[Yahoo-eng-team] [Bug 1316401] [NEW] Use of MagicMocks causes unnecessary memory use

2014-05-05 Thread Kevin Benton
Public bug reported: There are many locations throughout the code-base that make use of MagicMocks to capture and prevent external calls. However, several of them never have assertions made on them so the fact that they record all of the calls made to them is a waste of memory. This is especially

[Yahoo-eng-team] [Bug 1316382] [NEW] Big Switch: servermanager HTTPS fails on 2.6

2014-05-05 Thread Kevin Benton
Public bug reported: The HTTPS class with certificate validation in the Big Switch server manager module references a source_address attribute that isn't present in the python 2.6 httplib. ** Affects: neutron Importance: Undecided Status: New -- You received this bug notification

[Yahoo-eng-team] [Bug 1316373] [NEW] Can't force delete an errored instance with no info cache

2014-05-05 Thread Sam Morrison
Public bug reported: Sometimes when an instance fails to launch for some reason when trying to delete it using nova delete or nova force-delete it doesn't work and gives the following error: This is when using cells but I think it possibly isn't cells related. Deleting is expecting an info cache

[Yahoo-eng-team] [Bug 1316365] [NEW] NSX: stacktrace during port creation

2014-05-05 Thread Armando Migliaccio
Public bug reported: When using NSX-DHCP the following stack trace can be caused if a port is created with a non-empty device id, on a subnet with disabled dhcp services: 2014-05-05 17:12:22.099 18224 DEBUG neutron.openstack.common.rpc.amqp [req-06840c8e-e091-4627-b221-5faeecd1402f None] Sending

[Yahoo-eng-team] [Bug 1275256] Re: failed to reach SHELVED_OFFLOADED status during tempest-dsvm

2014-05-05 Thread Joe Gordon
** This bug is no longer a duplicate of bug 1276772 Instance deletion fails due to race condition on Hyper-V -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to OpenStack Compute (nova). https://bugs.launchpad.net/bugs/1275256 Titl

[Yahoo-eng-team] [Bug 1254238] Re: tempest.api.compute.servers.test_list_servers_negative.ListServersNegativeTestXML.test_list_servers_by_changes_since failed in gate

2014-05-05 Thread Joe Gordon
** Also affects: tempest Importance: Undecided Status: New -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to OpenStack Compute (nova). https://bugs.launchpad.net/bugs/1254238 Title: tempest.api.compute.servers.test_list

[Yahoo-eng-team] [Bug 1316334] [NEW] Duplicate module-rgx line in .pylintrc

2014-05-05 Thread Adam Harwell
Public bug reported: The module-rgx line in .pylintrc was updated to change from quantum to neutron, but the old line was not removed and is overriding the newer correct line. ** Affects: neutron Importance: Undecided Assignee: Adam Harwell (flux-adam) Status: New ** Changed i

[Yahoo-eng-team] [Bug 1316331] [NEW] ValueError when cookie pagesize is ""

2014-05-05 Thread Matthew D. Wood
Public bug reported: If a user somehow gets the value of 'horizon_pagesize' set to "", the method horizon.utils.functions.get_page_size will raise a ValueError because it can't convert the empty string to an int. Clearly this would also happen if the value gets set to any other non-valid int-lite

[Yahoo-eng-team] [Bug 1316323] [NEW] cloud-init-single-module-fails-to-merge

2014-05-05 Thread Marco Morais
Public bug reported: I have a yaml file in a local directory (cloud-init.yaml) which contains 2 modules (write_files, runcmd). I am exercising the write_files module using the following command: $ sudo cloud-init --debug --file cloud-init.yaml single --name cc_write_files >From the debug output

[Yahoo-eng-team] [Bug 1316234] [NEW] Image delete calls slow due to synchronous wait on backend store delete

2014-05-05 Thread nikhil komawar
Public bug reported: Issue:- --- Currently, the image delete calls can be very slow if the image being delete is large (consisting of multiple chunks esp. in case of swift) or if the backend store has slow response time. Solution:- - Make image delete call asynchronous

[Yahoo-eng-team] [Bug 1316233] [NEW] image data not deleted while deleting image in v2 api

2014-05-05 Thread nikhil komawar
Public bug reported: This is reopen for the bug 1039897 ( https://bugs.launchpad.net/glance/+bug/1039897 ). Description as per the old bugs description. Seems the code has been changed since that got merged in and needs a fix again. ** Affects: glance Importance: Undecided Status:

[Yahoo-eng-team] [Bug 1316226] [NEW] Autocomplete attribute for Workflow forms

2014-05-05 Thread Matthew D. Wood
Public bug reported: Currently, there's no mechanism to set a Workflow's form element's "autocomplete" attribute. This mechanism exists within modal-forms; through a template inheritance mechanism. That approach doesn't work for workflows because the template-inheritance is reversed. One good e

[Yahoo-eng-team] [Bug 1316222] [NEW] Nova host-evacuate lists the target_host as optional parameter though it is a required parameter

2014-05-05 Thread Sangeeta Singh
Public bug reported: steps: - stop nova-compute on the HV you want to evacuate your vms from (e.g. cloudtest.com) - wait until "sudo nova-manage service list" on api host shows the hv is down (nova-compute cloudtest002.com novaenabledXXX 2014-05-02 19:33:42) - nova host-ev

[Yahoo-eng-team] [Bug 1315150] Re: on VM launch Unexpected vif_type=binding_failed

2014-05-05 Thread Shambhu Dayal Rai
It turned out that extra white space in the neutron.conf file was causing the issue. After i removed the whitespace, it works fine.. ** Changed in: neutron Status: New => Invalid -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed

