March 2015 - Kimchi-devel - oVirt List Archives

RFC - Guest memory usage
by Daniel Henrique Barboza 27 Oct '15

27 Oct '15

I was looking at the following item in the backlog: Guests Stats: Display memory utilization (use virt-df or virt-top or ...) If I understood it right, the idea here is to show the -inner- memory allocation of the guest. If you have a VM with 4Gb of RAM running an Ubuntu, we want to know how much memory the Ubuntu OS and its processes are using. I've done an investigation and I haven't found any tool to accomplish this. "virt-top", "virsh dommemstat" and the libvirt API retrieves the information of the memory usage of the guest relative to the host. In the example mentioned before, supposing that the host has 64Gb of RAM, all these tools would show that the VM is using 12% of the host RAM. They do not dive in the VM and shows the actual mem usage of the Ubuntu and its processes running there. Haven't found anything useful in other MLs and forums. The common answer is 'run top in a terminal inside the VM', which of course does not suit us. My question is: any thoughts about how we can implement this feature? Because I am starting to think that, in the end, this kind of info is strict to the guest OS and can't be polled from the outside. Thanks!

5 5

[RFC] Split Kimchi into Base Framework and Virtualization layers
by Lucio Correia 05 May '15

05 May '15

Hi, Gustavo (on cc) and me have been analyzing how to split Kimchi Virtualization layer form its Base web framework. We would like to leverage all infrastructure kimchi base framework provides (URI routing, error handling, i18n, authentication, API validation, plugin tabs loading) in other projects, and contribute this back to kimchi-ginger project. The general idea for the web framework is that it provides basic UI with login and tabs menu, and each plugin installed extends the menu with some tabs. We started by making kimchi a plugin, like ginger is. We are way advanced on that and already have most of kimchi's UI running as a plugin over an almost fully running base web framework. Ginger needs minimal changes, mainly on import statements. We can provide changes needed back to ginger too. Currently, we are splitting kimchi specific stuff like this way: (after renaming the dir src/kimchi to src/backend *) * we are using the name "backend" just for disambiguation from virtualization layer (plugins/kimchi). We will need to define a better name for that basic framework. One suggestion is to call it "wok", the base where kimchi is cooked. {src/backend => plugins/kimchi}/API.json | 0 {src/backend/control/vm => plugins/kimchi/control}/Makefile.am | 7 +- {src/backend/control/vm => plugins/kimchi/control}/__init__.py | 2 +- {src/backend => plugins/kimchi}/control/config.py | 0 {src/backend => plugins/kimchi}/control/cpuinfo.py | 0 {src/backend => plugins/kimchi}/control/debugreports.py | 0 {src/backend => plugins/kimchi}/control/groups.py | 0 {src/backend => plugins/kimchi}/control/host.py | 0 {src/backend => plugins/kimchi}/control/interfaces.py | 0 {src/backend => plugins/kimchi}/control/networks.py | 0 {src/backend => plugins/kimchi}/control/peers.py | 0 {src/backend => plugins/kimchi}/control/storagepools.py | 0 {src/backend => plugins/kimchi}/control/storageservers.py | 0 {src/backend => plugins/kimchi}/control/storagevolumes.py | 0 {src/backend => plugins/kimchi}/control/templates.py | 0 {src/backend => plugins/kimchi}/control/users.py | 0 {src/backend => plugins/kimchi}/control/vm/Makefile.am | 0 {src/backend => plugins/kimchi}/control/vm/__init__.py | 0 {src/backend => plugins/kimchi}/control/vm/hostdevs.py | 0 {src/backend => plugins/kimchi}/control/vm/ifaces.py | 0 {src/backend => plugins/kimchi}/control/vm/snapshots.py | 0 {src/backend => plugins/kimchi}/control/vm/storages.py | 0 {src/backend => plugins/kimchi}/control/vms.py | 0 {src/backend => plugins/kimchi}/disks.py | 0 {src/backend => plugins/kimchi}/imageinfo.py | 0 {src/backend => plugins/kimchi}/iscsi.py | 0 {src/backend => plugins/kimchi}/isoinfo.py | 0 {src/backend => plugins/kimchi}/kvmusertests.py | 0 {src/backend => plugins/kimchi}/mockmodel.py | 0 {src/backend/control/vm => plugins/kimchi/model}/Makefile.am | 7 +- {src/backend/control/vm => plugins/kimchi/model}/__init__.py | 8 --- {src/backend => plugins/kimchi}/model/config.py | 0 {src/backend => plugins/kimchi}/model/cpuinfo.py | 0 {src/backend => plugins/kimchi}/model/debugreports.py | 0 {src/backend => plugins/kimchi}/model/diskutils.py | 0 {src/backend => plugins/kimchi}/model/featuretests.py | 0 {src/backend => plugins/kimchi}/model/groups.py | 0 {src/backend => plugins/kimchi}/model/host.py | 0 {src/backend => plugins/kimchi}/model/hostdev.py | 0 {src/backend => plugins/kimchi}/model/interfaces.py | 0 {src/backend => plugins/kimchi}/model/libvirtconnection.py | 0 {src/backend => plugins/kimchi}/model/libvirtstoragepool.py | 0 plugins/kimchi/model/model.py | 93 ++++++++++++++++++++++++++++ {src/backend => plugins/kimchi}/model/networks.py | 0 {src/backend => plugins/kimchi}/model/peers.py | 0 {src/backend => plugins/kimchi}/model/storagepools.py | 0 {src/backend => plugins/kimchi}/model/storageservers.py | 0 {src/backend => plugins/kimchi}/model/storagetargets.py | 0 {src/backend => plugins/kimchi}/model/storagevolumes.py | 0 {src/backend => plugins/kimchi}/model/templates.py | 0 {src/backend => plugins/kimchi}/model/users.py | 0 {src/backend => plugins/kimchi}/model/utils.py | 0 {src/backend => plugins/kimchi}/model/vmhostdevs.py | 0 {src/backend => plugins/kimchi}/model/vmifaces.py | 0 {src/backend => plugins/kimchi}/model/vms.py | 0 {src/backend => plugins/kimchi}/model/vmsnapshots.py | 0 {src/backend => plugins/kimchi}/model/vmstorages.py | 0 {src/backend => plugins/kimchi}/osinfo.py | 0 {src/backend => plugins/kimchi}/scan.py | 0 {config => plugins/kimchi/ui}/Makefile.am | 0 {config/ui => plugins/kimchi/ui/config}/Makefile.am | 0 config/ui/tabs.xml => plugins/kimchi/ui/config/tab-ext.xml | 0 {src/backend => plugins/kimchi}/vmtemplate.py | 0 The remaining generic stuff on the "backend" side, usable or already in use by other plugins, would be: backend |___model | |___model.py | |___plugins.py | |___tasks.py |___control | |___base.py | |___plugins.py | |___tasks.py | |___utils.py |___xmlutils | |___network.py | |___disk.py | |___utils.py | |___graphics.py | |___qemucmdline.py | |___interface.py |___asynctask.py |___auth.py |___basemodel.py |___cachebust.py |___distroloader.py |___exception.py |___i18n.py |___netinfo.py |___network.py |___objectstore.py |___proxy.py |___repositories.py |___rollbackcontext.py |___root.py |___screenshot.py |___server.py |___sslcert.py |___swupdate.py |___template.py |___utils.py |___vnc.py We don't want to make this a too long e-mail and lose the focus on the idea. We have further technical details to discuss in a follow up. Please let us know your thoughts on that, any directions and requirements you have on this, and how could we proceed with that plan. -- Lucio Correia Software Engineer IBM LTC Brazil

