SME Storage

Members
  • Posts

    6
  • Joined

  • Last visited

Posts posted by SME Storage

  1. My Unraid build is solitarily based on Toshiba N300 drives in regard to HDD's. (4/8 TB)

    Also had one X300 running until I swapped it out for a bigger one. N300 of course.

    Due to my experience does the X300 run a little more warm than the N300. Then again is X300 a cheaper drive.

    Once a new N300 drive failed during zeroing at first time use and got a RMA disk without any problem.

    No other drive N300 died on me yet. Admit to be very pleased with the N300 series.

  2. Of course it is possible to add additional functionality yourself by scripting those. On the other hand when you add custom functionality yourself then at some point this is going to work against you == additional maintenance and testing at every new Unraid version. 

    You might like custom modding and that is fine but for now like to address high level design. 

    By example the KISS principle by implementing out of the box solutions.   
    Hot spare functionality might be a game changer in regard to Unraid's design. Have no doubt. 

    An additional protection layer in securing data availability.  

    Like to think that it is only a matter of time for LimeTech to add hot spare functionality. The logical next step in Unraid's evolution.  

    Look around and see what happens in today's home with IOT. We see more and more automation and simplicity been added to our lives.

    Automation is the way forward. The difference between pro-activity and reactivity.

     

    • Like 1
  3. Just for the sake of Murphy's law while assuming a cooling problem of some sort (ambient) and one or more drives would run hot.

    How long do you want your drives to run hot before a notice is generated? 

    I know it is a bit far fetched but that is the time frame I personally like to work with.

    Have set my Tunable (poll_attributes) to 3 minutes. Only hope that is not too long.

  4. Thank you very much for your post.

     

    At work I manage allmost one petabyte of data. Used raid solution is based on traditional raid (raid-4-DP). The file system is striped over multiple raid arrays.

    This allows use of large file systems. Each raid array can deal with losing 2 data drives simultaniously before a raid array would go down. Hot spares are a must here.

     

     

    Thank you for your explaination about Unraid and how it works. I could not have put that in better writing.

    I am a little worried that I have not been able to find the right words to get the message over. 

    Personally I have no doubt in Unraids raid and file system concept. I feel Unraid is doing a splendit job!

    Double parity based array together with mirrored cache is a good enough solution in most cases.

    In general has data protection different protection levels. Parity is one, double parity is another one, Mirroring is one and backup is also an important one.

     

    Hypothetical speaking in regard to the risk for multi disk failure:

    It might seem over the top at first but when automatic assignment of a hot spare is possible in case a disk has died on you one would save time and by that reduce the chance for a multi disk failure even further. 

    Not even considering a scenario where a replacement disk is not available and still needs to be purchased before this disk can be added to the array.

     

    Hope that I was able to defend my hot spare opinion this time.

     

    Regards,

     

    Marcel

     

    • Upvote 1
  5. Dear Unraiders and Lime Tech,

     

    As a storage professional am I working with storage on a daily basis. Therefore was I searching for hot spare funtionality at Unraid and did not find it.

    (Feel free to correct me when I am wrong.)

     

     

    Definitions:

    Hot Spare: In case of a disk failure a hot spare disk is automatically added to the array and triggers the normal data/parity rebuild.

    Global Hot Spare: Hot spare disk to be used by the complete array.

    Local Hot Spare: Hot Spare intended for one single disk pool.

     

    Question is does Unraid require hot spare functionality??

    Just remember why we use NAS storage? We like to have some level of hardware redundancy for our data.

    Hot spares can be an addition to the overall package of hardware redundancy and when used raise the redundancy level. 

    There are many different (hardware) redundancy solutions one can add to your Unraid system. It is just a matter of how far you want to take it and how important your data is to you.

    Slider always moves between cost on one side and highest possible data redundancy on the other side.

    Benefit which hot spare feature would bring is the fact that the feature is fully automatic from the time a hot spare has been made available.

    Another benefit is that the time your array runs in degraded mode is reduced.

    Some posts expressed worry that in case unraid encounters a bad SATA connection a hot spare will kick in. (when available)

    Exactly what I would want! First priority is the health of the array.

    Other posts have mentioned the problem of having a hot spare available at disk array's with different disk capacities.

    Well that is true. It is a bit more work when you use disks with different capacity to come to the correct disk size you need. 

    In this case it is more easy to use disks with same capacity for every drive pool.

     

    If hot spare feature is optional one could choose to use this feature or not. 

    A choice to have a global or local hot spare would complete the whole.
     

    Would it not be a very peaceful thought when I am at work and I do not have to worry about my unraid system at home in case my Unraid system could have a hot spare available....

    Proactivity is the definition of being in control.

     

    Cheers,

     

    Marcel

     

     

     

     

     

    • Like 1
    • Upvote 1