageing Toshiba drives.

    Yes, that's the plan. I've already ordered some WD Red Plus 4TB, but I will wait with the replacement for now. Couple of weeks, I want to watch/delete some stuff to make the migration faster.

    Is this the correct procedure to replace the disk?

    External Content www.youtube.com
    Content embedded from external sources will not be displayed without your consent.
    Through the activation of external content, you agree that personal data may be transferred to third party platforms. We have provided more information on this in our privacy policy.

Participate now!

Don’t have an account yet? Register yourself now and be a part of our community!

Register Yourself Login