2 3

[RFC] Do not block operations due to ref_cnt
by Crístian Viana 03 Apr '15

03 Apr '15

Hi, I'd like to propose a change to how Kimchi uses the resource field "ref_cnt". Currently, "ref_cnt" - which stands for "reference count" - is one of the fields returned when looking up a storage volume. Its purpose is to indicate how many times that resource is being used at the moment. For example, if the resource /storagepools/pool/storagevolumes/vol has ref_cnt=1, it means that the disk is attached to 1 VM right now and thus it cannot be attached to another VM. I believe the original idea of this feature is to prevent the same resource from being attached more than once at the same time. However, IMO, that might not always be the desired behavior and there's no way to enforce it completely as those resources can be used outside of Kimchi, where "ref_cnt" doesn't exist. For example, if I have one VM which uses the disk "vol", I'm not able to attach it to another VM via Kimchi; but if I use another libvirt-based VM manager (e.g. virsh), I am able to attach that disk to a different VM. This becomes even trickier when we consider other operations, like snapshots, which can attach/detach disks while they're being reverted to. Also, suppose I might want to inspect one VM's disk from another VM, and then I'd need to attach one disk twice; Kimchi wouldn't allow that by stating that the disk is already in use. I propose Kimchi should stop using "ref_cnt" as a blocking method. The field may still exist for information/warning messages (e.g. "are you sure you want to attach this disk? it's already being used by another VM.") but no operation should be blocked because of it, as it is the case now. As inconsistencies with that value may happen and we have no way to make sure it will always work, we shouldn't annoy the user by stopping them from doing something that may be perfectly valid. Any feedback will be very welcome. Best regards, Crístian.

