]> review.fuel-infra Code Review - openstack-build/cinder-build.git/log
openstack-build/cinder-build.git
11 years agoAdded service and endpoint creation in cinder-api.
Thomas Goirand [Thu, 22 Nov 2012 17:13:42 +0000 (17:13 +0000)]
Added service and endpoint creation in cinder-api.

11 years agoAdded upstream changelog
Thomas Goirand [Tue, 20 Nov 2012 13:11:11 +0000 (13:11 +0000)]
Added upstream changelog

11 years agoRemoved duplicate get-orig-source: (it's in pkgos.make)
Thomas Goirand [Tue, 20 Nov 2012 13:07:51 +0000 (13:07 +0000)]
Removed duplicate get-orig-source: (it's in pkgos.make)

11 years agoRemove duplicate Build-Depends: openstack-pkg-tools
Thomas Goirand [Tue, 20 Nov 2012 12:56:42 +0000 (12:56 +0000)]
Remove duplicate Build-Depends: openstack-pkg-tools

11 years agoMove some stuff to pkgos.make
Thomas Goirand [Fri, 16 Nov 2012 14:58:53 +0000 (14:58 +0000)]
Move some stuff to pkgos.make

12 years agoMerge branch 'debian/experimental' of ssh://git.debian.org/git/openstack/cinder into...
Thomas Goirand [Thu, 1 Nov 2012 16:55:49 +0000 (16:55 +0000)]
Merge branch 'debian/experimental' of ssh://git.debian.org/git/openstack/cinder into debian/experimental

12 years agoNow auto-detects and prompts for a volume group using debconf.
Thomas Goirand [Thu, 1 Nov 2012 16:41:11 +0000 (16:41 +0000)]
Now auto-detects and prompts for a volume group using debconf.

12 years agoRemoved set -x
Roland Mas [Tue, 30 Oct 2012 13:16:48 +0000 (14:16 +0100)]
Removed set -x

12 years agoRemoved set-x in debian/cinder-common.postinst.in
Thomas Goirand [Tue, 30 Oct 2012 06:08:10 +0000 (06:08 +0000)]
Removed set-x in debian/cinder-common.postinst.in

12 years agoStart using factored pkgos_* stuff from openstack-pkg-tools
Roland Mas [Mon, 29 Oct 2012 14:33:37 +0000 (15:33 +0100)]
Start using factored pkgos_* stuff from openstack-pkg-tools

12 years ago* Now asking for Keystone credentials using the new pkg_os function.
Thomas Goirand [Fri, 26 Oct 2012 15:31:24 +0000 (15:31 +0000)]
* Now asking for Keystone credentials using the new pkg_os function.
* Patches api-paste.ini for syntax.

12 years agoCopy of pkgos_func from glance.
Thomas Goirand [Fri, 26 Oct 2012 15:18:39 +0000 (15:18 +0000)]
Copy of pkgos_func from glance.

12 years ago * Now uses pkgos_func functions for debconf and dbconfig-common handling.
Thomas Goirand [Fri, 26 Oct 2012 11:07:48 +0000 (11:07 +0000)]
  * Now uses pkgos_func functions for debconf and dbconfig-common handling.
  * Renames /etc/sudoers.d/cinder_sudoers as /etc/sudoers.d/cinder-common.
  * Debconf now users cinder/<debconf-screen-name> and not cinder-common/.
  * Reviewed long and short descriptions in debian/control.
  * Bumped Build-Depends-Indep: python-all to 2.6.6-3~ since we are using dh_python2.
  * Added everyone in the team as Uploaders:.
  * Changed Homepage field to http://cinder.openstack.org/.
  * Dependencies are now by alpha order.
  * Removed useless empty cinder-volume.postrm, cinder-api.postrm,
  cinder-scheduler.postinst, cinder-scheduler.postrm.
  * Pre-Depends: dpkg (>= 1.15.6~) because we use xz compression.

12 years agoFixes wrong handling (eg: policy violation) of /etc/cinder/cinder.conf
Thomas Goirand [Tue, 9 Oct 2012 15:41:23 +0000 (15:41 +0000)]
Fixes wrong handling (eg: policy violation) of /etc/cinder/cinder.conf

12 years agoAdded setuptools_git-*-py2.*.egg in .gitignore
Thomas Goirand [Tue, 9 Oct 2012 09:29:17 +0000 (09:29 +0000)]
Added setuptools_git-*-py2.*.egg in .gitignore

12 years agoAdded a gbp.conf file.
Thomas Goirand [Tue, 9 Oct 2012 09:16:17 +0000 (09:16 +0000)]
Added a gbp.conf file.

