Hi
I just went from M0 to M2 because I had no space left (512Mb consumed)
After migration, I have only 283Mb consumed
Although i believe this is not done on purpose, I still feel a bit like I’ve been tricked.
Can I now go back to my M0 plan?
What method did you use to move your data?
Are you sure all indexes have been recreated?
Thanks @steevej
I just upgraded my cluster from the Atlas website. That’s why it’s a bit frustrating
Hi @Martin_Ratinaud - Welcome to the community.
Can I now go back to my M0 plan?
Unfortunately shared tier clusters cannot be downgraded, and need to be migrated instead. You can use mongodump and mongorestore for this to migrate back to an M0 cluster as noted in the Backup & Restore considerations documentation.
Although i believe this is not done on purpose, I still feel a bit like I’ve been tricked.
I understand your frustrations here and it does sound a bit odd regarding the drop in logical size after the upgrade. I would highly recommend you contact the Atlas support team via the in-app chat to investigate this drop in logical size related to your Atlas account / project as they have further insight into the deployment(s).
Regards,
Jason
Thanks @Jason_Tran I’ve contacted them and here is their answer
the drop in the logical size you see is due to initial sync that ran during the time you upgraded your cluster.
In MongoDB, document deletion does not decrease the size of the DB storage size.
Atlas uses WiredTiger as the default storage engine and it will not release disk space created by deleting documents. However, following a checkpoint, WiredTiger will mark any space freed due to deletes or updates as being available for re-use.
The initial sync freed up the unused disk space to make it available for reuse.
it seems the “WiredTiger will mark any space freed due to deletes or updates as being available for re-use” part is not working that well
1 Like
This topic was automatically closed 5 days after the last reply. New replies are no longer allowed.