3 3

[kimchi-devel][PATCHv3 0/6] Upload storage volume
by lvroyce＠linux.vnet.ibm.com 03 Apr '15

03 Apr '15

From: Royce Lv <lvroyce(a)linux.vnet.ibm.com> known issue: When using api to send requests without setting up session, cherrypy server will report error after hundred of calls, because race condition below: https://bitbucket.org/cherrypy/cherrypy/pull-request/50/fix-race-condition-… But it works well with session setup. v3>v1, Use libvirt upload api instead of r/w file Use lock to synchronize upload to prevent Royce Lv (6): Update docs and json schema of storage volume upload Update controller to make update accept formdata params Add lock facility for storage volume upload Update model for storage volume update Fix incomplete record when uploading update test case for storage volume upload docs/API.md | 7 ++-- src/kimchi/API.json | 19 +++++++++++ src/kimchi/control/base.py | 6 ++-- src/kimchi/i18n.py | 4 +++ src/kimchi/isoinfo.py | 5 ++- src/kimchi/mockmodel.py | 8 ++--- src/kimchi/model/storagevolumes.py | 61 +++++++++++++++------------------- src/kimchi/model/utils.py | 46 ++++++++++++++++++++++++++ tests/test_model.py | 67 ++++++++++++++++++++++++++++++++++++++ 9 files changed, 177 insertions(+), 46 deletions(-) -- 2.1.0

3 20

[PATCH] RFC patch to make nginx proxy optional
by Julien Goodwin 03 Apr '15

03 Apr '15

Sending this as an RFC patch, I expect people will want changes from this. Implements Issue #570. Julien Goodwin (1): Initial prototype, make nginx proxy optional. docs/Makefile.am | 1 + docs/apache.conf.ex | 35 +++++++++++++++++++++++++++++++++++ src/kimchi.conf.in | 3 +++ src/kimchi/config.py.in | 1 + src/kimchi/proxy.py | 6 ++++++ 5 files changed, 46 insertions(+) create mode 100644 docs/apache.conf.ex -- 2.1.4

4 18

[RFC] New API to create a random guest console password
by Aline Manera 02 Apr '15

02 Apr '15

Today user may set/change the guest console password and also its expiration time through Kimchi API. When passing an empty password, a random password is automatically generated. curl -u <user:password> -H "Content-Type: application/json" -H "Accept: application/json" http://localhost:8010/vms/blah -X PUT -d'{"graphics": {"passwd": ""}}' That way is difficult to handle when user wants to reset the guest password. We have a similar issue when we automatically change the passwdValidTo when it is expired - increasing it in 30 seconds. My proposal is simple: only change "passwd" and "passwdValidTo" when user wants to do it. curl -u <user:password> -H "Content-Type: application/json" -H "Accept: application/json" http://localhost:8010/vms/blah -X PUT -d'{"graphics": {"passwd": "123456", "passwdValidTo": "<some datetime format>"}}' curl -u <user:password> -H "Content-Type: application/json" -H "Accept: application/json" http://localhost:8010/vms/blah -X PUT -d'{"graphics": {"passwdValidTo": "<some datetime format>"}}' And make sure the passwdValidTo is only acceptable when there is a passwd set. And to reset those values, we only need to send an empty string: curl -u <user:password> -H "Content-Type: application/json" -H "Accept: application/json" http://localhost:8010/vms/blah -X PUT -d'{"graphics": {"password": "", "passwdValidTo": ""}}' And create a new API: POST /vms/blah/ticket to automatically generate a random password valid only for 30 seconds. curl -u <user:password> -H "Content-Type: application/json" -H "Accept: application/json" http://localhost:8010/vms/blah/ticket -X POST -d'{}' What do you think about it? Regards, Aline Manera

2 3

[RFC] Hotplug a new VM disk
by Crístian Viana 01 Apr '15

01 Apr '15

Hi, I'd like to propose a new feature to hotplug a new disk to a VM. This feature will be implemented only on the UI side. The backend already has the methods needed to make it work. Suppose a user wants to add a new disk to a running VM. Currently, they have to create an empty disk using other tools outside of Kimchi and put it into an existing storage pool. This might be challenging for some users, as it requires the use of other tools, possibly involving the command line (e.g. qemu-img). With this new feature, the user will only select the storage pool and the size of the new disk they want to add; Kimchi will create the disk in the appropriate pool and attach it to the VM. In order to perform this feature, the UI will have to be updated to allow the following steps: 1) Select an existing VM, "Actions" > "Edit". 2) Switch to the tab "Storage". 3) Click on the button "+". 4) On "Device type", select "disk" . 5) On "Storage pool", select any storage pool. *6) On "Storage volume", select "New...".** **7) Enter the desired disk size.* 8) Click on the button "Attach". Steps 6 and 7 are the ones related to this feature and they need to be implemented. The other steps already exist in the current UI. When the user performs step 8, the UI should send the following requests to the server: POST /storagepools/<pool>/storagevolumes {'name': <vol>, 'capacity': <size>} POST /vms/<vm>/storages {'name': <vol>} Those requests already work, they just have to be executed in that order in order to perform this new feature. Any feedback will be very welcome. Best regards, Crístian.