12 years agoRe-added ubuntu upstart files using the .in extension.
Thomas Goirand [Tue, 9 Oct 2012 09:15:50 +0000 (09:15 +0000)]
Re-added ubuntu upstart files using the .in extension.

12 years agoAdded a get-vcs-source target and sets compression to use XZ level 9.
Thomas Goirand [Tue, 9 Oct 2012 09:07:29 +0000 (09:07 +0000)]
Added a get-vcs-source target and sets compression to use XZ level 9.

12 years agoNew upstream release 2012.2
Mehdi Abaakouk [Sat, 29 Sep 2012 07:23:55 +0000 (09:23 +0200)]
New upstream release 2012.2

12 years agoMerge tag '2012.2' into debian/experimental
Mehdi Abaakouk [Sat, 29 Sep 2012 07:21:04 +0000 (09:21 +0200)]
Merge tag '2012.2' into debian/experimental

2012.2

12 years agoMerge "Fix NetAppCmodeISCSIDriver._get_lun_handle() method" into milestone-proposed
Jenkins [Wed, 26 Sep 2012 00:48:40 +0000 (00:48 +0000)]
Merge "Fix NetAppCmodeISCSIDriver._get_lun_handle() method" into milestone-proposed

12 years agoMerge "Revert explicit usage of tgt-adm --conf option." into milestone-proposed
Jenkins [Tue, 25 Sep 2012 23:55:17 +0000 (23:55 +0000)]
Merge "Revert explicit usage of tgt-adm --conf option." into milestone-proposed

12 years agoFixes remove_export for IetAdm
MotoKen [Tue, 25 Sep 2012 05:08:34 +0000 (13:08 +0800)]
Fixes remove_export for IetAdm

* Corrects indentation.
* Uses specific iqn as keyword argument.

Fixes bug 1055401

Change-Id: Ie99dd8a34924f0d7f447cf41a305353659a51d11
(cherry picked from commit 5b61536fdf17332f48af4b422fe23c2e153934a1)

12 years agoFix NetAppCmodeISCSIDriver._get_lun_handle() method
Ben Swartzlander [Mon, 24 Sep 2012 16:30:48 +0000 (12:30 -0400)]
Fix NetAppCmodeISCSIDriver._get_lun_handle() method

bug 1055578

Change-Id: I06da6a9a59fa2320f1cae9698e5e63562e534916
(cherry picked from commit b2627b27dae521d719b90831dcc41695b389712f)

12 years agoRevert explicit usage of tgt-adm --conf option.
Eoghan Glynn [Tue, 25 Sep 2012 14:52:38 +0000 (15:52 +0100)]
Revert explicit usage of tgt-adm --conf option.

Fixes bug 1056246.

This reverts commit 0903603971033536751ba40df98a3c39175d3189.

This ensures that we fail-fast if the tgtd configuration does
not include the directory for volume iscsi target configs,
instead of only blowing up when tgtd is restarted.

Change-Id: I30aa3f1d8944eb33cd4a42d65e227031023406b4

12 years agoAdd missing entries in setup, fix up pip-requires
John Griffith [Mon, 24 Sep 2012 18:59:15 +0000 (12:59 -0600)]
Add missing entries in setup, fix up pip-requires

setup.py still has references to nova/compute and was missing some
requires statements, and more importantly didn't hvae the install_requires.

Also, tools/pip-requires needed lxml versions set, as well as
added the missing setuptools_git entry.

This was exposed when trying to run Grenade as the requires.txt
file was never generated.

Fixes bug #1055693

Change-Id: Iaf9669d66e89507989ef3fde630dc2b488a97cfc
(cherry picked from commit 7f00d13640edac8136241e546076f2106f7bfa61)

12 years agoFix cinder-volume-usage-audit
Julien Danjou [Thu, 20 Sep 2012 16:37:36 +0000 (18:37 +0200)]
Fix cinder-volume-usage-audit

It seems many point in the internal API and openstack.common changed, but
this binary has been forgotten.

Also, context ie needed when using the notifier, so import it.

This fixes bug #1053502

Change-Id: I306b8671e7b0ed8c2ce2548a51a4c5e6393e13cb
Signed-off-by: Julien Danjou <julien@danjou.info>
(cherry picked from commit 186143ecc81369c02aca51b81feb976259bcad41)

12 years agoReturn 400 if create volume snapshot force parameter is invalid
Rongze Zhu [Wed, 12 Sep 2012 10:18:31 +0000 (10:18 +0000)]
Return 400 if create volume snapshot force parameter is invalid

Fixes bug #1014689

* Add is_valid_boolstr function in utils.py
* Add force parameter check in SnapshotsController.create()
* Add unittest for invalid force parameter.

