• Nicholas
    0
    Hello,

    We have several clients with unfortunately horrid upload speeds. Each client has block-level, image based backups configured with a full backup scheduled every month.

    However, some clients can take more than a week to upload their data. During this time, the other daily local backups cannot run for this entire week as I face the "Another image based backup is running" warning.

    Is there a way to give the local backup precedence (only takes a few minutes), effectively pausing the cloud backup during the local backup, and then resuming the cloud backup once the local has completed? Or really any way for a local and cloud block-level, image based backups to play nicely with one another (hybrid doesn't fix this issue).

    If not, is there a way to seed a block-level, image based backup at another location? Does https://kb.cloudberrylab.com/cloud-vendors/seeding-data/seeding-data-via-cloudberry-backup-part-ii apply to block-level, image based backups as well?

    Not sure the best way to go about this. Any advice is appreciated!

  • Matt
    91
    Seeding image-based backup is possible, but it would still require full backups to keep versioning and purging in check. If you're using S3 or Wasabi it's better to use synthetic full in that case.
  • Nicholas
    0
    Hello Matt, thank you for your response, and answering the question if a seeding an image-based backup is possible.

    Our use case defines that we use Glacier for our storage, and so a synthetic full backup is not possible.

    Is there a way to give the local backup precedence, effectively pausing the cloud backup? The worry is that if a cloud backup is taking a while (say, more than a week), then the local, image-based backups will not occur for that length of time.

    Thanks for your response!
  • MattAccepted Answer
    91
    If you're using the so called "pure Glacier"(i.e. the one that uses Vaults) you won't be able to use any kind of seeding due to how it works. Note that restoration of image-based backups from Glacier could be a long and somewhat painful process, again, due to how this storage class works.

    As for local backup precedence, we currently don't have anything like that, so the best way would be to simply use local backups instead of the cloud ones if upgrading internet connection is not an option.
  • Nicholas
    0
    We're not using the "pure Glacier", just the storage class. Even then, the synthetic backup fails (support has told me that Glacier was the cause of the errors, and so I cannot use synthetic).

    Anyways, thanks for the response for pausing the local backup. Looks like we'll have to figure something else out. Thanks!
  • Matt
    91
    Yes, synthetic full won't work even with Glacier storage class due to API limitations, but we're working on that.
bold
italic
underline
strike
code
quote
ulist
image
url
mention
reveal
youtube
tweet
Add a Comment