2 1

[RFC] Hotplug a VM network interface
by Crístian Viana 01 Apr '15

01 Apr '15

Hi, I'd like to propose a new feature to hotplug a network interface to a running VM. A similar feature is already available for stopped VMs by selecting a VM and clicking on "Actions" > "Edit", tab "Interface"; the user is able to add or remove a network interface in that dialog. However, if the VM is running, that tab disables the buttons to add and remove network interfaces. This feature will allow those two operations to be performed also on running VMs. The API for adding and removing network interfaces to a VM already exist: POST /vms/<vm>/ifaces {<iface data>} DELETE /vms/<vm>/ifaces/<iface> However, they currently raise an exception when the VM is running. This feature will use those same commands in order to hotplug a network interface, keeping the existing REST interface. Any feedback will be very welcome. Best regards, Crístian.

2 1

[RFC] Add suport for VM suspend/resume
by Crístian Viana 01 Apr '15

01 Apr '15

Hi, I'd like to propose two new operations on virtual machines: *suspend* and *resume*: Suspend ---------- From libvirt.org: /Suspend an active domain, the process is frozen without further access to CPU resources and I/O but the memory used by the domain at the hypervisor level will stay allocated./ REST syntax: POST /vms/<vm-name>/suspend Return status: 200: OK 4xx: error Return data: None Resume ---------- From libvirt.org: /Resume a suspended domain, the process is restarted from the state where it was frozen by calling "suspend"./ REST syntax: POST /vms/<vm-name>/resume Return status: 200: OK 4xx: error Return data: None -------------------- This feature has been partially implemented in previous patches ("[PATCH 0/3] VM suspend/resume", from Jan 21st), and its progress should be resumed soon. The UI will have to be updated accordingly by adding buttons to suspend and resume a VM, somewhat like the buttons which start and stop a VM. Any feedback will be very welcome. Best regards, Crístian.

2 1

[PATCH v2 0/3] Fix issue #461 (backend)
by Crístian Viana 01 Apr '15

01 Apr '15

This is the changelog between this and the previous patchset: - Don't allow mirrorlist and metalink to be specified at the same time; according to the YUM documentation, that situation is invalid. Crístian Viana (3): Handle empty variables when updating YUM repository Use more generic message in repo mirror list error issue #461: Add 'metalink' support for YUM repositories src/kimchi/API.json | 5 +++++ src/kimchi/i18n.py | 6 ++++-- src/kimchi/repositories.py | 31 +++++++++++++++++++++++++------ 3 files changed, 34 insertions(+), 8 deletions(-) -- 2.1.0

2 4

[PATCH V2] Update host number of cpus and total physical memory
by Jose Ricardo Ziviani 01 Apr '15

01 Apr '15

- The number of cpus and the total physical memory information displayed in the host section can change, so them cannot be static. This commit updates those values whenever a GET request hits the backend looking for host information. --- src/kimchi/model/host.py | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/src/kimchi/model/host.py b/src/kimchi/model/host.py index 4419bb3..588bb28 100644 --- a/src/kimchi/model/host.py +++ b/src/kimchi/model/host.py @@ -87,8 +87,8 @@ class HostModel(object): res['cpu_model'] = line.split(':')[1].strip() break - res['cpus'] = psutil.NUM_CPUS - res['memory'] = psutil.TOTAL_PHYMEM + res['cpus'] = 0 + res['memory'] = 0L # Include IBM PowerKVM name to supported distro names _sup_distros = platform._supported_dists + ('ibm_powerkvm',) @@ -102,6 +102,8 @@ class HostModel(object): return res def lookup(self, *name): + self.host_info['cpus'] = psutil._psplatform.get_num_cpus() + self.host_info['memory'] = psutil.phymem_usage().total return self.host_info def swupdate(self, *name): -- 1.9.1

3 3

[PATCH V3] Number of CPUs and Physical Memory
by Jose Ricardo Ziviani 01 Apr '15

01 Apr '15

V3: psutil has changed its interface to retrieve the number of CPUs, this new V3 updates the code while supporting the old version (still needed). Jose Ricardo Ziviani (1): Update host number of cpus and total physical memory src/kimchi/model/host.py | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) -- 1.9.1

