disk cachemodes should be restricted with multiattached volumes
Bug #1788619 reported by
Eric Harney
This bug affects 2 people
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
Cinder |
Invalid
|
High
|
Unassigned | ||
OpenStack Compute (nova) |
Invalid
|
Undecided
|
Unassigned |
Bug Description
If using multiattach, the "writeback" and "unsafe" disk_cachemode options presumably break the semantics that an application writing to a clustered datastore would rely on for data consistency between multiple nodes.
Volumes should not be allowed to attach to multiple instances (multiattach) with unsafe cache modes.
(This may even include writethrough? I'm not sure.)
Changed in cinder: | |
importance: | Undecided → High |
tags: | added: cinder volumes |
To post a comment you must log in.
Have you tested this or just guessing that the libvirt driver in nova isn't doing the right thing? Because multiattach disks are always set to cache mode "none":
https:/ /github. com/openstack/ nova/blob/ 20bc0136d0665ba fdcd379f19389a0 a5ea7bf310/ nova/virt/ libvirt/ driver. py#L423- L426
# Shareable disks like for a multi-attach volume need to have the
conf. driver_ cache = 'none'
# driver cache disabled.
if getattr(conf, 'shareable', False):