openstack/nova - nova - OpenDev: Free Software Needs Free Tools

Fixes bug 934534.
The launch_index column in the database is used as the ami-launch-index value
provided by the metadata service.
If launch_index is not specified it defaults to 1. This is incorrect for the
single instance case (should be 0) and doesn't correctly handle multi-instance
starts.
This branch changes the default launch_index to 0 (to handle the single
instance case) and adds the launch_index in the schedulers for multi-instance
starts.
Change-Id: Ifc45abf4cd9f50f732ba4a4b68c0a6242a6c9710

bug 897726 Add "attaching" status for volumes and clear state on error
This closes a gap where multiple calls of attach for the same volume
can happen before the volume is actually attached.
Change-Id: I59d22ceda83729c1a455af9994c9ffec1912e23b

Also remove comment which is no longer relevant.
Change-Id: I33a951d08a34510b2a9dbacb9fb3ebf6bee978b4

Change-Id: Ie1f40fcce6ce6af4ab71961c725dcd626eda8aea

* enables hairpin_mode on virtual bridge ports on instance spawn
* adds conntrack DNAT state criteria to fixed/fixed SNAT exception so reflected traffic SNATs
* updates get_interface ElementTree to work with Python 2.6/2.7
* fixes bug 933640
Change-Id: I63b3e91b41898fcffda8a288be503f9b740b4b4e

Addresses Bug 708278
Add check for duplicates in api.py
Add DuplicateVlan to exception.py
Add test to raise DuplicateVlan in test_db_api.py
Add to Authors
Change-Id: I9d68d7b7c886071e38df3c9d7d53724758bdd84c

remove unused methods and classes in nova/virt/libvirt/firewall.py
and releative function tests. now,nova use nwfilter offered by
libvirt only for anti ARP and IP spoofing.other security policy use
iptables.
Change-Id: Ib9866802ef64668e2feba09124bdf7c7fee92f92

Fixes LP Bug #32154.
Change-Id: I9d6ddfedcc39308811ff5264879b45b4847ec4a8

CA filesystem now created by nova-cert
Also changing Author details
Change-Id: I456d985810d6f0312a7ef94b21637f347e933303

These compute API links were broken, previously referring to
Rackspace URLs with s/v1.0/v1.1/
The links now correctly reference long-lived locations on
docs.openstack.org.
Change-Id: I10efca2868a05982a1783688e374060acf1d0a69

Implements blueprint fast-cloning-for-xenserver
	1. use_cow_images flag is reused for xenserver to check if copy on write images should be used.
	2. image-id is used to tag an image which has already been streamed from glance.
	3. If cow is true, when an instance of an image is created for the first time on a given xenserver, the image is streamed from glance and copy on write disk is created for the instance.
	4. For subsequent instance creation requests (of the same image), a copy on write disk is created from the base image that is already present on the host.
	5. If cow is false, when an instance of an image is created for the first time on a host, the image is streamed from glance and its copy is made to create a virtual disk for the instance.
	6. For subsequent instance creation requests, a copy of disk is made for creating the disk for the instance.
	7. Snapshot creation code was updated to handle cow=true. Now there can be upto 3 disks in the chain. The base disk needs to be uploaded too.
	8. Also added a cache_images flag. Depending on whether the flag is turned on on not, images will be cached on the host.
Change-Id: I54838a24b061c134877f3479c925c6ee78da14bc

Add a new nova configuration flag, boolean, enable_instance_password.
When the flag is True (default), existing behavior is unchanged.
When the flag is False, responses from the create or
rebuild API calls don't include the adminPass attribute.
Change-Id: Icb2bd703770f3a39bb1e458dc31e1489d48da7c1

Change-Id: Iad8f66af18eb396f4737cd4ea168edcc77481ee6

fixes bug: 741972
Change-Id: I4a661a13c5baaa79ee647a6c1cedcef29a2642f6

- Syntax is OS-KSEC2:ec2Credentials (not ec2Credentials)
- adds to fix for bug 917408
- Syntax also requires a call to /tokens instead of /ec2tokens
- Changes are compatible (old syntax supported)
- Warning will be issued to log if URL syntax is old
Change-Id: I5ff9d1643122de2f383517e60f15852720c54490