[Yahoo-eng-team] [Bug 1316190] [NEW] When IPv6 addresses are added to the dnsmasq host file, dnsmasq stops issuing leases

2014-05-05 Thread Sean M. Collins
Public bug reported: When an IPv6 subnet is created, with ipv6_ra_mode not set, and ipv6_address_mode set to "slaac" - ipv6 addresses are added to the host file that dnsmasq uses, and this causes dnsmasq to cease responding to dhcp clients on the v4 side. ** Affects: neutron Importance: Unde

[Yahoo-eng-team] [Bug 1287983] Re: hyperv check failure

2014-05-05 Thread Alessandro Pilotti
** Changed in: neutron Status: New => Invalid -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to neutron. https://bugs.launchpad.net/bugs/1287983 Title: hyperv check failure Status in OpenStack Neutron (virtual network serv

[Yahoo-eng-team] [Bug 1316173] [NEW] nova.virt.imagehandler not used

2014-05-05 Thread Jay Pipes
Public bug reported: jaypipes@cranky:~/repos/openstack/nova$ ack-grep load_image_handlers --ignore-dir tests nova/virt/imagehandler/__init__.py 70:def load_image_handlers(driver): jaypipes@cranky:~/repos/openstack/nova$ ack-grep handle_image --ignore-dir tests nova/virt/imagehandler/__init__.py

[Yahoo-eng-team] [Bug 1316167] [NEW] mock.assert_not_called() is not a thing

2014-05-05 Thread Matt Riedemann
Public bug reported: We have 14 hits of tests using mock.assert_not_called() which is not a real method for mocks: nova nova tests api openstack test_common.py 395: href_link_mock.assert_not_called() compute test_compute_utils.py (2 matches) 764: mock_log.warning.assert_not_called() 775: mock_l

[Yahoo-eng-team] [Bug 1291570] Re: "Default" naming for default security group can confuse end users

2014-05-05 Thread Liz Blanchard
** Description changed: It would be great to rename the "Default" security group to be "Default security group" so that when users click on this name as a link in the - table it will be clear that this is a security group. Also, it will help - when they are selecting this security group from t

[Yahoo-eng-team] [Bug 1316152] [NEW] nova cell showing should raise more information on v2 API

2014-05-05 Thread ugvddm
Public bug reported: nova cell showing should raise more information on v2 API, instead of "The resource could not be found. " ** Affects: nova Importance: Undecided Assignee: ugvddm (271025598-9) Status: New ** Changed in: nova Assignee: (unassigned) => ugvddm (271025598

[Yahoo-eng-team] [Bug 1316137] [NEW] glance --location images handling is hiding HTTP errors and uses incorrect HTTP codes...

2014-05-05 Thread Jaroslav Henner
Public bug reported: 1) Glance is hiding error messages when used with --location. If the URL returns HTTP 404, glance is just ignoring it. 2) Reply to GET request to http://CENSORED:9292/v1/images/2e34a168-62ca- 412d-84bb-852fcaf2a391 contains header:    location: http://CENSORED:9292/v1/images

[Yahoo-eng-team] [Bug 1316128] [NEW] Move code in ldap which delete elements of tree to common method

2014-05-05 Thread Sergey Nikitin
Public bug reported: Code which uses 'search_s' and 'delete_s' in ldap backends should be moved to a new method to remove duplication of code. Also should be added handling of ldap.NO_SUCH_OBJECT exception because now it's just ignored. ** Affects: keystone Importance: Undecided Assig

[Yahoo-eng-team] [Bug 1316122] [NEW] GET requests on neutron API do not take into account the tenant specified in the token

2014-05-05 Thread Matthieu Maquevice
Public bug reported: Hey, GET requests on Neutron API (/networks, /floatingips...) return a perimeter of data with multiple tenants, even if the token used in the request specified a tenant id. Is it the normal behavior ? If so, how can we limit our scope to our tenant ? Thanks ! ** Affects: ne

[Yahoo-eng-team] [Bug 1307412] Re: The API v2 reserved property error is not raised when there's nothing to update

2014-05-05 Thread Flavio Percoco
turns out the client is not sending the owner field update, which IMHO, is wrong. These kind of optimizations belong to the server, not the client. I'll move this bug to the client and update the title. ** Project changed: glance => python-glanceclient ** Summary changed: - The API v2 reserved p

[Yahoo-eng-team] [Bug 1316077] [NEW] live migration fails on NoValidHost although the host seems very valid

2014-05-05 Thread Ami Jeain
Public bug reported: I am trying to perform a live-migration of one instance from one server to another, and while it fails, I see in the conductor log file: File "/usr/lib/python2.7/site-packages/nova/scheduler/filter_scheduler.py", line 140, in select_destinations raise exception.NoValidHost(r

[Yahoo-eng-team] [Bug 1316074] [NEW] nova network dhcpbridge has direct DB access

2014-05-05 Thread Hans Lindgren
Public bug reported: nova-network is currently broken due to direct DB access in dhcpbridge. This issue was found using a multihost devstack setup where the non- controller node has an empty sql connection string. ** Affects: nova Importance: High Assignee: Hans Lindgren (hanlind)

[Yahoo-eng-team] [Bug 1314705] Re: image-download fails causes terminal text to gibberish

2014-05-05 Thread Flavio Percoco
This is the expected behavior and it has nothing to do with ssh. `image- download` default's output is stdout. In order to avoid 'gibberishing' the terminal, you should redirect the output to a file. Unfortunately, we can't do anything about the terminal being completely messed up after the image