Does someone has real experience using snapshots and backup/restore? We have periodical snapshots and backup, but we never had the need to perform a restore. Of course tested before going in production, it worked perfectly, but it was a limited testbed, now we have about 10000 endpoints and a dual pod infrastructure.
The question is: what is the impact of a rollback in terms of interruption of the communication? Consider this example: we perform a snapshot, then some reconfiguration for some entities (e.g. an L3 out in a tenant and a couple of physical interfaces used in other tenants). If we perform a rollback of the complete fabric, we would like to get impact only in the entities involved in changes done, with thousands of endpoints not involved that continue to communicate without interruptions. What is the real impact of a fabric rollback? Real experiences?
Another point: what is the difference between snapshot and backup?
Yes, I did it right yesterday with almost no interruption.
By exporting snapshots before and after making configuration changes, you have the ability to roll back configuration changes that were applied between two snapshots.
The rollback feature provides an "undo" function that reverts changes made between one snapshot archive and a later snapshot archive.
You can optionally enable the preview mode to generate and view a rollback before implementing it.
(Optional) Specifies that the rollback changes are generated and previewed but not applied. When preview mode is enabled, no changes to the configuration are made. After previewing rollback changes, use the no preview command to exit preview mode and enable the rollback to be applied when you reenter the trigger snapshot rollback commands.
I hope it will be helpful.
Regards,
Ali
Learn, share, save
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
