2

The documentation claims that snapshots taken by ops manager are delta (differential) between successive snapshots.

  • How do you then restore?
  • I mean if you are able to delete some snapshots according to retention, how do you connect the dots in between to restore to a point in time.
  • What do they mean by base snapshot? Is that whole backup?
  • How can a given snapshot become full and ready for restore if any snapshot before it is deleted?

I understand one can replay oplog to go forward to a point in time from a given snapshot. This question is if any snapshot before the snapshot where restore begins gets deleted how can you restore?

From MongoDB docs:

"...The snapshot only stores the differences between successive snapshots. Snapshots use only a fraction of the disk space required for full snapshots."

Any help is much appreciated.

SqlWorldWide
  • 13,687
  • 3
  • 30
  • 54
AAgurmu
  • 21
  • 2

1 Answers1

0

I'm not sure which version of Ops Manager you were running, but there's been a lot of improvements in the last year. The latest docs explain the space-spacing approaches of the different storage choices:

  • MongoDB blockstore and AWS S3 bucket - Each successive snapshot stores only the differences from the previous snapshot. Compression and block-level de-duplication reduce the size of snapshot data.
  • File system store - uses compression only.
Nicholas Tolley Cottrell
  • 111
  • 3