find_multipath_device_path being called needlessly by linuxscsi.extend_volume
Bug #1832247 reported by
Lee Yarwood
This bug affects 1 person
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
os-brick |
Fix Released
|
Undecided
|
Lee Yarwood |
Bug Description
This results in a delay between the SCSI rescan and domain disk block resize being called:
This is currently allowing a race between the kernel, QEMU, libvirtd and n-cpu to lead to failures when running the tempest.
While removing this delay might workaround the above race it isn't the underlying cause.
Changed in os-brick: | |
assignee: | nobody → Lee Yarwood (lyarwood) |
status: | New → In Progress |
To post a comment you must log in.
Reviewed: https:/ /review. opendev. org/664418 /git.openstack. org/cgit/ openstack/ os-brick/ commit/ ?id=9cf135c6864 0f81838e2898b4a ef3e0445770089
Committed: https:/
Submitter: Zuul
Branch: master
commit 9cf135c68640f81 838e2898b4aef3e 0445770089
Author: Lee Yarwood <email address hidden>
Date: Mon Jun 10 19:02:19 2019 +0100
linuxscsi: Stop waiting for multipath devices during extend_volume
Previously linuxscsi. extend_ volume would always attempt to wait for
multipath devices to appear regardless of multipath actually being used
by the connector.
This change corrects this by passing down the existing use_multipath extend_ volume.
attribute from the iSCSI and FC connectors into linuxscsi.
The same attribute is introduced to the NVMe connector to also allow it
to skip this search for multipath devices.
Change-Id: I29d65ae036957f 3a63cba93dd330b 14e3361a1b9
Closes-bug: #1832247