2 2

[PATCH][RFC] Kimchi and relative URLs
by Frédéric Bonnard 01 Apr '15

01 Apr '15

From: Frederic Bonnard <frediz(a)linux.vnet.ibm.com> Hi, many people and distros use a subdirectory in the configuration of the system's webserver to configure different websites. The usual way to setup a website is thus to : 1. use the distro's webserver to serve all the websites 2. have a subdirectory to copy the configuration file that makes available a website by only restarting the webserver (and removing the website is just done by removing the configuration file and restart). Concerning this point, the classical configurations can be : a) a sub web location : https://website.org/webapp b) a virtual host : http://webapp.website.org a) and b) correspond each to different configuration files that are just dropped into : - Ubuntu/Debian : /etc/{apache,nginx,..}/sites-available/{webapp_subweb,webapp_virthost}.conf - Fedora/Opensuse : /etc/{httpd,nginx,..}/conf.d/ At the moment kimchi is launched with a private instance of nginx, not the system's installed one and it would be nice to have this improved and this is covered in another thread : http://lists.ovirt.org/pipermail/kimchi-devel/2015-February/009642.html Provided that previous patch, one can use the distro's webserver to run proxy kimchi and that fullfills 1. Then for 2. one need a configuration file for either b) (provided for apache in the link above ; I have also one for nginx ) and a), which I provide in this patch. But this nginx configuration file is not enough for a sub web location configuration as kimchi relies on absolute path based on / from the http://server/. I tried to make URLs paths relative so that kimchi doesn't have to know where it has been placed on the webserver, for example https://server/kimchi/ which will probably be the URL used in distros. I'm not a web developer and I tried to modify all failing requests with a a) configuration with relative paths in mind. So I'd like comments on the way this is done, and this may need more extensive testing (paths I didn't test). Though I used that code back on a virtualhost configuration and it worked as well. Thanks for your help, F. Frederic Bonnard (1): Making urls relative docs/Makefile.am | 1 + docs/nginx.conf.subsite.ex | 37 +++++++ src/kimchi/screenshot.py | 2 +- ui/css/theme-default/template_add.css | 20 ++-- ui/css/theme-default/topbar.css | 2 +- ui/js/src/kimchi.api.js | 176 +++++++++++++++++----------------- ui/js/src/kimchi.login.js | 2 +- ui/pages/guest.html.tmpl | 2 +- ui/pages/help/dita-help.xsl | 4 +- ui/pages/kimchi-ui.html.tmpl | 4 +- ui/pages/storagepool-add.html.tmpl | 2 +- ui/pages/tabs/storage.html.tmpl | 2 +- ui/pages/template-add.html.tmpl | 2 +- 13 files changed, 147 insertions(+), 109 deletions(-) create mode 100644 docs/nginx.conf.subsite.ex -- 1.9.1

4 9

[PATCH v3] test_model: Power arch fixes
by Daniel Henrique Barboza 01 Apr '15

01 Apr '15

changes in v3: - rebase to latest master changes in v2: - changed iso_path to reflect the new ISO being created (Ubuntu 14.04) Daniel Henrique Barboza (1): test/test_model: Power architecture fixes tests/iso_gen.py | 3 ++- tests/test_model.py | 12 +++++++----- 2 files changed, 9 insertions(+), 6 deletions(-) -- 1.9.3

2 3

[kimchi-devel][PATCH] Issue #623: Fix mismatch in host device details query
by lvroyce＠linux.vnet.ibm.com 01 Apr '15

01 Apr '15

From: Royce Lv <lvroyce(a)linux.vnet.ibm.com> Commit #902d2957 began to use splitted devices names, while get detailed device still use original names. Mismatched name results in error raised when listing host devices. Fix it by using splitted device name as well. Signed-off-by: Royce Lv <lvroyce(a)linux.vnet.ibm.com> --- src/kimchi/disks.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/kimchi/disks.py b/src/kimchi/disks.py index a092c17..5cf6885 100644 --- a/src/kimchi/disks.py +++ b/src/kimchi/disks.py @@ -64,7 +64,7 @@ def _get_dev_major_min(name): dev_list = _get_lsblk_devs(keys) for dev in dev_list: - if dev['name'] == name: + if dev['name'].split()[0] == name: maj_min = dev['maj:min'] break else: -- 1.9.1

2 2

[PATCH 0/3] Fix backend VM, POOL, NETWORK naming validation
by Rodrigo Trujillo 01 Apr '15