Change-Id: I0f64326f33eb4fad1cf384bd825f56f09e935f40
(cherry picked from commit c678b045c9839ed30e13fe5e7e655e3cb2e2c3f9)

12 years agoMerge "Stop double logging to the console" into milestone-proposed
Jenkins [Fri, 21 Sep 2012 05:45:15 +0000 (05:45 +0000)]
Merge "Stop double logging to the console" into milestone-proposed

12 years agoStop double logging to the console
Vishvananda Ishaya [Thu, 20 Sep 2012 16:52:51 +0000 (16:52 +0000)]
Stop double logging to the console

The code in nova for logging added an extra root handler to be
able to log messages to the console during log setup. This handler
was removed in the setup method. The common setup method no longer
removes this handler, so don't create it. Note that there may be
a small period before the logging setup is finished where messages
will not appear.

Fixes bug 1053512

Change-Id: I559cca8ffc2a33beff22025e42a9713487245980
(cherry picked from commit a1cbbe2e9e5d7a5ec55ec1197571cd68fcdbad15)

12 years agoRemove cinder gating hack
Vishvananda Ishaya [Fri, 21 Sep 2012 00:32:56 +0000 (00:32 +0000)]
Remove cinder gating hack

The hack was necessary while we switched the name used for az's
in the gating tests. Now it can be removed

Change-Id: I970205910ac0950ab4a961aa0537eb964d7f0d3c
(cherry picked from commit bc0fb113af225878b4305da7d834673a6b7f5e97)

12 years agoMerge "Set the default availability zone back to nova" into milestone-proposed
Jenkins [Fri, 21 Sep 2012 04:04:03 +0000 (04:04 +0000)]
Merge "Set the default availability zone back to nova" into milestone-proposed

12 years agoSet the default availability zone back to nova
Vishvananda Ishaya [Thu, 20 Sep 2012 16:34:59 +0000 (16:34 +0000)]
Set the default availability zone back to nova

Currently nova and cinder have to have the same set of az's for
the ec2 api to work properly. Therefore, set the default az for
cinder to nova so they will match.

Includes a hack to az to set cinder to nova to work around gate
issues. The hack can be removed once the gate sets the zone to
nova properly.

Fixes bug 1053508

Change-Id: I9c30ccc5c79b9c4e6c1ebeb3e80d0098e1ffbb11
(cherry picked from commit d051ca9cdf828b1e47ef5158c4b83beb9484109e)

12 years agoAdd lun number (0) to model_update in HpSanDriver
John Griffith [Thu, 20 Sep 2012 23:27:13 +0000 (17:27 -0600)]
Add lun number (0) to model_update in HpSanDriver

The HpSanDriver was not setting the lun number in model_update
and as a result default value of 1 was being used.  Trouble is the
Lun number used by LeftHand is 0, so the connect info would be
wrong and fail when trying to attach the volume.

Fixes bug #1053717

Change-Id: If82a66c6a78ab03da09223a7b79abe559a6f702f
(cherry picked from commit f55198570fd59b7a2a4ddd2f5e77b3d03d0a3d64)

12 years agoFixes to the SolarisISCSI Driver
Yosef Berman [Thu, 20 Sep 2012 17:34:24 +0000 (10:34 -0700)]
Fixes to the SolarisISCSI Driver

Even when the san_is_local config option was set to false, the
SolarisISCSIDriver's _execute method was accidentally set to
util.execute by the VolumeDriver's __init__ method.

Fix bug 1053041

Change-Id: Ie3299f5065e2b084644ec2956f0c6ebae4264a49
(cherry picked from commit 3b8a6303e6c04f0c7b45043c2fcd53948910d37e)

12 years agoUpdate MANIFEST.in patch
Mehdi Abaakouk [Thu, 20 Sep 2012 09:48:40 +0000 (11:48 +0200)]
Update MANIFEST.in patch

12 years agoMerge snapshot 6b31057
Mehdi Abaakouk [Thu, 20 Sep 2012 09:39:32 +0000 (11:39 +0200)]
Merge snapshot 6b31057

12 years agoMerge remote-tracking branch 'upstream/milestone-proposed' into debian/experimental
Mehdi Abaakouk [Thu, 20 Sep 2012 09:38:47 +0000 (11:38 +0200)]
Merge remote-tracking branch 'upstream/milestone-proposed' into debian/experimental

