parameters max_tries and wait_between of method ComputeManager._await_block_device_map_created should be configurable

Bug #1280357 reported by Christian Berendt
18
This bug affects 3 people
Affects Status Importance Assigned to Milestone
OpenStack Compute (nova)
Won't Fix
Medium
Unassigned

Bug Description

When using a weak storage backend and initiating the creation of a lot of new instances using volumes as backend (directly created from an image) I got a lot of InvalidBDM: Block Device Mapping is Invalid. After I had a look on the method _await_block_device_map_created (in nova/manager/ComputeManager) the solution was pretty easy: increasing the max_tries and/or wait_between parameters solved the issue. The storage backend could simply not provide this mass of volumes in a very short time (100 seconds on my testing system).

Matt Riedemann (mriedem)
tags: added: compute volumes
melanie witt (melwitt)
Changed in nova:
importance: Undecided → Medium
status: New → Confirmed
Revision history for this message
Nikola Đipanov (ndipanov) wrote :

As discussed on several proposed patches around this (see https://review.openstack.org/#/c/80619/ or https://review.openstack.org/#/c/80619/ which actually rejects this solution).

I will move this bug to "won't fix", and will raise a BP targeted for Juno to use some of the code added in https://blueprints.launchpad.net/nova/+spec/admin-event-callback-api to make interactions between nova and cinder better and avoid the need for a configurable timeout.

Changed in nova:
status: Confirmed → Won't Fix
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.