February 2016 - Kimchi-devel - oVirt List Archives

[PATCH] [Kimchi] Add maxvCpu support UI for guest and template
by Socorro Stoppler 23 Feb '16

23 Feb '16

- Give the option to set maxvCpu when editing guest and template Socorro Stoppler (1): MaxvCpu Initial Work; made some adjustments to style ui/css/kimchi.css | 66 +++++++++++++++++++++++++++++----- ui/css/src/modules/_edit-guests.scss | 22 +++++++++++- ui/css/src/modules/_templates.scss | 25 +++++++++++-- ui/js/src/kimchi.guest_edit_main.js | 30 +++++++++++++--- ui/js/src/kimchi.template_edit_main.js | 53 ++++++++++++++++++++------- ui/pages/guest-edit.html.tmpl | 9 ++++- ui/pages/i18n.json.tmpl | 2 ++ ui/pages/template-edit.html.tmpl | 29 ++++++++------- 8 files changed, 195 insertions(+), 41 deletions(-) -- 1.9.1

2 2

Locking for multi-user env?
by Harshal Patil 23 Feb '16

23 Feb '16

3 2

[PATCH] [Wok] Fixed i18n popup message issue
by rajgupta＠linux.vnet.ibm.com 22 Feb '16

22 Feb '16

From: Rajat Gupta <rajat.triumph(a)gmail.com> Fixed i18n messages display issue for modal dialogs --- ui/js/src/wok.message.js | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/ui/js/src/wok.message.js b/ui/js/src/wok.message.js index d602805..b74d188 100644 --- a/ui/js/src/wok.message.js +++ b/ui/js/src/wok.message.js @@ -57,10 +57,10 @@ wok.message.error = function(msg, node, closeable) { "use strict"; wok.message(msg, 'alert-danger', node, closeable); }; -wok.message.error.code = function(code) { +wok.message.error.code = function(code, node, closeable) { "use strict"; var msg = code + ": " + i18n[code]; - wok.message(msg, 'alert-danger'); + wok.message(msg, 'alert-danger', node, closeable); }; wok.message.success = function(msg, node, closeable) { "use strict"; -- 2.4.3

2 1

[PATCH] [Wok] Reverted wok.combobox and wok.filterSelect
by sguimaraes943＠gmail.com 22 Feb '16

22 Feb '16

From: Samuel Guimarães <sguimaraes943(a)gmail.com> This patch reverts two widgets that were removed and are still being used by Storages modals in Kimchi. samhenri (1): Reverted wok.combobox and wok.filterSelect ui/js/src/wok.combobox.js | 129 ++++++++++++++++++++++++++++++++++++++ ui/js/src/wok.filter-select.js | 138 +++++++++++++++++++++++++++++++++++++++++ 2 files changed, 267 insertions(+) create mode 100644 ui/js/src/wok.combobox.js create mode 100644 ui/js/src/wok.filter-select.js -- 1.9.3

2 2

[PATCH] [Wok] Added support for wok-loading-icon inside form-groups
by sguimaraes943＠gmail.com 22 Feb '16

22 Feb '16

From: Samuel Guimarães <sguimaraes943(a)gmail.com> This commit fixes wok-loading-icon position when inside a form-group with has-feedback class. samhenri (1): Added support for wok-loading-icon inside form-groups ui/css/bootstrap.custom.css | 641 ++++++++++++++++++++++++++----------- ui/css/src/modules/_wok-forms.scss | 25 ++ ui/css/wok.css | 23 ++ 3 files changed, 496 insertions(+), 193 deletions(-) -- 1.9.3

2 2

[PATCH ][Wok]Issue # 28 : Tab order in Host is not consistent
by atreyee＠linux.vnet.ibm.com 22 Feb '16

22 Feb '16

From: Atreyee Mukhopadhyay <atreyee(a)linux.vnet.ibm.com> Second level tabs in Host functionality are sorted using a unique number assigned to align them in desired order irrespective of the plugin installed Unique numbers are assigned to tabs in gingerbase and ginger plugin Atreyee Mukhopadhyay (1): Issue # 28 : Tab order in Host is not consistent ui/js/src/wok.main.js | 822 +++++++++++++++++++++++++------------------------- 1 file changed, 416 insertions(+), 406 deletions(-) -- 2.1.0

2 2

[RFC] [Wok] Display new Wok Logos
by Samuel Henrique De Oliveira Guimaraes 22 Feb '16

22 Feb '16

Hi team, I've done some tests with the new logos and I'm attaching the mockups with a footer in three sizes, normal, smaller and smallest. I'm also attaching what the About window would look like if the user has all plugins installed. In my opinion if we are really going forward with a gray footer, then we could show the logos in white or light gray instead of colors. We would also have to shrink the footer a little more because in the smallest variation it has 60px. Regards, Samuel

3 4

[PATCH][Kimchi 0/3] Support to edit max memory
by Rodrigo Trujillo 21 Feb '16

21 Feb '16

This patchset implements the support in backend to allow user to change maxMemory xml tag. It also changes the memory API from 'memory: XXX' to 'memory': {'current': XXX, 'maxmemory': YYY} maxMemory xml tag is not created by default anymore, when a guest is created. Tag will be created only when the user changes maxmemory manually, and if the value is different (higher) of 'current' memory. So in order to performe memory hotplug, an user has to stop guest, modify maxmemory and then start the guest. Tests were fixed and new ones were included. Rodrigo Trujillo (3): Add support to edit max memory in Templates Add support to edit max memory in Guest Changes and add tests to support new memory API (templates/guest) API.json | 66 +++++++++--- docs/API.md | 29 ++++-- i18n.py | 11 +- model/templates.py | 58 ++++++++++- model/vms.py | 245 ++++++++++++++++++++++++++------------------ osinfo.py | 16 ++- template.conf | 10 +- tests/test_livemigration.py | 6 +- tests/test_model.py | 22 ++-- tests/test_rest.py | 23 ++++- tests/test_template.py | 38 +++++-- tests/test_vmtemplate.py | 26 +++-- vmtemplate.py | 74 ++++++------- 13 files changed, 411 insertions(+), 213 deletions(-) -- 2.1.0

