Invalid argument in threadgroup.Thread.wait

Bug #1415184 reported by Jonathan Halterman
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
OpenStack DBaaS (Trove)
Expired
Medium
Unassigned
keystonemiddleware
Invalid
Undecided
Unassigned
oslo.concurrency
Expired
Undecided
Unassigned

Bug Description

When using Trove an Invalid argument error was encountered via Oslo's threadgroup.Thread.wait function (called via Keystone middleware). This results in Trove's wsgi process dying and being restarted. See the call stacks at the bottom of:

http://paste.openstack.org/show/162889/

To reproduce in Trove: make an /instances call for a valid tenant id (in the URI) that does not match the valid X-Auth-Token. Example:

http://10.0.1.25:35357/v1.0/tenant-joe/instances

Header: X-Auth-Token: mikes-token

There are probably more straightforward ways to reproduce this, but this is how I encountered it.

--

For trove specifically, my api-paste.ini config:

[filter:authtoken]
paste.filter_factory = keystonemiddleware.auth_token:filter_factory
auth_host = 10.0.1.40
auth_port = 35357
auth_protocol = http
admin_tenant_name = service
admin_user = trove

From trove.conf:

[keystone_authtoken]
admin_user = trove
admin_tenant_name = service
auth_uri = http://10.0.1.40:5000/v2.0
identity_uri = http://10.0.1.40:35357
auth_protocol = http
auth_port = 35357
auth_host = 10.0.140

description: updated
description: updated
Changed in trove:
status: New → Triaged
importance: Undecided → Medium
milestone: none → ongoing
description: updated
description: updated
Revision history for this message
Morgan Fainberg (mdrnstm) wrote :

Based on conversations with Jonathan, I'm marking this as incomplete in Keystonemiddleware. We can't cleanly duplicate it, but if it resurfaces / etc we'll revisit/reopen.

Changed in keystonemiddleware:
status: New → Incomplete
Revision history for this message
Doug Hellmann (doug-hellmann) wrote :

The traceback in http://paste.openstack.org/show/162889/http://paste.openstack.org/show/162889/ makes it look like the wait() call is down inside eventlet somewhere. Am I mis-reading it?

Changed in oslo.concurrency:
status: New → Incomplete
Revision history for this message
Steve Martinelli (stevemar) wrote :

marking this as invalid for keystonemiddleware as it's been months with no feedback and likely not the project causing the error, since the error looks eventlet related..

Changed in keystonemiddleware:
status: Incomplete → Invalid
Revision history for this message
Amrith Kumar (amrith) wrote :

marking incomplete for trove since it has been months without this occurrence or update.

Changed in trove:
milestone: ongoing → none
status: Triaged → Incomplete
Revision history for this message
Launchpad Janitor (janitor) wrote :

[Expired for oslo.concurrency because there has been no activity for 60 days.]

Changed in oslo.concurrency:
status: Incomplete → Expired
Revision history for this message
Launchpad Janitor (janitor) wrote :

[Expired for OpenStack DBaaS (Trove) because there has been no activity for 60 days.]

Changed in trove:
status: Incomplete → Expired
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.