[ubuntu-havana-R1.06-50] ECMP: instance cleanup failed after deleting SI

Bug #1333810 reported by Prashant Shetty
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Juniper Openstack
Status tracked in Trunk
R1.1
Won't Fix
Critical
Rudra Rugge
Trunk
Fix Committed
Critical
Rudra Rugge

Bug Description

In ECMP regression we are seeing cleanup failure. We have test which launches 8 instance in a SI.
When we delete SI after test, few instance are not deleted and still have reference to service instance.

Can you please take a look at below setup.

CFGM : [u'nodea35', u'nodea34']
Control Nodes : [u'nodea35', u'nodec53']
Compute Nodes : [u'nodec54', u'nodec55', u'nodec56']
Collector : [u'nodea35', u'nodec53']
WebUI : nodec53
OpenstackUI : nodea34

Logs:

root@nodea35:~/logs/0000_#curl http://127.0.0.1:8095/virtual-machine/01121ba2-5000-41c6-b1a9-15e068eaf91b | python -m json.tool
.
.
"service_instance_refs": [ { "attr": null, "href": "http://127.0.0.1:8095/service-instance/bd54034d-ae76-4768-8013-ec376c8e54b8", "to": [ "ERROR" ], "uuid": "bd54034d-ae76-4768-8013-ec376c8e54b8"
 } ],

.
.

root@nodea35:~/logs/0000_# curl http://127.0.0.1:8095/service-instances | python -m json.tool | more % Total % Received % Xferd Average Speed Time Time Time Current Dload Upload Total Spent
 Left Speed 100 25 100 25 0 0 7916 0 --:--:-- --:--:-- --:--:-- 12500 { "service-instances": [] } root@nodea35:~/logs/0000_#

Revision history for this message
Nagabhushana R (bhushana) wrote :

Ganesha
  can you check this on the latest 1.1 Regression and update..?

Changed in juniperopenstack:
milestone: r1.06-fcs → r1.10-fcs
assignee: nobody → Rudra Rugge (rudrarugge)
assignee: Rudra Rugge (rudrarugge) → Ganesha HV (ganeshahv)
Changed in juniperopenstack:
milestone: r1.10-fcs → none
status: New → Incomplete
Ganesha HV (ganeshahv)
Changed in juniperopenstack:
status: Incomplete → Confirmed
Revision history for this message
Ganesha HV (ganeshahv) wrote :

Seeing the bug again during test_run.
Rudra, can you please take a look?

Changed in juniperopenstack:
assignee: Ganesha HV (ganeshahv) → nobody
assignee: nobody → Rudra Rugge (rudrarugge)
importance: Medium → High
information type: Proprietary → Public
Revision history for this message
Ganesha HV (ganeshahv) wrote :

Seeing the same issue in 1.20-55~icehouse ubutu run as well

tags: added: blocker sanity
Revision history for this message
OpenContrail Admin (ci-admin-f) wrote : A change has been merged

Reviewed: https://review.opencontrail.org/4517
Committed: http://github.org/Juniper/contrail-controller/commit/75a56bf73e2a24e78299c78e81896c204c47cc58
Submitter: Zuul
Branch: master

commit 75a56bf73e2a24e78299c78e81896c204c47cc58
Author: Rudra Rugge <email address hidden>
Date: Tue Nov 11 16:55:10 2014 -0800

Fix service monitor virtual machine manager

- Ensure delete of VMs only if find is succesful
- Fix shared instance-ip delete on VMI delete. Ensure that all
VMIs for the instance IP are deleted before deleting instance-ip

Change-Id: I3297bf4c509d3c66dc5845664f699a771b90c008
Closes-Bug: #1333810

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.