• EN
Solved

K10 Import policy testing and full reset

  • 10 November 2023
  • 5 comments
  • 70 views
B

I am currently developing a DR solution.  I currently have the backup on the source side exporting snapshots.  On the destination side, I have an import job configured to restore workloads with exclusions of certain objects.

Issue:

  1. I run the the import job for the 1st time today (within 24 hour clock).
  2. K10 imports the snapshot from the external S3 target.
  3. Then restores the workloads

After some testing, I wipe out all the restored workloads, including the namespace to reset the destination for another test.

When I re-run the import job:

  1. K10 imports the restore point
  2. The job marks as complete
    1. Without restoring any workloads

Question:
How can I re-run the import job multiple times to fully test my DR solution?

icon

Best answer by FRubens 13 November 2023, 23:50

View original

5 comments

FRubens
Userlevel 4
Badge +2

Hello @brawin,

Thanks for using Kasten by Veeam community!

The import policy can be executed as much as needed, and it should restore the workloads again if you have selected to restore workloads after import. I will run some testing to check if there are any issues in K10 latest version on my side and update you.


There is a way to manually run the restore after the import, since you have already executed the import all restore points related with the application should be available in the DR cluster, check on K10 dashboard > applications and under "removed" applications filter you will find the imported application. 

If you are on 6.0.12, click in the 3 dots and restore, after that choose one restore point to restore, in previous K10 versions you can click in the restore button in the application card.


Rubens

B

@FRubens 

btw… Running K10-v6.0.11
the issue with running the restore like you pointed out is I also have to now add back my exclusions to the one-time restore job (possible but a p.i.t.a)

So today, my scheduled import had failed.  This time, I did not perform any object deletions on the kubeadm side (just left as-is).  Tried to run the policy again but same results with only the ‘import’ phase was executed.  It is like the ‘link’ between the steps of ‘import’ and ‘restore’ is not being triggered but cannot find anything that says why.

 

 

B

So did some testing and validated my suspicion.

An import job with the flag to ‘restore after import’ will not trigger the restore phase if the restore point already exists on the local system.

If you want to re-run the ‘entire’ import policy, you need to delete the restore point on the local system.

 

I can see the logic of this from different points.

 

Feature suggestion:

Add a flag within import jobs that will ALWAYS restore workload regardless if the restore already exist or has just been imported.

FRubens
Userlevel 4
Badge +2

Hello @brawin,

Thanks for sharing, I have found the same but I am still checking if this is an expected behaviour in this case.

Will inform here when I have the information.

Rubens

FRubens
Userlevel 4
Badge +2

Hello @brawin,
This is by design, that means if there is no new Restore Points imported, there is no new data to restore, since data was restored from the latest Restore Point imported in the last time.

K10 will trigger the restore only if there is a new Restore Point imported, if you have select this option.

Please let me know if this clarify your question.

Comment


Terms & Conditions