bulk termination of instances

[Matthias Runge]

while terminating a bulk of instances from the horizon, the first instance termination failed, the other instances all stayed in available status.

Steps to Reproduce:
1. Delete a bulk of instances
2. The first deletion process should fail & the instance status should turn to Error

Actual results:
The other instances are not been terminated

Expected results:
The termination of the instances should continue to the other instances

[Gary W. Smith (gary-w-smith)]

In step 2 above, are you saying that whenever you try to do a bulk termination, that the first one always fails; or are you saying that you are somehow manipulating the instance so that the first one termination intentionally fails?

When I try bulk termination it works properly, so I cannot reproduce the former.
If it is the latter, then please give additional details in how I can force the termination to fail.

[Launchpad Janitor (janitor)]

[Expired for OpenStack Dashboard (Horizon) because there has been no activity for 60 days.]

[Matthias Runge (mrunge)]

The issue is here: bulk termination stops, when an instance is stopped e.g via command line manually after displaying the list of instances.

[Matthias Runge (mrunge)]

If you have 2.g. two persons working in the same project, this can happen quite easily.

I wouldn't call 2 persons stopping a list of instances at the same time manipulating.

[Ritesh Paiboina (rsritesh)]

Hi,
I have tried to reproduce this scenario but could not able to do so.
Have tried some changes in the nova side, to block the delete a particular instances, but during the bulk termination it works as expected [ The termination of the instances should continue to the other instances ]

It would be better to if you could describe more how we can fail the deletion process of the first instances.