01 Apr '15

This patchset fixes the backend JSON name checkings when user creates a new VM, POOL or NETWORK. Rodrigo Trujillo (3): Prohibits slashes '/' in VM name Move slash checking in storagepool name from UI to backend Fix Network create name checking in backend (slashes and quotes) src/kimchi/API.json | 4 ++++ src/kimchi/i18n.py | 6 +++--- src/kimchi/model/utils.py | 3 ++- ui/js/src/kimchi.storagepool_add_main.js | 5 ----- ui/pages/i18n.json.tmpl | 1 - 5 files changed, 9 insertions(+), 10 deletions(-) -- 2.1.0

2 7

[PATCH] Github issue #630: fixing None.strip() error
by Daniel Henrique Barboza 31 Mar '15

31 Mar '15

This simple patch fixes the scenario in repositories.py where baseurl or mirrorlist is None and the code was executing baseurl.strip() and mirrorlist.strip(), leading to errors in tests/test_model.py (test_repository_update) and probably other bugs. Signed-off-by: Daniel Henrique Barboza <dhbarboza82(a)gmail.com> --- src/kimchi/repositories.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/kimchi/repositories.py b/src/kimchi/repositories.py index f1e1eb3..19c5810 100644 --- a/src/kimchi/repositories.py +++ b/src/kimchi/repositories.py @@ -275,10 +275,10 @@ class YumRepo(object): config = params.get('config', {}) mirrorlist = config.get('mirrorlist', None) - if len(baseurl.strip()) == 0: + if baseurl is not None and len(baseurl.strip()) == 0: baseurl = None - if len(mirrorlist.strip()) == 0: + if mirrorlist is not None and len(mirrorlist.strip()) == 0: mirrorlist = None if baseurl is None and mirrorlist is None: -- 1.9.3

2 3

[PATCH] issue #628: Set ref_cnt=1 on cloned disks
by Crístian Viana 31 Mar '15

31 Mar '15

Cloned disks currently don't have ref_cnt stored locally, so they may be attached again to another VM - or even to the same one. When cloning a VM, set their disks' ref_cnt to 1 to prevent them from being reattached. Fix issue #628 (Storage volume ref_cnt is not updated after vm clone). Signed-off-by: Crístian Viana <vianac(a)linux.vnet.ibm.com> --- src/kimchi/model/vms.py | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/src/kimchi/model/vms.py b/src/kimchi/model/vms.py index 86c173f..af5e765 100644 --- a/src/kimchi/model/vms.py +++ b/src/kimchi/model/vms.py @@ -449,6 +449,10 @@ class VMModel(object): vir_conn = self.conn.get() + def _delete_disk_from_objstore(path): + with self.objstore as session: + session.delete('storagevolume', path) + for i, path in enumerate(all_paths): try: vir_orig_vol = vir_conn.storageVolLookupByPath(path) @@ -518,6 +522,11 @@ class VMModel(object): xml = xml_item_update(xml, XPATH_DOMAIN_DISK_BY_FILE % path, new_vol['path'], 'file') + # set the new disk's ref_cnt + with self.objstore as session: + session.store('storagevolume', new_vol['path'], {'ref_cnt': 1}) + rollback.prependDefer(_delete_disk_from_objstore, new_vol['path']) + # remove the new volume should an error occur later rollback.prependDefer(self.storagevolume.delete, new_pool_name, new_vol_name) -- 2.1.0

2 2

[PATCH 0/3] Fix issue #461 (backend)
by Crístian Viana 31 Mar '15

31 Mar '15

This patchset fixes the backend part of issue #461. The UI should also be updated in order to use the metadata property. Other related fixes have been committed as well. Crístian Viana (3): Handle empty variables when updating YUM repository Use more generic message in repo mirror list error issue #461: Add 'metalink' support for YUM repositories src/kimchi/API.json | 5 +++++ src/kimchi/i18n.py | 5 +++-- src/kimchi/repositories.py | 25 +++++++++++++++++++------ 3 files changed, 27 insertions(+), 8 deletions(-) -- 2.1.0

3 9

RFC - New option 'Serial Console' in the template UI
by Jose Ricardo Ziviani 30 Mar '15

30 Mar '15

Hello everybody! I'd like to hear from you if it would be nice to have a new option in the Template UI to redirect the output to a serial console. Today, there is a field called 'Graphics' with two options [VNC, Spice]. What if we have [VNC, Spice, Serial Console (no-graphic)]? A drawback is the need of an external tool (like virsh console) to be able to take the console, as Kimchi doesn't provide one. What do you think? Thank you! -- Jose Ricardo Ziviani ----------------------------- Software Engineer Linux Technology Center - IBM

7 9

[PATCH] Fix a ZeroDivisionError when starting kimchi service in Qemu.
by Jose Ricardo Ziviani 30 Mar '15