4 15

[RFC] [Wok] Feature: User Activity Log
by Lucio Correia 19 Feb '16

19 Feb '16

User Activity Log functionality ------------------------------- Host admin wants to have access to all actions performed by Wok users, either through API, web user interface, with the option to download User Acitivity Log in a text file. REST API -------- Only one API will be exposed: - GET /logs - Collection of log entries for all installed plugins. Examples of filtering: /logs?plugin=kimchi - All log entries for Kimchi plugin /logs?user=admin - All log entries from admin user /logs?request=PUT - All log entries relative to PUT requests /logs?date=20160110 - All Jan 10th's log entries Backend design -------------- - Define activity texts for each currently implemented activity of all plugins - Implement method to retrieve the activity texts in Wok and all plugins - Log events to a JSON file in the host file system. Each user request received by Wok will generate an entry in the respective log file. Fields to be saved for each entry: plugin, datetime, user, request type (PUT, POST, GET, DELETE), activity text (i.e. "create guest 'vm-sample'") - Implement code for retrieving and filtering log events by the fields above Web User interface design ------------------------- Create a new tab: "User Activity" - available only to Wok admins - Search fields to filter log events by installed plugin, date, user, and/or request type - Show search results in a table containing plugin, datetime, user, request type, activity text - User may sort the table by plugin, datetime, user or request type - Button for downloading current search results as a txt file -- Lucio Correia Software Engineer IBM LTC Brazil

5 21

[PATCH ][Gingerbase] Changes for gingerbase to fix the Issue# 28 in Wok.
by atreyee＠linux.vnet.ibm.com 19 Feb '16

19 Feb '16

From: Atreyee Mukhopadhyay <atreyee(a)linux.vnet.ibm.com> Unique numbers are assigned to tabs details in gingerbase to fix the tab ordering issue in Host functionality. --- ui/config/tab-ext.xml | 2 ++ 1 file changed, 2 insertions(+) diff --git a/ui/config/tab-ext.xml b/ui/config/tab-ext.xml index 69dcca0..dcea62d 100644 --- a/ui/config/tab-ext.xml +++ b/ui/config/tab-ext.xml @@ -5,12 +5,14 @@ <access role="admin" mode="admin"/> <access role="user" mode="read-only"/> <title>Dashboard</title> + <order>0</order> <path>plugins/gingerbase/tabs/host-dashboard.html</path> </tab> <tab> <access role="admin" mode="admin"/> <access role="user" mode="none"/> <title>Updates</title> + <order>20</order> <path>plugins/gingerbase/tabs/host-update.html</path> </tab> </tabs-ext> -- 2.1.0

2 1

[PATCH v4] [Kimchi 0/8] Network API change + VEPA network support
by dhbarboza82＠gmail.com 18 Feb '16

18 Feb '16

From: Daniel Henrique Barboza <dhbarboza82(a)gmail.com> v4: - added one interface only restriction to macvtap and bridge networks v3: - Lucio suggestions made: docs/API.md and 'bridge' connection in test_template.py v2: - changes on the existing network backend This patch set implements two major changes in the network backend and API of Kimchi: - first 4 patches: changed the API to allow networks to be created with more than one interface Changing 'interface' to 'interfaces' in the network backend to allow more than one interface to be added to a network. Existing networks that use only one interface were adapted to use interfaces[0] instead. UI changes were made to comply with this change. - last 4 patches: second version of the VEPA network patch set. VEPA networks are somewhat similar to macvtap networks that tunnels traffic between VMs by using an external VEPA-enabled switch that will process and forward the frames faster than the host CPU/OS can do. The UI support is similar to what we already have for the macvtap network. ATM the UI doesn't support multiple physical devices being assigned to a single VEPA network but the backend does. This can be enhanced in the future with an UI patch. The idea was to have some UI support to get the feature alive ASAP. To create a VEPA network the process is similar to the bridge network, allowing more interfaces to be declared in the new 'interfaces' parameter of the networks backend. This is the XML created when adding a VEPA network using one interface: <network> <name>vepa_net</name> <uuid>d6a3c86b-0592-4ccf-abf0-0abb353870b3</uuid> <forward dev='enp0s25' mode='vepa'> <interface dev='enp0s25'/> </forward> </network> This is how a VM can declare a interface that belongs to this network: <interface type='network'> <mac address='52:54:00:9c:be:bd'/> <source network='vepa_net'/> <model type='virtio'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> This is how the XML looks like when turning on the VM. Libvirt will take care of the creation of the VEPA tunnel by using the assigned physical interface in the network. If more interfaces were added to the network, libvirt will do a load balance between them and the VMs: <interface type='direct'> <mac address='52:54:00:9c:be:bd'/> <source network='vepa_net' dev='enp0s25' mode='vepa'/> <target dev='macvtap0'/> <model type='virtio'/> <alias name='net0'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> Daniel Henrique Barboza (8): Changing network API 'interface' to array: docs and API Changing network API: control and model changes Changing network API: unit test changes Changing network API: UI changes VEPA network support: API and i18n changes VEPA network support - models and xmlutils changes VEPA network support: additional backend unit tests VEPA network support: UI changes API.json | 8 ++--- control/networks.py | 4 +-- docs/API.md | 14 ++++---- i18n.py | 8 +++-- model/networks.py | 65 ++++++++++++++++++++++++------------ tests/test_mock_network.py | 11 ++---- tests/test_model_network.py | 38 ++++++++++++++++++--- tests/test_networkxml.py | 44 +++++++++++++++++++++++- tests/test_template.py | 12 ++++--- ui/js/src/kimchi.network.js | 5 +-- ui/js/src/kimchi.network_add_main.js | 12 +++---- ui/pages/network-add.html.tmpl | 5 +-- xmlutils/network.py | 10 +++++- 13 files changed, 169 insertions(+), 67 deletions(-) -- 2.5.0

2 9

[PATCH v3] [Kimchi 0/8] Network API change + VEPA network support
by dhbarboza82＠gmail.com 17 Feb '16