12 years agoDon't suppress upstream file on dh_clean
Mehdi Abaakouk [Thu, 20 Sep 2012 09:25:57 +0000 (11:25 +0200)]
Don't suppress upstream file on dh_clean

12 years agodump version to folsom rc1
Mehdi Abaakouk [Thu, 20 Sep 2012 08:51:18 +0000 (10:51 +0200)]
dump version to folsom rc1

12 years agoMerge tag 'folsom-rc1' into debian/experimental
Mehdi Abaakouk [Thu, 20 Sep 2012 08:50:50 +0000 (10:50 +0200)]
Merge tag 'folsom-rc1' into debian/experimental

folsom-rc1

12 years agoAdd rules to generate quickly the manifest.in
Mehdi Abaakouk [Thu, 20 Sep 2012 08:50:11 +0000 (10:50 +0200)]
Add rules to generate quickly the manifest.in

12 years agoSet important path
Mehdi Abaakouk [Thu, 20 Sep 2012 08:49:34 +0000 (10:49 +0200)]
Set important path

12 years agoMerge "Correct IetAdm remove_iscsi_target" into milestone-proposed
Jenkins [Thu, 20 Sep 2012 03:14:43 +0000 (03:14 +0000)]
Merge "Correct IetAdm remove_iscsi_target" into milestone-proposed

12 years agoAdd nova migrate_version check to cinder import
John Griffith [Tue, 18 Sep 2012 19:29:20 +0000 (13:29 -0600)]
Add nova migrate_version check to cinder import

Cinder-manage migrate import only works from Nova/Folsom--->Cinder/Folsom
this change adds an explicit check of the nova migrate_version (133)
to make sure we have all of the volume id/uuid changes that are needed
to be compatable, and presents an error message if that's not the case.

Addresses bug #1052598

Change-Id: I7f32cf35e33320f4348b7bc44a8e92b5716b71d8
(cherry picked from commit 321d32379cd6044378f2f9c32c068de18f514df5)

12 years agoCorrect IetAdm remove_iscsi_target
John Griffith [Wed, 19 Sep 2012 17:18:51 +0000 (11:18 -0600)]
Correct IetAdm remove_iscsi_target

Remove logical units should be done before delete the target while
unexport iSCSI volume. Or ietd will show "No such file or directory"
error.

Fix bug 1043107

Change-Id: Iab5b97044c5a30c788fcabce827923ff82a84b91
(cherry picked from commit f14e955735065a998cd78280661f6d33c850d9e2)

12 years agoChange two remaining rename mistake
Mehdi Abaakouk [Tue, 18 Sep 2012 20:36:06 +0000 (22:36 +0200)]
Change two remaining rename mistake

12 years agoMark Cinder 2012.2 Final as we near RC1
Thierry Carrez [Tue, 18 Sep 2012 18:12:45 +0000 (20:12 +0200)]
Mark Cinder 2012.2 Final as we near RC1

Mark Final=True as we prepare to publish Folsom RC1.

Change-Id: Ie95dd290aef49633586a4b483c3c80dca3e72aa2

12 years agoMerge "Clean up db.volume_create()"
Jenkins [Tue, 18 Sep 2012 16:30:19 +0000 (16:30 +0000)]
Merge "Clean up db.volume_create()"

12 years agoMerge "Fix volume deletion when device mapper is used"
Jenkins [Tue, 18 Sep 2012 08:12:09 +0000 (08:12 +0000)]
Merge "Fix volume deletion when device mapper is used"

12 years agoClean up db.volume_create()
John Griffith [Mon, 17 Sep 2012 21:36:10 +0000 (15:36 -0600)]
Clean up db.volume_create()

 Adding the metadata return to db.volume_create() introduced some
 messy and unnecessary repitition in the code.  This patch
 cleans that up and makes use of existing volume_get functions rather
 than duplicating the code in volume_create.

 This syncs cinder up with the patch as it's been submitted for nova
 in https://review.openstack.org/#/c/10461/

 Addresses bug #1052176

Change-Id: I17aca069fbb0240770613f176430a1b2fcf2b25a

12 years agoMerge "Update quota when deleting volume that failed to be scheduled"
Jenkins [Mon, 17 Sep 2012 20:37:07 +0000 (20:37 +0000)]
Merge "Update quota when deleting volume that failed to be scheduled"

12 years agoFix volume deletion when device mapper is used
Rafi Khardalian [Wed, 5 Sep 2012 05:51:43 +0000 (05:51 +0000)]
Fix volume deletion when device mapper is used

Call dmsetup remove if there is a /dev/mapper/nova--volumes-
element present.

