[HPE] Failed to create multiple instances with boot volumes at the same time in version 20.0.2.dev11.

Bug #1995863 reported by tomasz
22
This bug affects 4 people
Affects Status Importance Assigned to Milestone
Cinder
New
Low
Unassigned

Bug Description

Description:

Env Info:
HPE 3par 8K
Wsapi: 1.6.15
Image Cache: True

Problem with:
python-3parclient 4.2.12

cinder 20.0.2.dev11

In such a scenario, below errors may occur for some vms:

Error: Failed to perform requested operation on instance "test-10", the instance has an error status: Please try again later [Error: Build of instance 9eb0f656-8401-44b6-9cd0-742bb49ca605 aborted:
Unable to update attachment.(Bad or unexpected response from the storage volume backend API: Driver initialize connection failed (error: Forbidden (HTTP 403) 6 - invalid session key).). (HTTP].

Error list:
Bad or unexpected response from the storage volume backend API: Driver initialize connection failed (error: Forbidden (HTTP 403) 6 - invalid session key).). (HTTP].
or
Exception during message handling: cinder.exception.VolumeBackendAPIException: Bad or unexpected response from the storage volume backend API: Driver initialize connection failed (error: 'NoneType' object has no attribute 'login').
or
Exception during message handling: cinder.exception.VolumeBackendAPIException: Bad or unexpected response from the storage volume backend API: Driver initialize connection failed (error: 'NoneType' object has no attribute 'logout').
or
Exception during message handling: cinder.exception.VolumeBackendAPIException: Bad or unexpected response from the storage volume backend API: Driver initialize connection failed (error: Conflict (HTTP 409) 34 - resource in use).

From Nova perspective:
Instance Status: Error
Volumes Attached To a47805b4-aa00-4e83-80ac-a10e20e865b8 on /dev/vda

From Cinder perspective:
ID: a47805b4-aa00-4e83-80ac-a10e20e865b8
Volume Attached To: Not Attached

Same env/storage with one exception (cinder version):

python-3parclient 4.2.12

cinder 19.1.1

No problem at all.

Changed in cinder:
importance: Undecided → Low
summary: - Failed to create multiple instances with boot volumes at the same time.
+ Failed to create multiple instances with boot volumes at the same time
+ in version 20.0.2.dev11.
tags: added: boot drivers hpe
Eric Harney (eharney)
tags: added: 3par
Revision history for this message
Sofia Enriquez (lsofia-enriquez) wrote :

Hello, we need a 3par driver owners to look at this bug please. Thanks, Sofia

summary: - Failed to create multiple instances with boot volumes at the same time
- in version 20.0.2.dev11.
+ [HPE] Failed to create multiple instances with boot volumes at the same
+ time in version 20.0.2.dev11.
Revision history for this message
J (unrec0gnized) wrote :

Hi, I have the same issue. I've tested also on Zed release and the problem persists.

Revision history for this message
tomasz (chendynski) wrote (last edit ):

Hello,

I would like to add some new information:
If I roll back changes to
Commit:
on Sep 2, 2021 d3e8af0 traghavendra HPE 3PAR: Add HPE Alletra 9k related information

only on two specific files

hpe_3par_base.py
hpe_3par_common.py

Link:
https://github.com/openstack/cinder/tree/d3e8af04a314f102f91e77224b84ea3c21172757/cinder/volume/drivers/hpe

That change omits:
Merge "HPE 3PAR: Reuse existing session"
Move nimble driver code to hpe folder
HPE 3PAR: In multi host env, fix multi-detach operation

Then:
My tests show that the problem is gone.

Revision history for this message
tomasz (chendynski) wrote :
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.