17 Feb '16

From: Daniel Henrique Barboza <dhbarboza82(a)gmail.com> v3: - Lucio suggestions made: docs/API.md and 'bridge' connection in test_template.py v2: - changes on the existing network backend This patch set implements two major changes in the network backend and API of Kimchi: - first 4 patches: changed the API to allow networks to be created with more than one interface Changing 'interface' to 'interfaces' in the network backend to allow more than one interface to be added to a network. Existing networks that use only one interface were adapted to use interfaces[0] instead. UI changes were made to comply with this change. - last 4 patches: second version of the VEPA network patch set. VEPA networks are somewhat similar to macvtap networks that tunnels traffic between VMs by using an external VEPA-enabled switch that will process and forward the frames faster than the host CPU/OS can do. The UI support is similar to what we already have for the macvtap network. ATM the UI doesn't support multiple physical devices being assigned to a single VEPA network but the backend does. This can be enhanced in the future with an UI patch. The idea was to have some UI support to get the feature alive ASAP. To create a VEPA network the process is similar to the bridge network, allowing more interfaces to be declared in the new 'interfaces' parameter of the networks backend. This is the XML created when adding a VEPA network using one interface: <network> <name>vepa_net</name> <uuid>d6a3c86b-0592-4ccf-abf0-0abb353870b3</uuid> <forward dev='enp0s25' mode='vepa'> <interface dev='enp0s25'/> </forward> </network> This is how a VM can declare a interface that belongs to this network: <interface type='network'> <mac address='52:54:00:9c:be:bd'/> <source network='vepa_net'/> <model type='virtio'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> This is how the XML looks like when turning on the VM. Libvirt will take care of the creation of the VEPA tunnel by using the assigned physical interface in the network. If more interfaces were added to the network, libvirt will do a load balance between them and the VMs: <interface type='direct'> <mac address='52:54:00:9c:be:bd'/> <source network='vepa_net' dev='enp0s25' mode='vepa'/> <target dev='macvtap0'/> <model type='virtio'/> <alias name='net0'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> Daniel Henrique Barboza (8): Changing network API 'interface' to array: docs and API Changing network API: control and model changes Changing network API: unit test changes Changing network API: UI changes VEPA network support: API and i18n changes VEPA network support - models and xmlutils changes VEPA network support: additional backend unit tests VEPA network support: UI changes API.json | 8 ++--- control/networks.py | 4 +-- docs/API.md | 14 +++++---- i18n.py | 7 +++-- model/networks.py | 61 +++++++++++++++++++++++------------- tests/test_mock_network.py | 11 ++----- tests/test_model_network.py | 13 +++++--- tests/test_networkxml.py | 44 +++++++++++++++++++++++++- tests/test_template.py | 12 ++++--- ui/js/src/kimchi.network.js | 5 +-- ui/js/src/kimchi.network_add_main.js | 12 +++---- ui/pages/network-add.html.tmpl | 5 +-- xmlutils/network.py | 10 +++++- 13 files changed, 139 insertions(+), 67 deletions(-) -- 2.5.0

3 11

[PATCH] [Kimchi] Bugfix 804: Disable connect vnc when VM is powered off
by Jose Ricardo Ziviani 17 Feb '16

17 Feb '16

- When a VM is powered off, the option to connect to its vnc should not be displayed under the actions menu otherwise users will get errors. Signed-off-by: Jose Ricardo Ziviani <joserz(a)linux.vnet.ibm.com> --- ui/js/src/kimchi.guest_main.js | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/ui/js/src/kimchi.guest_main.js b/ui/js/src/kimchi.guest_main.js index bfd62a8..83ee3f0 100644 --- a/ui/js/src/kimchi.guest_main.js +++ b/ui/js/src/kimchi.guest_main.js @@ -652,7 +652,8 @@ kimchi.createGuestLi = function(vmObject, prevScreenImage, openMenu) { var consoleActions = guestActions.find("[name=vm-console]"); var consoleLinkActions = result.find(".vnc-link"); - if ((vmObject.graphics['type'] == 'vnc') || (vmObject.graphics['type'] == 'spice')) { + if (((vmObject.graphics['type'] == 'vnc') || (vmObject.graphics['type'] == 'spice')) + && (!vmPoweredOffBool)) { consoleActions.on("click", function(event) { event.preventDefault(); kimchi.openVmConsole(event); -- 1.9.1

3 2

[PATCH] [Kimchi] Bugfix 843: Add sourceURL comment for kimchi.min.js
by Jose Ricardo Ziviani 17 Feb '16

17 Feb '16

- browser javascript debuggers cannot inspect dynamic loaded javascript because such scripts have no name (they are created with eval). The comment //# sourceURL=name is a way to solve it. - https://developer.chrome.com/devtools/docs/javascript-debugging Signed-off-by: Jose Ricardo Ziviani <joserz(a)linux.vnet.ibm.com> --- ui/js/Makefile.am | 1 + 1 file changed, 1 insertion(+) diff --git a/ui/js/Makefile.am b/ui/js/Makefile.am index c9d1218..c24feeb 100644 --- a/ui/js/Makefile.am +++ b/ui/js/Makefile.am @@ -23,5 +23,6 @@ dist_js_DATA = kimchi.min.js $(filter-out kimchi.min.js, $(wildcard *.js)) kimchi.min.js: src/*.js cat $(sort $^) > $@ + echo "//# sourceURL=kimchi.min.js" >> $@ CLEANFILES = kimchi.min.js -- 1.9.1