Resolves bug 979020

Change-Id: Iddaaed411a77dda4bd32f9a97687ff17744119eb

12 years agoMerge "Fix FLAGS.volumes_dir help message"
Jenkins [Mon, 17 Sep 2012 18:14:34 +0000 (18:14 +0000)]
Merge "Fix FLAGS.volumes_dir help message"

12 years agoUpdate quota when deleting volume that failed to be scheduled
Zhiteng Huang [Mon, 17 Sep 2012 09:26:09 +0000 (17:26 +0800)]
Update quota when deleting volume that failed to be scheduled

If one volume was failed to get scheduled, removing such volume should
also clean up reservation.
Also when create_volume is ready to send to scheduler, reservation
should be committed no matter backend can successfully create that
volume or not since deleting volume call will do a minus reservation
even on volume with 'error' status.

Fix bug 1052052

Change-Id: Iaec9926024ad5adb38b08b07772095f49b8593c2

12 years agoAdd files that need to be installed
Mehdi Abaakouk [Mon, 17 Sep 2012 09:00:58 +0000 (11:00 +0200)]
Add files that need to be installed

12 years agoAdd debconf po directory and dependency
Mehdi Abaakouk [Mon, 17 Sep 2012 08:12:35 +0000 (10:12 +0200)]
Add debconf po directory and dependency

12 years agoFix error in configuration file
Mehdi Abaakouk [Mon, 17 Sep 2012 08:11:45 +0000 (10:11 +0200)]
Fix error in configuration file

12 years agoClean /var/lib/cinder on purge
Mehdi Abaakouk [Mon, 17 Sep 2012 08:11:14 +0000 (10:11 +0200)]
Clean /var/lib/cinder on purge

12 years agoMispelling in init script arguments
Mehdi Abaakouk [Sun, 16 Sep 2012 08:31:54 +0000 (10:31 +0200)]
Mispelling in init script arguments

12 years agoAdd dbconfig-common support
Mehdi Abaakouk [Sun, 16 Sep 2012 08:30:59 +0000 (10:30 +0200)]
Add dbconfig-common support

12 years agoFix some lintian warning
Mehdi Abaakouk [Sun, 16 Sep 2012 08:29:29 +0000 (10:29 +0200)]
Fix some lintian warning

12 years agoAdd po rules
Mehdi Abaakouk [Sun, 16 Sep 2012 08:28:51 +0000 (10:28 +0200)]
Add po rules

12 years agoRemove update-rc.d, it's already done by DEBHELPER
Mehdi Abaakouk [Sun, 16 Sep 2012 08:27:57 +0000 (10:27 +0200)]
Remove update-rc.d, it's already done by DEBHELPER

12 years agoSync a change to rpc from openstack-common.
Russell Bryant [Fri, 14 Sep 2012 17:50:47 +0000 (13:50 -0400)]
Sync a change to rpc from openstack-common.

This commit pulls in the following change from openstack-common:

    ee44222 Change the qpid_heartbeat default to 60 seconds.

Fix bug 1050661.

Change-Id: I999e16c18227f9aab4dc8c2ff88cd6907c7c8191

12 years agoMerge "Sync notifier changes from openstack-common"
Jenkins [Fri, 14 Sep 2012 03:43:28 +0000 (03:43 +0000)]
Merge "Sync notifier changes from openstack-common"

12 years agoMerge "Add a resume delete on volume manager startup"
Jenkins [Fri, 14 Sep 2012 03:43:06 +0000 (03:43 +0000)]
Merge "Add a resume delete on volume manager startup"

12 years agoMerge "Improve entity validation in volumes APIs"
Jenkins [Fri, 14 Sep 2012 03:41:14 +0000 (03:41 +0000)]
Merge "Improve entity validation in volumes APIs"

12 years agoAdd a resume delete on volume manager startup
John Griffith [Thu, 13 Sep 2012 16:53:52 +0000 (10:53 -0600)]
Add a resume delete on volume manager startup

  Currently if for some reason the volume service was stopped
  during the zero out operation of a volume delete there was
  no way to get the volume removed from the system (it would
  be present in deleting status forever).

  This change adds a simple check of volumes in the DB with status
  of deleting, and if any are found it restarts the delete process
  on them.

  addresses bug #1011150

Change-Id: Id4c4a3bc61f95245ebc6658234b4b88029956562

12 years agoMerge "Add entity body validation helper"
Jenkins [Thu, 13 Sep 2012 16:45:10 +0000 (16:45 +0000)]
Merge "Add entity body validation helper"

12 years agoImprove entity validation in volumes APIs
Mark McLoughlin [Wed, 12 Sep 2012 11:51:40 +0000 (12:51 +0100)]
Improve entity validation in volumes APIs

Fixes bug #1048565

Use the new Controller.is_valid_body() helper to validate the entity
body in various volumes related POST/PUT handlers and return 422
as appropriate.