* Updates methods in nova.compute.instance_types to catch
 InstanceTypeNotFound exceptions and re-raise them as ApiError.
 - Previously, the methods in instance_types were raising ApiError in
 response to DBError, but the underlying db calls have since been
 updated to detect empty results and throw more specific exceptions.
 * Updates get_instance_type_by_flavor_id to not try to catch DBError,
 as the underlying db call only ever raises FlavorNotFound.
 get_instance_type_by_flavor_id should not raise ApiError, as callers
 expect the FlavorNotFound exception and know how to handle it.
 * Fixes bug 872618
Change-Id: I8652f04889adb448017afd253ecdcc840fbb660f

Fixes bug 902162
Change-Id: I1b73943aab338bde90b4d47bc015964e9981af5d

Changed the responses to match that of Amazon EC2
Changed test to expect that response
Change-Id: Ia03a0ea663614944067c0a0193058ba91e149ad2

This change is intended to make log messages including instance
uuids consistent. This eases operational support, as when a
customer asks for the history of there instance it is a simple
grep operation, instead of the current situation. I have included
a sample use, but will work through adding this to relevant log
messages over time.
Change-Id: I1f61c04f32dbb960471950ac7231313d9d9ced12

* Adds new SolidFire driver that subclasses nova.volume.san.SanISCSIDriver
* Adds unit tests for new driver
* Adds new exception subclasses in nova.exception
* Adds John Griffith to Authors
Implements solidfire-san-iscsidriver
Change-Id: I4dc7508ba08f5333cde74d4cfeaae3939c5d2b02

1. Provides dom0 IPtables driver to implement the Provider firewall rules.
 2. Existing libvirt code has been refactored to reduce the amount of duplicated code to a minimum
 3. The three provider apis in ec2/admin.py file are now fixed the following way:
 a. remove_external_address_block returned 'OK' on removing blocks which didn't exist. This is now fixed.
 b. block_external_addresses raised exception earlier on duplicate network blocks. Now the exception is logged and failed status message is returned.
 c. all the three provider apis now logs for invalid and improper inputs and return uniform (a dictionary ) and proper status messages for all cases.
 4. appropriate unit tests added to cover the same
Change-Id: I27d83186f850423a6268947aed0c9a349d8f8d65

Implements blueprint start-stop-methods-support-in-os-servers-xapi
Change-Id: I7aa241ffdd2ec3a7c9800411268f4adc47960be8

Implements blueprint apache-qpid-rpc.
This patch adds a new implementation of the nova.rpc API that uses Qpid
as the messaging backend. We first considered trying to add Qpid
support to kombu, but the kombu API did not map very well to Qpid, which
is based on a newer version of AMQP. It was easier to just map the nova
rpc API to Qpid.
For more information about how to use Qpid with Nova, see this wiki
page:
 http://wiki.openstack.org/QpidSupport
The structure of impl_qpid is largely based on impl_kombu, but adapted
to use the Qpid APIs as necessary. This patch also factors out some of
the code shared with impl_kombu into nova.rpc.common to try to cut down
on duplicated code.
The unit tests were written from scratch instead of taking advantage
of the common rpc unit tests. The common unit tests only work if qpidd
is running, but they do pass if enabled. The unit tests for impl_qpid
instead use mox to mock out the Qpid objects and ensure the right
operations happen on them when the rpc API is exercised.
This patch was a joint effort between myself and William Henry, which is
why he was added to the Authors file in this patch.
Change-Id: Ibacaa956e016ef96f014443074e2a4622e31f090

This allows an extension to add new actions for create and delete
operations to an existing collection.
Currently when extending an existing collection (API namespace), an
extension can register brand new 'actions', and extend any preexisting API
functionality, but unfortunately create and delete are special cases that
don't fall under the 'actions' classification, meaning the infrastructure
can't handle them unless extending an existing impl.
Stubbing out the create/delete methods in the original resource with
the equiv of a NotImplementedError doesn't work, since 'extend'ing
requires the original implementation to exit correctly.
Whitelist 'create' and 'delete' and handle them appropriately.
v2:
 Add myself to Authors
 Update wsgi.action docs
 Add a test case
v3:
 Actually update Authors
Change-Id: I550ae93c5e200f18644042ac81656bc76dbe8955

