Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


Scrubbing a SAN
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

Scrubbing a SAN

SpencerSpencer Member
edited April 2013 in General

I have a gigenetcloud VPS and it has been sloooow lately. So I submitted a ticket and the first ticket they said they are looking into it. Then about 10 hours later they said they they are Scrubbing the SANS. Correct me if im wrong but isnt scrubbing a HDD when you remove all the files on it? But my SAN knowledge is very minimal so would you scrub a SAN to increase performance or something?

Comments

  • Maybe they are giving the San some TLC in the bath?

  • @BradND said: Maybe they are giving the San some TLC in the bath?

    Bow chica bow wow.

  • SurgeSurge Member

    Scrubbing means rereading, checking checksums and then sometimes writing again all of the data on the disk to ensure cold data is not corrupted.

  • danodano Member

    With ZFS filesystems, this is common -

    http://en.wikipedia.org/wiki/ZFS#Resilvering_and_scrub

  • netomxnetomx Moderator, Veteran

    @HalfEatenPie said: Bow chica bow wow.

    hahahah i needed to laugh, thanks

  • mnpeepmnpeep Member
    edited April 2013

    "scrub checks everything, including metadata and the data. The effect can be observed by comparing fsck to scrub times, sometimes a fsck on a large raid completes in a few minutes, which means only the metadata was checked. To traverse all metadata and data on a large raid, takes many hours, which is exactly what scrub does."

    That's why it takes so long @Spencer.

  • JacobJacob Member

    Using a single SAN is just trouble, all these horror stories, It makes more sense to just have small hypervisor's, which when one goes down it can fallover to another hypervisor.

  • mikhomikho Member, Host Rep

    @Jacob and how do you failover data if there is no shared storage?

  • JacobJacob Member
    edited April 2013

    NFS.

    @MikHo said: how do you failover data if there is no shared storage?

  • mikhomikho Member, Host Rep

    So you are using NFS to a shared storage where all your vm's are?

    So you are using a NAS instead of SAN?

  • JacobJacob Member

    No.

    @MikHo said: So you are using NFS to a shared storage where all your vm's are?

    No.

    @MikHo said: So you are using a NAS instead of SAN?

    Just stating that rather than having a single huge SAN, these providers should look into multiple smaller SAN's or using NFS between hypervisors.

  • mikhomikho Member, Host Rep

    I dont think that would be very cost effective.

  • JacobJacob Member

    Then it just makes all the cloud hype even more BS.

    NFS is cost effective and if you are multi homed, redundant power + networking gear then it's all covered.

  • lumaluma Member

    @Jacob said: NFS is cost effective and if you are multi homed, redundant power + networking gear then it's all covered.

    with NFS you still need storage somewhere. So you still run into the same problem. A single SAN or a single NFS box? I would rather go with the SAN.

    Now a days decent SAN's have not only multiple PSU's but also multiple controllers which includes cpu, ram and networking.

  • Nfsmwilcbe trash as a San replacement

Sign In or Register to comment.