30 Mar '15

When using kimchi inside a virtual machine it's possible to get a ZeroDivisionError exception because self.sockets is 0, so the division self.cores_per_socket = self.cores_present/self.sockets will fail. In this case, self.sockets is 0 because cores_present = 1 and threads_per_core = 4, which will truncate 0.25 to 0 during the integer division. This is a low priority corner case since kimchi is obviously not supposed to run under a virtual machine but for a plugin development, which will not use any virtualization capability, it might become handy. Signed-off-by: Jose Ricardo Ziviani <joserz(a)linux.vnet.ibm.com> --- src/kimchi/model/cpuinfo.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/src/kimchi/model/cpuinfo.py b/src/kimchi/model/cpuinfo.py index 3411ef5..d063425 100644 --- a/src/kimchi/model/cpuinfo.py +++ b/src/kimchi/model/cpuinfo.py @@ -84,6 +84,8 @@ class CPUInfoModel(object): if not rc: self.threads_per_core = int(out.split()[-1]) self.sockets = self.cores_present/self.threads_per_core + if self.sockets == 0: + self.sockets = 1 self.cores_per_socket = self.cores_present/self.sockets else: # Intel or AMD -- 1.9.1

2 2

Kimchi 1.4.1 is released!
by Aline Manera 27 Mar '15

27 Mar '15

On behalf of everyone who has worked hard on this release, I am pleased to announce the availability of *Kimchi 1.4.1*! This release adds many new features including: ✔ Bug fixes ✔ Add robot.txt file ✔ Disable SSLv3 ✔ Recognize Arch Linux ISO image ✔ Basic ppc65el support ✔ Add man page for kimchid command We have worked hard to ensure that Kimchi runs well on the most popular Linux distributions including: Fedora 21, Ubuntu 14.10, openSUSE 13.2, and RHEL 7.1. Kimchi uses standard Linux interfaces so it should run well on many other distributions too. You can easily grab this release in tarball format or via git: ✔https://github.com/kimchi-project/kimchi/archive/1.4.1.tar.gz <https://github.com/kimchi-project/kimchi/archive/1.4.0.tar.gz> ✔ git clonehttps://github.com/kimchi-project/kimchi.git There are also some packages available at: ✔http://kimchi-project.github.io/kimchi/downloads/ Go ahead! Give it a try and let us know what you think! Regards, Aline Manera

1 0

[kimchi-devel][PATCHv2] Fix update vcpu count: Check available vcpu count before update
by lvroyce＠linux.vnet.ibm.com 26 Mar '15

26 Mar '15

From: Royce Lv <lvroyce(a)linux.vnet.ibm.com> Currently we just check vcpu count does not exceed host available count when topology specified. Actually vcpu count also need to be checked when only vcpu specified. Move this check outside topology check and add it to param update. Signed-off-by: Royce Lv <lvroyce(a)linux.vnet.ibm.com> --- src/kimchi/model/cpuinfo.py | 8 ++++++-- src/kimchi/model/templates.py | 6 +++--- tests/test_model.py | 2 +- tests/test_rest.py | 1 - 4 files changed, 10 insertions(+), 7 deletions(-) diff --git a/src/kimchi/model/cpuinfo.py b/src/kimchi/model/cpuinfo.py index 3411ef5..54bfad4 100644 --- a/src/kimchi/model/cpuinfo.py +++ b/src/kimchi/model/cpuinfo.py @@ -114,11 +114,15 @@ class CPUInfoModel(object): cores = topology['cores'] threads = topology['threads'] + self.check_vcpus(vcpus) + if not self.guest_threads_enabled: raise InvalidOperation("KCHCPUINF0003E") if vcpus != sockets * cores * threads: raise InvalidParameter("KCHCPUINF0002E") - if vcpus > self.cores_available * self.threads_per_core: - raise InvalidParameter("KCHCPUINF0001E") if threads > self.threads_per_core: raise InvalidParameter("KCHCPUINF0002E") + + def check_vcpus(self, vcpus): + if vcpus > self.cores_available * self.threads_per_core: + raise InvalidParameter("KCHCPUINF0001E") diff --git a/src/kimchi/model/templates.py b/src/kimchi/model/templates.py index 6bc8aca..14e5b37 100644 --- a/src/kimchi/model/templates.py +++ b/src/kimchi/model/templates.py @@ -53,7 +53,7 @@ class TemplatesModel(object): {'filename': iso, 'user': user, 'err': excp}) - cpu_info = params.get('cpu_info') + cpu_info = params.setdefault('cpu_info', dict()) if cpu_info: topology = cpu_info.get('topology') # Check, even though currently only topology @@ -68,8 +68,8 @@ class TemplatesModel(object): # exception if a topology is invalid. CPUInfoModel(conn=self.conn).\ check_topology(params['cpus'], topology) - else: - params['cpu_info'] = dict() + if params.get('cpus'): + CPUInfoModel(conn=self.conn).check_vcpus(params['cpus']) conn = self.conn.get() pool_uri = params.get(u'storagepool', '') diff --git a/tests/test_model.py b/tests/test_model.py index 0020022..138a7da 100644 --- a/tests/test_model.py +++ b/tests/test_model.py @@ -577,7 +577,7 @@ class ModelTests(unittest.TestCase): inst = model.Model(None, objstore_loc=self.tmp_store) - orig_params = {'name': 'test', 'memory': '1024', 'cpus': '1', + orig_params = {'name': 'test', 'memory': '1024', 'cpus': 1, 'cdrom': UBUNTU_ISO} inst.templates_create(orig_params) diff --git a/tests/test_rest.py b/tests/test_rest.py index 07e5733..4ecf3ce 100644 --- a/tests/test_rest.py +++ b/tests/test_rest.py @@ -932,7 +932,6 @@ class RestTests(unittest.TestCase): task_info = model.storagevolumes_create('pool-3', params) wait_task(self._task_lookup, task_info['id']) - storagevolume = json.loads(self.request( '/storagepools/kimchi_isos/storagevolumes/').read())[0] self.assertEquals('fedora.iso', storagevolume['name']) -- 1.9.1