This commit introduces sub for a host maintenance_mode option into
the osapi. It is required for the host aggregates work.
This is part of a series of commits that have started with change:
https://review.openstack.org/#change,3035
Change-Id: Ib462ef9ad6b641bf1083cc176dfba6645020ccec

Add configuration sanity to QuantumManager. Logs error message
if a parameter not used by QuantumManager's create_networks()
method is passed into it.
Change-Id: I1d6daa7f77590a61b542122ada33ed1c188df19f

I've Updated the calls to nova.virt.libvirt.utils.fetch_image() to match the rest of the class.
Additionally, nova.virt.images.fetch() currently makes no use of the user/project id supplied.
Change-Id: I2db4b8c1b6542909fcc2b452fc961e22fe2557bf

Fixed bug #877417
Change-Id: Iddb9a6d4eb798b0f41f0ef1ab01cb02d16d16dbe

Fixes Bug #917003
The latest version of greenlet is installed in install_venv.py while
the version required is set to 0.3.1 in pip-requires.
Change-Id: I145deb46a8edc41e87634dc5ca839baf5acd8065

Fix based on revewer's comment
upgraded the migration version
nova/db/sqlalchemy/migrate_repo/versions/069_block_migration.py
rebase on master
Change-Id: Ia762f8dec761c3d595bc6fcd39f127f6d92306d2

Rather than just giving users the sudo password prompt immediately,
actually tell them what is about to be installed, so they know
whether it is reasonable to be entering their password.
Change-Id: Ic0c1de812be119384753895531a008075b13494e
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>

fixes bug #890099
Change-Id: I2b516c4b41d387846f8edb9a7b82f3016f0e0eac

Fixes bug #910008
Change-Id: I7914fe1eaffeb0be5feb414f3ea0ad563a36c21f

Change-Id: I4c7ab8512fe3a00dd555d3a183b2c2348cef9e21

Implements import_key_pair instead of import_public_key.
Change-Id: I256b76aeb4ff94030a58692bfd12b2d7c53ae461

A foreign key constraint needs to be removed in order to succeed in the
DB migration. An earlier submitted fix for this uses a mysql drop
statement. I think it's cleaner to use sqlalchemy to do this, as it's
done this way in other version migrations as well. Fix for bug 907254.
Update: IBM from 1928 called, they want their 80 CPL limit
back. - on a more serious note, fixed jenkins PEP8 failure
Update2: Replaced tab by a spaces
Update3: Don't do Foreign Key stuff if the engine is sqlite
Update4: Add myself to the Authors file
Update5: With the correct mail address
Change-Id: Ib021e0ddb2c80ee63888435a9e3761c053534160

Rather than providing two mutually exlusive image
access methods (loop and qemu-nbd), try each in turn.
This is to prepare for a follow up patch which will
add libguestfs as a method to try.
* nova/virt/mount.py: A new Mount class to abstract the
devce allocation, partition mapping and file sys mounting,
for each access type.
* nova/virt/disk/loop.py: A specialization of the base Mount class
to provide loop back mounting support.
* nova/virt/disk/nbd.py: A specialization of the base Mount class
to provide qemu-nbd mounting support.
* nova/virt/disk/api.py: A new file containing the nova.virt.disk
module interface.
(img_handlers): A new list of access methods to try,
with the order being honored.
(_DiskImage): An internal helper class that uses the plugin classes
above, to provide the operations available on a disk image file.
When mounting, iterate over each access method until one succeeds.
If a hint is provided about a CoW format image, the list of
methods to try will be reduced accordingly.
Note expected errors are no longer raised as exceptions during mounting.
Instead, on failure to mount an image, errors are collated and raised.
Interveining errors are logged in debug mode for successful mounts.
* nova/virt/libvirt/connection.py: Adjust the function parameter
names to be more general, rather than referencing specific
implementations like 'nbd' and 'tune2fs'.
Simplify the destroy_container() by storing and passing
back a reference to the _DiskImage object, which has the
necessary state to unmount.
* nova/utils.py (trycmd): A helper function to both deal with,
commands that issue ignorable warnings to stderr,
and commands that EXIT_SUCCESS while issuing errors to stderr.
nova/virt/xenapi/vm_utils.py: Adjust for the moved virt.disk package
Change-Id: If3a4b1c8f4e2f2e7300a21071340dcc839cb36d7