Change-Id: I04127972981522c1ed81903893396c4f9665bcd3

12 years agoAdd entity body validation helper
Mark McLoughlin [Wed, 12 Sep 2012 11:51:23 +0000 (12:51 +0100)]
Add entity body validation helper

Add a _valid_body() helper for the to avoid repeating the same tests in
multiple methods. Include a check that the entity itself is actually a
dict.

(Cherry-picks part of commit d1ad73ee from Nova)

Change-Id: If8114cc76d68567005c85c803f29e30e034db89a

12 years agoShould've added super().tearDown() in test_iscsi
John Griffith [Thu, 13 Sep 2012 14:50:09 +0000 (08:50 -0600)]
Should've added super().tearDown() in test_iscsi

  Made changes to use tempfiles and implemented teardown,
  but should have also called super().tearDown as well.

  Associated with chage: I9b9a2400

Change-Id: I2e9d1cb7bfbe78323630eb2f3270698a7ce7e4c4

12 years agoAdd debian dir
Mehdi Abaakouk [Thu, 13 Sep 2012 14:14:57 +0000 (16:14 +0200)]
Add debian dir

12 years agoFixes bug 1050135.
Avishay Traeger [Tue, 11 Sep 2012 14:11:41 +0000 (17:11 +0300)]
Fixes bug 1050135.

Regenerate cinder.conf.sample and fix a couple related bugs.

Change-Id: I3e472f9c69ad4888dc15f901f4bc7cd825fb5ba1

12 years agoFix FLAGS.volumes_dir help message
Joe Gordon [Thu, 13 Sep 2012 01:06:34 +0000 (18:06 -0700)]
Fix FLAGS.volumes_dir help message

And other nova/volume/iscsi.py typos

Change-Id: I781a07b4f11b6cf336c278a6994bd5575ae1a70a

12 years agoMerge "Use tmpdir and avoid leaving test files behind"
Jenkins [Thu, 13 Sep 2012 00:16:08 +0000 (00:16 +0000)]
Merge "Use tmpdir and avoid leaving test files behind"

12 years agoMerge "Sync log format changes from openstack-common."
Jenkins [Thu, 13 Sep 2012 00:15:40 +0000 (00:15 +0000)]
Merge "Sync log format changes from openstack-common."

12 years agoMerge "Update rpc from openstack-common."
Jenkins [Thu, 13 Sep 2012 00:15:16 +0000 (00:15 +0000)]
Merge "Update rpc from openstack-common."

12 years agoUse tmpdir and avoid leaving test files behind
John Griffith [Wed, 12 Sep 2012 22:52:02 +0000 (16:52 -0600)]
Use tmpdir and avoid leaving test files behind

 We were just creating a file for testing iscsi persist files and
 weren't cleaning up after the test.

 Change this to use a tmpdir and make sure we leave no tracks.

 Fixes bug #1050086

Change-Id: I9b9a2400a172d52987d76f3132793cb042fd201c

12 years agoMerge "Add volume quota in volume/api.py and olume/manager.py"
Jenkins [Wed, 12 Sep 2012 23:24:55 +0000 (23:24 +0000)]
Merge "Add volume quota in volume/api.py and olume/manager.py"

12 years agoMerge "Fixes bug 1049446."
Jenkins [Wed, 12 Sep 2012 17:40:45 +0000 (17:40 +0000)]
Merge "Fixes bug 1049446."

12 years agoSync log format changes from openstack-common.
Russell Bryant [Wed, 12 Sep 2012 16:34:43 +0000 (12:34 -0400)]
Sync log format changes from openstack-common.

This patch pulls in the following change from openstack-common:

    9f88d84 Improve logging of process id

Change-Id: I6c5be346ed332d14357441b2b147ad1f09157f41

12 years agoUpdate rpc from openstack-common.
Russell Bryant [Wed, 12 Sep 2012 16:30:04 +0000 (12:30 -0400)]
Update rpc from openstack-common.

Fix bug 1049843.

This commit syncs the following commit from openstack-common:

    commit 7e9f72bb28456c912aa80945dbdb8d200f81b462
    Make projects define 'control_exchange'.

This change allows cinder to set a project-specific default of 'cinder'
for the 'control_exchange' option.  The option is now defined in
cinder.flags.