3 2

[PATCH] [Wok] Allow user disables HTTP port
by Aline Manera 17 Feb '16

17 Feb '16

This patch adds a new option to wok.conf file - https_only - to allow user disable HTTP port. The default value for https_only option is false, which means HTTP and HTTPS are allowed. When set to true, all the connections will be done through HTTPS. Signed-off-by: Aline Manera <alinefm(a)linux.vnet.ibm.com> --- src/nginx/wok.conf.in | 7 +------ src/wok.conf.in | 6 +++++- src/wok/config.py.in | 1 + src/wok/proxy.py | 16 +++++++++++++++- src/wokd.in | 4 ++++ tests/utils.py | 10 +++++----- 6 files changed, 31 insertions(+), 13 deletions(-) diff --git a/src/nginx/wok.conf.in b/src/nginx/wok.conf.in index 501377e..8dd5d7c 100644 --- a/src/nginx/wok.conf.in +++ b/src/nginx/wok.conf.in @@ -32,7 +32,6 @@ events { } http { - log_format main '$remote_addr - $remote_user [$time_local] "$request" ' '$status $body_bytes_sent "$http_referer" ' '"$http_user_agent" "$http_x_forwarded_for"'; @@ -88,9 +87,5 @@ http { proxy_set_header Connection $connection_upgrade; } } - - server { - listen ${host_addr}:${proxy_port}; - rewrite ^/(.*)$ https://$host:${proxy_ssl_port}/$1 redirect; - } + ${http_config} } diff --git a/src/wok.conf.in b/src/wok.conf.in index 7d479d3..77a79b6 100644 --- a/src/wok.conf.in +++ b/src/wok.conf.in @@ -9,9 +9,13 @@ # Port to listen on #port = 8000 -# If present, start an SSL-enabled server on the given port +# Start an SSL-enabled server on the given port #ssl_port = 8001 +# Allow user disables HTTP port. In that case, all the connections +# will be done directly through HTTPS port (values: true|false) +#https_only = false + # Cherrypy server port #cherrypy_port = 8010 diff --git a/src/wok/config.py.in b/src/wok/config.py.in index 5d32ba4..40fbcda 100644 --- a/src/wok/config.py.in +++ b/src/wok/config.py.in @@ -232,6 +232,7 @@ def _get_config(): config.set("server", "host", "0.0.0.0") config.set("server", "port", "8000") config.set("server", "ssl_port", "8001") + config.set("server", "https_only", "false") config.set("server", "cherrypy_port", "8010") config.set("server", "websockets_port", "64667") config.set("server", "ssl_cert", "") diff --git a/src/wok/proxy.py b/src/wok/proxy.py index c7bc665..9d39dbd 100644 --- a/src/wok/proxy.py +++ b/src/wok/proxy.py @@ -33,6 +33,14 @@ from wok import sslcert from wok.config import paths +HTTP_CONFIG = """ +server { + listen %(host_addr)s:%(proxy_port)s; + rewrite ^/(.*)$ https://$host:%(proxy_ssl_port)s/$1 redirect; +} +""" + + def _create_proxy_config(options): """Create nginx configuration file based on current ports config @@ -75,6 +83,12 @@ def _create_proxy_config(options): # Setting up Diffie-Hellman group with 2048-bit file dhparams_pem = os.path.join(config_dir, "dhparams.pem") + http_config = '' + if options.https_only == 'false': + http_config = HTTP_CONFIG % {'host_addr': options.host, + 'proxy_port': options.port, + 'proxy_ssl_port': options.ssl_port} + # Read template file and create a new config file # with the specified parameters. with open(os.path.join(nginx_config_dir, "wok.conf.in")) as template: @@ -82,8 +96,8 @@ def _create_proxy_config(options): data = Template(data) data = data.safe_substitute(user=user_proxy, host_addr=options.host, - proxy_port=options.port, proxy_ssl_port=options.ssl_port, + http_config=http_config, cherrypy_port=options.cherrypy_port, websockets_port=options.websockets_port, cert_pem=cert, cert_key=key, diff --git a/src/wokd.in b/src/wokd.in index 59a81f8..7255d3c 100644 --- a/src/wokd.in +++ b/src/wokd.in @@ -46,6 +46,7 @@ def main(options): host = config.config.get("server", "host") port = config.config.get("server", "port") ssl_port = config.config.get("server", "ssl_port") + https_only = config.config.get("server", "https_only") cherrypy_port = config.config.get("server", "cherrypy_port") websockets_port = config.config.get("server", "websockets_port") runningEnv = config.config.get("server", "environment") @@ -59,6 +60,9 @@ def main(options): help="Port to listen on (default %s)" % port) parser.add_option('--ssl-port', type="int", default=ssl_port, help="Port to enable SSL (default %s)" % ssl_port) + parser.add_option('--https_only', type="choice", default=https_only, + choices=['false', 'true'], + help="Disable HTTP port (default %s)" % ssl_port) parser.add_option('--cherrypy_port', type="int", default=cherrypy_port, help="Cherrypy server port (default %s)" % cherrypy_port) parser.add_option('--websockets_port', type="int", default=websockets_port, diff --git a/tests/utils.py b/tests/utils.py index bb9efb0..d158ba1 100644 --- a/tests/utils.py +++ b/tests/utils.py @@ -115,11 +115,11 @@ def run_server(host, port, ssl_port, test_mode, cherrypy_port=None, args = type('_', (object,), {'host': host, 'port': port, 'ssl_port': ssl_port, - 'cherrypy_port': cherrypy_port, 'max_body_size': '4*1024', - 'websockets_port': 64667, 'ssl_cert': '', - 'ssl_key': '', 'test': test_mode, 'access_log': '/dev/null', - 'error_log': '/dev/null', 'environment': environment, - 'log_level': 'debug'})() + 'https_only': 'false', 'cherrypy_port': cherrypy_port, + 'websockets_port': 64667, 'ssl_cert': '', 'ssl_key': '', + 'max_body_size': '4*1024', 'test': test_mode, + 'access_log': '/dev/null', 'error_log': '/dev/null', + 'environment': environment, 'log_level': 'debug'})() if model is not None: setattr(args, 'model', model) -- 2.5.0

2 2

[PATCH] [Wok] Fixed radio and checkbox focus event v3
by sguimaraes943＠gmail.com 17 Feb '16

17 Feb '16

From: Samuel Guimarães <sguimaraes943(a)gmail.com> This patch fixes focus event and keyboard navigation with radio and checkboxes input fields. v2 - Fixes checkbox pointer cursor v3 - Fixes focus ring color when checkbox is active samhenri (1): Fixed radio and checkbox focus event ui/css/src/modules/_wok-forms.scss | 11 ++++++-- ui/css/wok.css | 57 ++++++++++++++++++++++++++++++-------- 2 files changed, 54 insertions(+), 14 deletions(-) -- 1.9.3

2 2

[PATCH] [Wok] Fix pep8 E126 (continuation line) issues
by Jose Ricardo Ziviani 17 Feb '16

