13:01:24 <alinefm> #startmeeting
13:01:24 <kimchi-bot> Meeting started Wed Jan 21 13:01:24 2015 UTC.  The chair is alinefm. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:01:24 <kimchi-bot> Useful Commands: #action #agreed #help #info #idea #link #topic.
13:01:24 <alinefm> #meetingname scrum
13:01:24 <kimchi-bot> The meeting name has been set to 'scrum'
13:01:34 <alinefm> #info Agenda 1) Status 2) Open discussion
13:01:35 <alinefm> anything else?
13:02:33 <alinefm> #topic Status
13:02:33 <alinefm> #info Please provide your status using the #info command: #info <nickname> <status>
13:02:40 <YuXin> #info YuXin Patch of Tab Container Widget
13:02:52 <alinefm> #info alinefm is working on storage pool and storage volumes test cases
13:02:59 <wenwang> #info wenwang sent patch of kimchi ui widget of gauge
13:03:06 <wenwang> #info wenwang sent patch of kimchi ui widget of messagebar
13:03:12 <wenwang> #info wenwang sent patch of kimchi ui widget of message
13:03:22 <wenwang> #info wenwang sent patch of kimchi ui widget of dialog
13:03:24 <royce> #info royce debuged with problem of leaking file handler of upload, it turned to be kimchi server fh leak, sent series patch fixing leak, now resolving libvirt stream locking
13:03:33 <vianac> #info vianac worked on patches to allow creating/reverting a snapshot of a running domain, and to allow VM suspended state; reviewed patches on the mailing list
13:03:40 <wenwang> #info wenwang sent patch of kimchi ui widget of button
13:03:40 <YuXin> #info YuXin Investigated 6 Javascript Charting Library, Identified 3 that Kimchi can leverage, Tend to use ChartJS, Will prototype with it
13:03:53 <alinefm> #info alinefm got a Kimchi presentation accepted for LF Collab Summit
13:04:11 <royce> Great!
13:05:04 <alinefm> yeap
13:05:14 <alinefm> royce, anything you need help with the upload issues?
13:05:27 <YuXin> #info YuXin Patch of Web Font Icon Library
13:05:29 <alinefm> vianac, weren't the patches ready?
13:06:20 <vianac> alinefm, I just need to finish the tests for the snapshot patches
13:06:36 <vianac> and the one related to the suspended state is still being done
13:06:37 <pvital> #info pvital submitted patch to remove firewalld as Debian dependency from README and sources to create deb packages
13:07:22 <alinefm> vianac, ok
13:07:28 <royce> alinefm, when I resolved the fh leak, found libvirt upload stream can cause volume corruption when doing concurrently, do we need to give each volume a lock, and limit the concurrent upload to like 5 uploads?
13:08:16 <alinefm> royce, I suppose the issue in only uploading many files to the same pool
13:08:25 <alinefm> or it does not matter?
13:08:58 <royce> and about the fh leak of PAM validation, I felt putting in a process is urgly, but found no better solution, any idea?
13:09:36 <royce> alinefm, let me find the code
13:09:52 <alinefm> ok
13:10:53 <royce> two issues 1. file handler fixing 2. uploading lock
13:11:58 <royce> wait a sec, vpn blocks me
13:12:07 <alinefm> sure
13:13:30 <alinefm> let's move on to open discuss so we can continue
13:13:34 <alinefm> #topic Open Discussion
13:14:00 <alinefm> anyone else has more topics?
13:14:00 <YuXin> about web font icon library, aline, do you still have any concern?