Change-Id: Ia91c172ca0665798d616f9faf5880770074235c8

12 years agoAdd volume quota in volume/api.py and olume/manager.py
Rongze Zhu [Wed, 12 Sep 2012 07:10:53 +0000 (07:10 +0000)]
Add volume quota in volume/api.py and olume/manager.py

Fixes bug #1048158

volume.api.API.create() call QUOTAS.reserve(), then
volume.manager.VolumeManager.create_volume call QUOTAS.commit().
volume.manager.VolumeManager.delete_volume call QUOTAS.reserve() and
QUOTAS.commit().

It also fixes bug #1049459.

Change-Id: I764aeec83cce93595fbe9aa00205a7d8c9e13c35

12 years agoFixes bug 1049446.
Avishay Traeger [Tue, 11 Sep 2012 10:31:01 +0000 (13:31 +0300)]
Fixes bug 1049446.

Removed the "vtype" option which did not work in the driver, and
added the "easytier" option. Added another check to _check_flags
and made sure -warning flag is passed properly. Added unit tests
to check all option variations.

Change-Id: I4e66436f8f62d851a079ec80209fe9d6b27fb5b4

12 years agoMerge "Remove null_kernel option."
Jenkins [Tue, 11 Sep 2012 22:58:09 +0000 (22:58 +0000)]
Merge "Remove null_kernel option."

12 years agoMerge "Revert "Don't zero out snapshot volume on snapshot_delete""
Jenkins [Tue, 11 Sep 2012 22:15:46 +0000 (22:15 +0000)]
Merge "Revert "Don't zero out snapshot volume on snapshot_delete""

12 years agoRevert "Don't zero out snapshot volume on snapshot_delete"
John Griffith [Tue, 11 Sep 2012 21:41:30 +0000 (21:41 +0000)]
Revert "Don't zero out snapshot volume on snapshot_delete"

This reverts commit 1b3322d45fe2c5ed72cc7f8674e5e319928065ad

Turns out that although it's less likely we can still see this issue when zeroing out regular volumes.

12 years agoMerge "Remove unused utils.wrap_exception"
Jenkins [Tue, 11 Sep 2012 17:20:55 +0000 (17:20 +0000)]
Merge "Remove unused utils.wrap_exception"

12 years agoMerge "Nail the pip requirement at 1.1."
Jenkins [Mon, 10 Sep 2012 18:40:17 +0000 (18:40 +0000)]
Merge "Nail the pip requirement at 1.1."

12 years agoMerge "rename nova.pot => cinder.pot, nova.po => cinder.po"
Jenkins [Mon, 10 Sep 2012 18:26:00 +0000 (18:26 +0000)]
Merge "rename nova.pot => cinder.pot, nova.po => cinder.po"

12 years agoNail the pip requirement at 1.1.
Dan Prince [Fri, 7 Sep 2012 01:54:16 +0000 (21:54 -0400)]
Nail the pip requirement at 1.1.

The most recent version of pip (1.2.1 as of a couple days ago) fails
to install the required Cinder .venv packages. Nailing the installed
pip version when using run_tests.sh seems like a reasonable solution
to this issue.

Fixes LP Bug #1047120.

Change-Id: I8d831f747b0d25b5d0f6e74a234e764ff448965b

12 years agoMerge "Don't zero out snapshot volume on snapshot_delete"
Jenkins [Mon, 10 Sep 2012 02:48:37 +0000 (02:48 +0000)]
Merge "Don't zero out snapshot volume on snapshot_delete"

12 years agoMerge "Recent changes to SolidFire API changed iqn format"
Jenkins [Mon, 10 Sep 2012 02:19:56 +0000 (02:19 +0000)]
Merge "Recent changes to SolidFire API changed iqn format"

12 years agoMerge "Prevent from bug #1008866 is reverted"
Jenkins [Sun, 9 Sep 2012 17:45:46 +0000 (17:45 +0000)]
Merge "Prevent from bug #1008866 is reverted"

12 years agoClean up .gitignore
Rongze Zhu [Sun, 9 Sep 2012 10:49:19 +0000 (18:49 +0800)]
Clean up .gitignore

remove:
* nova.egg-info
* .nova-venv
* tools/conf/nova.conf*

Change-Id: I6578a75314b86628a5afbd4c4282fb271b9a830e

12 years agoPrevent from bug #1008866 is reverted
Rongze Zhu [Sun, 9 Sep 2012 07:35:26 +0000 (15:35 +0800)]
Prevent from bug #1008866 is reverted

Fixes bug #1047841.