17 Feb '16

Signed-off-by: Jose Ricardo Ziviani <joserz(a)linux.vnet.ibm.com> --- tests/test_config.py.in | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/tests/test_config.py.in b/tests/test_config.py.in index 28c7a57..23eeca2 100644 --- a/tests/test_config.py.in +++ b/tests/test_config.py.in @@ -136,77 +136,77 @@ class ConfigTests(unittest.TestCase): 'tools.nocache.on': False, 'tools.staticfile.on': True, 'tools.staticfile.filename': - '%s/fontawesome-webfont.ttf' % paths.fontawesome_dir + '%s/fontawesome-webfont.ttf' % paths.fontawesome_dir }, '/fonts/OpenSans-LightItalic.ttf': { 'tools.wokauth.on': False, 'tools.nocache.on': False, 'tools.staticfile.on': True, 'tools.staticfile.filename': - '%s/OpenSans-LightItalic.ttf' % paths.opensans_dir + '%s/OpenSans-LightItalic.ttf' % paths.opensans_dir }, '/fonts/OpenSans-Bold.ttf': { 'tools.wokauth.on': False, 'tools.nocache.on': False, 'tools.staticfile.on': True, 'tools.staticfile.filename': - '%s/OpenSans-Bold.ttf' % paths.opensans_dir + '%s/OpenSans-Bold.ttf' % paths.opensans_dir }, '/fonts/OpenSans-Italic.ttf': { 'tools.wokauth.on': False, 'tools.nocache.on': False, 'tools.staticfile.on': True, 'tools.staticfile.filename': - '%s/OpenSans-Italic.ttf' % paths.opensans_dir + '%s/OpenSans-Italic.ttf' % paths.opensans_dir }, '/fonts/OpenSans-Semibold.ttf': { 'tools.wokauth.on': False, 'tools.nocache.on': False, 'tools.staticfile.on': True, 'tools.staticfile.filename': - '%s/OpenSans-Semibold.ttf' % paths.opensans_dir + '%s/OpenSans-Semibold.ttf' % paths.opensans_dir }, '/fonts/OpenSans-Light.ttf': { 'tools.wokauth.on': False, 'tools.nocache.on': False, 'tools.staticfile.on': True, 'tools.staticfile.filename': - '%s/OpenSans-Light.ttf' % paths.opensans_dir + '%s/OpenSans-Light.ttf' % paths.opensans_dir }, '/fonts/OpenSans-SemiboldItalic.ttf': { 'tools.wokauth.on': False, 'tools.nocache.on': False, 'tools.staticfile.on': True, 'tools.staticfile.filename': - '%s/OpenSans-SemiboldItalic.ttf' % paths.opensans_dir + '%s/OpenSans-SemiboldItalic.ttf' % paths.opensans_dir }, '/fonts/OpenSans-ExtraBold.ttf': { 'tools.wokauth.on': False, 'tools.nocache.on': False, 'tools.staticfile.on': True, 'tools.staticfile.filename': - '%s/OpenSans-ExtraBold.ttf' % paths.opensans_dir + '%s/OpenSans-ExtraBold.ttf' % paths.opensans_dir }, '/fonts/OpenSans-BoldItalic.ttf': { 'tools.wokauth.on': False, 'tools.nocache.on': False, 'tools.staticfile.on': True, 'tools.staticfile.filename': - '%s/OpenSans-BoldItalic.ttf' % paths.opensans_dir + '%s/OpenSans-BoldItalic.ttf' % paths.opensans_dir }, '/fonts/OpenSans-ExtraBoldItalic.ttf': { 'tools.wokauth.on': False, 'tools.nocache.on': False, 'tools.staticfile.on': True, 'tools.staticfile.filename': - '%s/OpenSans-ExtraBoldItalic.ttf' % paths.opensans_dir + '%s/OpenSans-ExtraBoldItalic.ttf' % paths.opensans_dir }, '/fonts/OpenSans-Regular.ttf': { 'tools.wokauth.on': False, 'tools.nocache.on': False, 'tools.staticfile.on': True, 'tools.staticfile.filename': - '%s/OpenSans-Regular.ttf' % paths.opensans_dir + '%s/OpenSans-Regular.ttf' % paths.opensans_dir } } -- 1.9.1

3 2

[PATCH] [Wok] Bugfix 33: Implement a multiprocess file handler for log system
by Jose Ricardo Ziviani 17 Feb '16

17 Feb '16

- This commit adds a new watched file handler, used in Wok, but it uses a recursive lock before emitting the log register to avoid possible concurrency issues since multiple processes will write in the same log file. Signed-off-by: Jose Ricardo Ziviani <joserz(a)linux.vnet.ibm.com> --- *NOTE*: based on '[PATCH] [Wok] Bugfix 30: Duplicated log messages' src/wok/safewatchedfilehandler.py | 44 +++++++++++++++++++++++++++++++++++++++ src/wok/server.py | 4 ++-- 2 files changed, 46 insertions(+), 2 deletions(-) create mode 100644 src/wok/safewatchedfilehandler.py diff --git a/src/wok/safewatchedfilehandler.py b/src/wok/safewatchedfilehandler.py new file mode 100644 index 0000000..c5fad85 --- /dev/null +++ b/src/wok/safewatchedfilehandler.py @@ -0,0 +1,44 @@ +# +# Project Kimchi +# +# Copyright IBM, Corp. 2016 +# +# This library is free software; you can redistribute it and/or +# modify it under the terms of the GNU Lesser General Public +# License as published by the Free Software Foundation; either +# version 2.1 of the License, or (at your option) any later version. +# +# This library is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +# Lesser General Public License for more details. +# +# You should have received a copy of the GNU Lesser General Public +# License along with this library; if not, write to the Free Software +# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + +from logging.handlers import WatchedFileHandler +from multiprocessing import RLock + + +class SafeWatchedFileHandler(WatchedFileHandler): + + def __init__(self, filename, mode='a', encoding=None, delay=0): + WatchedFileHandler.__init__(self, filename, mode, encoding, delay) + self._lock = RLock() + + def close(self): + self._lock.acquire(timeout=2) + try: + WatchedFileHandler.close(self) + + finally: + self._lock.release() + + def emit(self, record): + self._lock.acquire(timeout=2) + try: + WatchedFileHandler.emit(self, record) + + finally: + self._lock.release() diff --git a/src/wok/server.py b/src/wok/server.py index 13851e8..75b41d5 100644 --- a/src/wok/server.py +++ b/src/wok/server.py @@ -35,6 +35,7 @@ from wok.control import sub_nodes from wok.model import model from wok.proxy import start_proxy, terminate_proxy from wok.root import WokRoot +from wok.safewatchedfilehandler import SafeWatchedFileHandler from wok.utils import get_enabled_plugins, import_class LOGGING_LEVEL = {"debug": logging.DEBUG, @@ -122,8 +123,7 @@ class Server(object): cherrypy.log.access_log.addHandler(h) # Create handler to error log file - h = logging.handlers.WatchedFileHandler(options.error_log, 'a', - delay=1) + h = SafeWatchedFileHandler(options.error_log, 'a', delay=1) h.setLevel(logLevel) h.setFormatter(cherrypy._cplogging.logfmt) -- 1.9.1