2 1

[PATCH] Fix PowerOptions availability
by Jose Ricardo Ziviani 26 Mar '15

26 Mar '15

- When we run tuned-adm without parameters it returns code 2, which will fail our check condition and Power Options would never be displayed. This commit add a parameter that simply list the profiles, this will return 0. --- models/powermanagement.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/models/powermanagement.py b/models/powermanagement.py index 0bb3dcc..7600024 100644 --- a/models/powermanagement.py +++ b/models/powermanagement.py @@ -62,7 +62,7 @@ class PowerProfilesModel(object): return profiles def is_feature_available(self): - output, error, returncode = run_command(['tuned-adm']) + output, error, returncode = run_command(['tuned-adm', 'list']) return returncode == 0 -- 1.9.1

1 1

[PATCH v] Making urls relative
by Frédéric Bonnard 26 Mar '15

26 Mar '15

From: Frederic Bonnard <frediz(a)linux.vnet.ibm.com> On Wed, 11 Mar 2015 12:39:48 -0300, Aline Manera <alinefm(a)linux.vnet.ibm.com> wrote > My point is: if the main idea behind it is to allow user to connects > > Kimchi to the external web server, why do not add an option to do it for > > > the user? > 2) Add an option to Kimchi config file to allow user connects to an > > external web server. I'm a bit confused : do you mean, why don't we add an option so that kimchi starts nginx ? If this is your question, in a distro we would have kimchi package have a dependency on [nginx | apache]. So when you would install kimchi, it would install a web server as well and distros usually starts all services once installed, so, nginx would be already started. That is why, we would need an option, not to start the web server instead. I'm not sure I got you point :) sorry. In the same topic, that's why I'm using the patch in thread "RFC patch to make nginx proxy optional" [1] > And yes - we launch nginx (if we should or not do that it is other > > discussion) and we will continue doing that until something changes what > > > is not in our plans by now. I'm not willing to change that default :) . It'll just be nice to have an option to change that behaviour in other environments [1]. Default would be as it already is, but we could specify with that option to disable launching an external webserver or specify another one like apache. as Royce suggested [2] ? Here follows the patch without the configuration file I put the previous time. F. 1: http://lists.ovirt.org/pipermail/kimchi-devel/2015-February/009642.html 2: http://lists.ovirt.org/pipermail/kimchi-devel/2015-February/009655.html Frederic Bonnard (1): Making urls relative src/kimchi/screenshot.py | 2 +- ui/css/theme-default/template_add.css | 20 ++-- ui/css/theme-default/topbar.css | 2 +- ui/js/src/kimchi.api.js | 176 +++++++++++++++++----------------- ui/js/src/kimchi.login.js | 2 +- ui/pages/guest.html.tmpl | 2 +- ui/pages/help/dita-help.xsl | 4 +- ui/pages/kimchi-ui.html.tmpl | 4 +- ui/pages/storagepool-add.html.tmpl | 2 +- ui/pages/tabs/storage.html.tmpl | 2 +- ui/pages/template-add.html.tmpl | 2 +- 11 files changed, 109 insertions(+), 109 deletions(-) -- 1.9.1

2 5