Commit 2f5360753308eb8b10581fc3c026c1b66f42ebdc (Adds new volume API
extensions) reverted a part of commit
99456bd690445443ae05c0d4fe1ec43ba6090f6f (Send 'create volume from
snapshot' to the proper host), so bug #1008866 is reproduced. I make
API.create_volume to call _cast_create_volume in cinder/volume/api.py,
it Prevent from bug #1008866 is reverted.

Change-Id: I1bf0b7c5fc47da756bce95128f8fd770d14399b0

12 years agorename nova.pot => cinder.pot, nova.po => cinder.po
Hui Cheng [Sun, 9 Sep 2012 05:04:09 +0000 (13:04 +0800)]
rename nova.pot => cinder.pot, nova.po => cinder.po

Fixes Bug #1047195

renamed:    cinder/locale/bs/LC_MESSAGES/nova.po -> cinder/locale/bs/LC_MESSAGES/cinder.po
renamed:    cinder/locale/nova.pot -> cinder/locale/cinder.pot
renamed:    cinder/locale/cs/LC_MESSAGES/nova.po -> cinder/locale/cs/LC_MESSAGES/cinder.po
renamed:    cinder/locale/da/LC_MESSAGES/nova.po -> cinder/locale/da/LC_MESSAGES/cinder.po
renamed:    cinder/locale/de/LC_MESSAGES/nova.po -> cinder/locale/de/LC_MESSAGES/cinder.po
renamed:    cinder/locale/en_AU/LC_MESSAGES/nova.po -> cinder/locale/en_AU/LC_MESSAGES/cinder.po
renamed:    cinder/locale/en_GB/LC_MESSAGES/nova.po -> cinder/locale/en_GB/LC_MESSAGES/cinder.po
renamed:    cinder/locale/es/LC_MESSAGES/nova.po -> cinder/locale/es/LC_MESSAGES/cinder.po
renamed:    cinder/locale/fr/LC_MESSAGES/nova.po -> cinder/locale/fr/LC_MESSAGES/cinder.po
renamed:    cinder/locale/it/LC_MESSAGES/nova.po -> cinder/locale/it/LC_MESSAGES/cinder.po
renamed:    cinder/locale/ja/LC_MESSAGES/nova.po -> cinder/locale/ja/LC_MESSAGES/cinder.po
renamed:    cinder/locale/ko/LC_MESSAGES/nova.po -> cinder/locale/ko/LC_MESSAGES/cinder.po
renamed:    cinder/locale/pt_BR/LC_MESSAGES/nova.po -> cinder/locale/pt_BR/LC_MESSAGES/cinder.po
renamed:    cinder/locale/ru/LC_MESSAGES/nova.po -> cinder/locale/ru/LC_MESSAGES/cinder.po
renamed:    cinder/locale/tl/LC_MESSAGES/nova.po -> cinder/locale/tl/LC_MESSAGES/cinder.po
renamed:    cinder/locale/tr/LC_MESSAGES/nova.po -> cinder/locale/tr/LC_MESSAGES/cinder.po
renamed:    cinder/locale/uk/LC_MESSAGES/nova.po -> cinder/locale/uk/LC_MESSAGES/cinder.po
renamed:    cinder/locale/zh_CN/LC_MESSAGES/nova.po -> cinder/locale/zh_CN/LC_MESSAGES/cinder.po
renamed:    cinder/locale/zh_TW/LC_MESSAGES/nova.po -> cinder/locale/zh_TW/LC_MESSAGES/cinder.po

Change-Id: I3a7894ac90793f270fb27444034b197b7c4f2841

12 years agoDon't zero out snapshot volume on snapshot_delete
John Griffith [Wed, 5 Sep 2012 19:27:48 +0000 (19:27 +0000)]
Don't zero out snapshot volume on snapshot_delete

 When trying to zero out an LVM snapshot on precise the kernel sometimes
 hangs when performing the dd.  Also the dd process itself can take
 an extremely long time even when it does succesfully complete.
 This can be up to 30 minutes for a 1 Gig volume/snapshot.

 I believe this is a kernel specific issue with LVM snapshots.  The
 zeroing process is unreliable and can cause kernel hangs to let's remove it.

 In order to protect against data leakage we'll implement the zeroing process
 on volume creation.  This doesn't seem to have a significant impact and doesn't
 suffer from the same isues tha zeroing out an LVM snapshot does.

 No reason to continue zero on delete, the zero on creation should
 probably be sufficient.

 Doesn't seem to cause any timing issues but need to keep this in mind.

 Fixes bug 1023755

Change-Id: I56209d8e5973cffa997b4ec3e51c3361838386de