3 2

[PATCH] [Wok] Bugfix 30: Duplicated log messages
by Jose Ricardo Ziviani 17 Feb '16

17 Feb '16

- today we have two file handlers pointing to the same file, this commit closes the standard file handler to use the watched file handler. Signed-off-by: Jose Ricardo Ziviani <joserz(a)linux.vnet.ibm.com> --- src/wok/server.py | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/src/wok/server.py b/src/wok/server.py index 80c1da2..13851e8 100644 --- a/src/wok/server.py +++ b/src/wok/server.py @@ -101,6 +101,17 @@ class Server(object): if dev_env: cherrypy.log.screen = True + # close standard file handlers because we are going to use a + # watchedfiled handler, otherwise we will have two file handlers + # pointing to the same file, duplicating log enries + for handler in cherrypy.log.access_log.handlers[:]: + if isinstance(handler, logging.FileHandler): + cherrypy.log.access_log.removeHandler(handler) + + for handler in cherrypy.log.error_log.handlers[:]: + if isinstance(handler, logging.FileHandler): + cherrypy.log.error_log.removeHandler(handler) + # Create handler to access log file h = logging.handlers.WatchedFileHandler(options.access_log, 'a', delay=1) -- 1.9.1

3 2

[PATCH v2] [Kimchi 0/8] Network API change + VEPA network support
by dhbarboza82＠gmail.com 16 Feb '16

16 Feb '16

From: Daniel Henrique Barboza <dhbarboza82(a)gmail.com> v2: - changes on the existing network backend This patch set implements two major changes in the network backend and API of Kimchi: - first 4 patches: changed the API to allow networks to be created with more than one interface Changing 'interface' to 'interfaces' in the network backend to allow more than one interface to be added to a network. Existing networks that use only one interface were adapted to use interfaces[0] instead. UI changes were made to comply with this change. - last 4 patches: second version of the VEPA network patch set. VEPA networks are somewhat similar to macvtap networks that tunnels traffic between VMs by using an external VEPA-enabled switch that will process and forward the frames faster than the host CPU/OS can do. The UI support is similar to what we already have for the macvtap network. ATM the UI doesn't support multiple physical devices being assigned to a single VEPA network but the backend does. This can be enhanced in the future with an UI patch. The idea was to have some UI support to get the feature alive ASAP. To create a VEPA network the process is similar to the bridge network, allowing more interfaces to be declared in the new 'interfaces' parameter of the networks backend. This is the XML created when adding a VEPA network using one interface: <network> <name>vepa_net</name> <uuid>d6a3c86b-0592-4ccf-abf0-0abb353870b3</uuid> <forward dev='enp0s25' mode='vepa'> <interface dev='enp0s25'/> </forward> </network> This is how a VM can declare a interface that belongs to this network: <interface type='network'> <mac address='52:54:00:9c:be:bd'/> <source network='vepa_net'/> <model type='virtio'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> This is how the XML looks like when turning on the VM. Libvirt will take care of the creation of the VEPA tunnel by using the assigned physical interface in the network. If more interfaces were added to the network, libvirt will do a load balance between them and the VMs: <interface type='direct'> <mac address='52:54:00:9c:be:bd'/> <source network='vepa_net' dev='enp0s25' mode='vepa'/> <target dev='macvtap0'/> <model type='virtio'/> <alias name='net0'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> Daniel Henrique Barboza (8): Changing network API 'interface' to array: docs and API Changing network API: control and model changes Changing network API: unit test changes Changing network API: UI changes VEPA network support: API and i18n changes VEPA network support - models and xmlutils changes VEPA network support: additional backend unit tests VEPA network support: UI changes API.json | 8 ++--- control/networks.py | 4 +-- docs/API.md | 15 +++++---- i18n.py | 7 +++-- model/networks.py | 61 +++++++++++++++++++++++------------- tests/test_mock_network.py | 11 ++----- tests/test_model_network.py | 13 +++++--- tests/test_networkxml.py | 44 +++++++++++++++++++++++++- tests/test_template.py | 12 ++++--- ui/js/src/kimchi.network.js | 5 +-- ui/js/src/kimchi.network_add_main.js | 12 +++---- ui/pages/network-add.html.tmpl | 5 +-- xmlutils/network.py | 10 +++++- 13 files changed, 140 insertions(+), 67 deletions(-) -- 2.5.0

3 14

[PATCH] [Wok] Fixed radio and checkbox focus event v2
by sguimaraes943＠gmail.com 16 Feb '16

16 Feb '16

From: Samuel Guimarães <sguimaraes943(a)gmail.com> This patch fixes focus event and keyboard navigation with radio and checkboxes input fields. v2 - Fixes checkboxes Samuel Guimarães (1): Fixed radio and checkbox focus event ui/css/src/modules/_wok-forms.scss | 11 ++++++-- ui/css/wok.css | 57 ++++++++++++++++++++++++++++++-------- 2 files changed, 54 insertions(+), 14 deletions(-) -- 1.9.3

2 2

Scrum meeting time
by Aline Manera 15 Feb '16

15 Feb '16

Hi all, As discussed in the last scrum meeting, the scrum meeting time will change from now on. It will be 10AM BRT/BRST (UTC-3/UTC-2). See you there on Wednesday. Regards, Aline Manera

1 0

[PATCH] [Wok] Fixed radio and checkbox focus event
by sguimaraes943＠gmail.com 15 Feb '16

15 Feb '16

From: Samuel Guimarães <sguimaraes943(a)gmail.com> This patch fixes focus event and keyboard navigation with radio and checkboxes input fields. Samuel Guimarães (1): Fixed radio and checkbox focus event ui/css/src/modules/_wok-forms.scss | 9 ++++++- ui/css/wok.css | 55 ++++++++++++++++++++++++++++++-------- 2 files changed, 52 insertions(+), 12 deletions(-) -- 1.9.3

1 1

[PATCH] [Kimchi] [RFC] Max memory front-end
by sguimaraes943＠gmail.com 15 Feb '16

15 Feb '16

From: Samuel Guimarães <sguimaraes943(a)gmail.com> This patch adds a hidden "Max Memory" numbered input in Edit Guests. It also changes memory "text" input in Edit Templates to "number" to match Edit Guests ui. Validation (form submission) wasn't added as the selected guest json doesn't have a "max memory" key. Signed-off-by: Samuel Guimarães <sguimaraes943(a)gmail.com> --- ui/css/kimchi.css | 27 ++++++++++++++++++++------- ui/css/src/modules/_edit-guests.scss | 14 +++++++++++++- ui/js/src/kimchi.guest_edit_main.js | 9 ++++++++- ui/pages/guest-edit.html.tmpl | 9 ++++++++- ui/pages/i18n.json.tmpl | 2 ++ ui/pages/template-edit.html.tmpl | 2 +- 6 files changed, 52 insertions(+), 11 deletions(-) diff --git a/ui/css/kimchi.css b/ui/css/kimchi.css index 88ae7de..d032c4c 100644 --- a/ui/css/kimchi.css +++ b/ui/css/kimchi.css @@ -78,7 +78,7 @@ #guest-add-window.modal-content label.box-iso-outer span.box-iso-border { display: block; border: 3px solid transparent; - transition: all .1s ease-in-out; + transition: all 0.1s ease-in-out; } #template-add-window.modal-content label.box-iso-outer .iso-radio-hidden:checked + span.box-iso-border, @@ -100,7 +100,7 @@ display: block; border: 1px solid transparent; background: #fff; - transition: all .1s ease-in-out; + transition: all 0.1s ease-in-out; } #template-add-window.modal-content ul.list-template, @@ -666,10 +666,20 @@ position: relative; } -#guest-edit-window #form-guest-edit-general #guest-edit-memory-textbox { +#guest-edit-window #form-guest-edit-general #guest-max-memory-panel { + display: none; +} + +#guest-edit-window #form-guest-edit-general #guest-edit-memory-textbox, +#guest-edit-window #form-guest-edit-general #guest-edit-max-memory-textbox { width: 590px; } +#guest-edit-window #form-guest-edit-general #guest-edit-memory-textbox, +#guest-edit-window #form-guest-edit-general #guest-show-max-memory { + display: inline-block; +} + #guest-edit-window #form-guest-edit-interface .column-actions { width: 23%; } @@ -1007,7 +1017,7 @@ border: 1px solid #eee !important; margin-bottom: 3px; display: block; - padding: .2em .6em .3em; + padding: 0.2em 0.6em 0.3em; font-weight: 700; line-height: 1; text-align: left; @@ -1250,7 +1260,7 @@ height: 664px; width: 2164px; left: 0; - transition: left .2s ease-in-out; + transition: left 0.2s ease-in-out; } #template-add-window.modal-content p { @@ -1526,7 +1536,9 @@ padding-right: 35px !important; } -#templates-root-container .wok-vm-gallery .item-hidden.column-type, #templates-root-container .wok-vm-gallery .item-hidden.column-version, #templates-root-container .wok-vm-gallery .item-hidden.column-processors { +#templates-root-container .wok-vm-gallery .item-hidden.column-type, +#templates-root-container .wok-vm-gallery .item-hidden.column-version, +#templates-root-container .wok-vm-gallery .item-hidden.column-processors { padding-bottom: 11px; } @@ -2067,7 +2079,8 @@ white-space: nowrap; } -.storage-modal .filter-select.popable .popover ul li:hover, .storage-modal .filter-select.popable .popover ul li:focus, +.storage-modal .filter-select.popable .popover ul li:hover, +.storage-modal .filter-select.popable .popover ul li:focus, .storage-modal .storage-target-input .popover ul li:hover, .storage-modal .storage-target-input .popover ul li:focus, .storage-modal .storage-add-input .popover ul li:hover, diff --git a/ui/css/src/modules/_edit-guests.scss b/ui/css/src/modules/_edit-guests.scss index 4d642f3..057aad9 100644 --- a/ui/css/src/modules/_edit-guests.scss +++ b/ui/css/src/modules/_edit-guests.scss @@ -25,9 +25,21 @@ position: relative; } #form-guest-edit-general { - #guest-edit-memory-textbox { + + #guest-max-memory-panel { + display: none; + } + + #guest-edit-memory-textbox, + #guest-edit-max-memory-textbox { width: 590px; } + + #guest-edit-memory-textbox, + #guest-show-max-memory { + display: inline-block; + } + } #form-guest-edit-interface { .column-actions { diff --git a/ui/js/src/kimchi.guest_edit_main.js b/ui/js/src/kimchi.guest_edit_main.js index 136469b..aa09b65 100644 --- a/ui/js/src/kimchi.guest_edit_main.js +++ b/ui/js/src/kimchi.guest_edit_main.js @@ -669,9 +669,16 @@ kimchi.guest_edit_main = function() { wok.window.open('plugins/kimchi/guest-storage-add.html', 'extendCreateStorage'); }); if ((kimchi.thisVMState === "running") || (kimchi.thisVMState === "paused")) { - $("#form-guest-edit-general input").not("#guest-edit-memory-textbox").prop("disabled", true); + $("#form-guest-edit-general input").not("#guest-edit-memory-textbox, #guest-edit-max-memory-textbox").prop("disabled", true); } + $('#guest-show-max-memory').on('click', function(e) { + e.preventDefault; + $('#guest-max-memory-panel').slideToggle(); + var text = $('#guest-show-max-memory span.text').text(); + $('#guest-show-max-memory span.text').text(text == i18n['KCHVMED6008M'] ? i18n['KCHVMED6009M'] : i18n['KCHVMED6008M']); + }); + var onAttached = function(params) { refreshCDROMs(); }; diff --git a/ui/pages/guest-edit.html.tmpl b/ui/pages/guest-edit.html.tmpl index 0ad635c..b3e5817 100644 --- a/ui/pages/guest-edit.html.tmpl +++ b/ui/pages/guest-edit.html.tmpl @@ -50,7 +50,14 @@ </div> <div class="form-group"> <label for="guest-edit-memory-textbox">$_("Memory (MB)")</label> - <input id="guest-edit-memory-textbox" class="form-control" name="memory" type="number" min="1024" step="1024" /> + <div id="guest-memory"> + <input id="guest-edit-memory-textbox" class="form-control" name="memory" type="number" min="1024" step="1024" /> + <button id="guest-show-max-memory" class="btn btn-primary" type="button"><i class="fa fa-plus-circle"></i> <span class="text">$_("More")</span></button> + </div> + </div> + <div id="guest-max-memory-panel" class="form-group"> + <label for="guest-edit-max-memory-textbox">$_("Max Memory (MB)")</label> + <input id="guest-edit-max-memory-textbox" class="form-control" name="max-memory" type="number" min="1024" step="1024" /> </div> <div class="form-group"> <label for="guest-edit-icon-textbox">$_("Icon")</label> diff --git a/ui/pages/i18n.json.tmpl b/ui/pages/i18n.json.tmpl index 8bc8683..c06b9be 100644 --- a/ui/pages/i18n.json.tmpl +++ b/ui/pages/i18n.json.tmpl @@ -78,6 +78,8 @@ "KCHVMED6005M": "$_("drive_type:")", "KCHVMED6006M": "$_("model:")", "KCHVMED6007M": "$_("Affected devices:")", + "KCHVMED6008M": "$_("More")", + "KCHVMED6009M": "$_("Less")", "KCHNET6001E": "$_("The VLAN id must be between 1 and 4094.")", diff --git a/ui/pages/template-edit.html.tmpl b/ui/pages/template-edit.html.tmpl index 478ccd5..47d4175 100644 --- a/ui/pages/template-edit.html.tmpl +++ b/ui/pages/template-edit.html.tmpl @@ -82,7 +82,7 @@ <input id="template-edit-version-textbox" class="form-control" name="os_version" type="text" disabled="disabled" /> </div> <div class="template-edit-wrapper-controls"> - <input id="template-edit-memory-textbox" class="form-control" name="memory" type="text" /> + <input id="template-edit-memory-textbox" class="form-control" name="memory" type="number" step="1" min="1" /> </div> <div class="template-edit-wrapper-controls templ-edit-cdrom"> <input id="template-edit-cdrom-textbox" class="form-control" name="cdrom" type="text" disabled="disabled" /> -- 1.9.3

4 5

[PATCH v5] [Kimchi 0/8] Web serial console
by Jose Ricardo Ziviani 15 Feb '16

15 Feb '16

v5: - fixed test config + '/serial': { + 'tools.nocache.on': True, + 'tools.wokauth.on': True, + 'tools.staticdir.dir': paths.serial_dir, + 'tools.staticdir.on': True + }, - fixed parent method calling from websocket.py def get_target(self, target_plugin, path): - target = super(CustomHandler, self).get_target(target_plugin, path) + if issubclass(CustomHandler, object): + target = super(CustomHandler, self).get_target(target_plugin, + path) + else: + target = request_proxy.get_target(self, target_plugin, path) + v4: - fix patch encoding to utf-8 v3: - applied code review - removed favicon from this patch v2: - applied code review - updated the build system This is the initial version of a web serial console for kimchi/libvirt VMs. When a VM is turned on, a new action is displayed named "Connect Serial", this will open a new tab with an interface like the existing novnc. That interface opens a websocket to kimchi webserver (nginx), then it is redirected to websockify. Websockify will proxy that connection to a local unix socket server to finally communicate with the guest console. I chose to create one server per guest because that's the way websockify was designed (it was born inside novnc), so I could reuse the token security plugin implemented in websockify. In order to avoid wasting resources I decided to user unix socket, a local/lightweight/reliable socket if compared with internet sockets, this uses files instead of ports to accept connections. When a connection is established no one else can get that console (I'm not multiplexing it but it's possible in future). The serial console will be available again if the current user does one of: - type ctrl+q - close the tab - 2 min. timeout Thanks Jose Ricardo Ziviani (8): Rename vnc.py to websocket.py Implement the web serial console server Implement the backend to support web serial console Import term.js to Kimchi project Implement the web serial console front-end Implement the Kimchi front-end for the web serial console Update the build system to make the serial console Add test case for the socket server COPYING | 2 + Makefile.am | 4 +- config.py.in | 6 +- configure.ac | 3 + control/vms.py | 3 +- i18n.py | 3 + model/vms.py | 43 +- root.py | 6 +- serialconsole.py | 315 +++ tests/test_config.py.in | 6 + tests/test_model.py | 28 +- ui/Makefile.am | 2 +- ui/js/src/kimchi.api.js | 27 + ui/js/src/kimchi.guest_main.js | 14 +- ui/pages/guest.html.tmpl | 5 +- ui/serial/Makefile.am | 21 + ui/serial/html/Makefile.am | 21 + ui/serial/html/serial.html | 99 + ui/serial/libs/Makefile.am | 21 + ui/serial/libs/term.js | 5973 ++++++++++++++++++++++++++++++++++++++++ vnc.py | 92 - websocket.py | 127 + 22 files changed, 6711 insertions(+), 110 deletions(-) create mode 100644 serialconsole.py create mode 100644 ui/serial/Makefile.am create mode 100644 ui/serial/html/Makefile.am create mode 100644 ui/serial/html/serial.html create mode 100644 ui/serial/libs/Makefile.am create mode 100644 ui/serial/libs/term.js delete mode 100644 vnc.py create mode 100644 websocket.py -- 1.9.1

2 11