13:14:03 <pvital> alinefm, I have a topic
13:14:31 <YuXin> I replied a mail with its license statements
13:14:56 <YuXin> all files in that patch are needed
13:16:44 <royce> +            st = self.conn.get().newStream(0)
13:16:45 <royce> +            st.upload(vol, pos, chunk_size, 0)
13:16:45 <royce> +            st.send(params['chunk'].fullvalue())
13:16:45 <royce> +
13:16:45 <royce> +            st.finish()
13:16:46 <alinefm> YuXin, I am just waiting from a reply from baude
13:16:51 <pvital> alinefm, kimchi is using some python libraries that are not available as packages to Debian Stable (version 7 - codiname: wheezy) - the current README, shows apt-get command that works only on Ubuntu and Debian Testing (next version 8 - codiname: jessy)
13:16:57 <YuXin> ok
13:17:38 <alinefm> pvital, we don't have anyone testing kimchi on debian AFAIK
13:17:56 <alinefm> pvital, we have 2 distro focal points: for Fedora and openSUSE
13:18:08 <pvital> alinefm, my sugestion is express on README to officially support Debian starting on  Jessy version
13:18:08 <alinefm> so probably there will be problem on debian
13:18:38 <alinefm> pvital, if you already confirmed that, please, send a patch
13:19:28 <royce> The above peice of code need a lock to synchronize because each volume can just have one active stream, problem is if two concurrent requests come in, we need to distinguish they are for two volumes or for the same volume, which means one lock per volume
13:19:43 <pvital> alinefm, OK! I'm trying to see if there's some other package on stable version that works. But will send a patch!
13:20:14 <wenwang> alinefm: also, after talking with YuXin, we don't think we have enough time to create sample for each widget since time is limited. I have a suggestion that we can write the comment to show how it works in kimchi, is that ok?
13:20:26 <alinefm> pvital, thanks
13:20:50 <alinefm> royce, got it
13:21:30 <alinefm> wenwang, how are you testing the code without a sample?
13:21:49 <wenwang> we are building the widget based on kimchi
13:21:58 <royce> but we don't want so many locks, we can just afford some locks I call it the active uploading limit
13:22:32 <royce> the lock is global because we want it survive between requests
13:22:33 <YuXin> we write the widget on top of current kimchi UI content, so we can see the display, but only send patch with widget content
13:23:40 <alinefm> royce, yeap
13:23:48 <YuXin> after need week, widget work will be done. then we will start to work on the Kimchi UI content by using these widgets created
13:24:15 <YuXin> aline, I think currently you do not need to merge the widget patches
13:24:22 <alinefm> YuXin, wenwang, so input in the commit message, the code to use the widget
13:24:30 <alinefm> so I can create a sample from it
13:24:41 <alinefm> YuXin, why not?
13:25:20 <wenwang> sure alinefm
13:25:29 <YuXin> when we start to build kimchi UI content with these widgets, each time when we using a widget, we will re-send the patch with both kimchi UI and the widget
13:25:59 <YuXin> then team can review the widget and the kimchi funtional UI content at the same time
13:26:14 <YuXin> this way will save time, is that ok?
13:26:35 <alinefm> YuXin, the reason why I think the samples are important is because they can be reused by plugins and so
13:26:42 <alinefm> so anyone would be able to reuse them
13:26:58 <alinefm> I prefer to merge the widgets patches as soon as they are ready to avoid any delay
13:27:15 <YuXin> but these widgets are kimchi specific, they can only be used inside kimchi
13:27:18 <alinefm> if you input the widget usage in the commit message I can send a separated patch with a sample
13:27:32 <YuXin> if we create sample, in the sample
13:27:48 <YuXin> 1. we need a html which is url addressable to be loaded from cherry web server
13:28:19 <alinefm> the widget samples will not be accessible through kimchi
13:28:34 <YuXin> 2. in this html, we need to reference kimchi css files and js files, so this html need to be placed somethere in kimchi source that can access kimchi web resoures
13:29:06 <YuXin> if the sample can not be accessed from kimchi, how it can be tested
13:29:16 <alinefm> firefox sample.html
13:29:39 <YuXin> load a local html file with filebox, you mean?
13:29:51 <alinefm> yes
13:29:54 <alinefm> just for testing
13:30:10 <alinefm> it will not be packaged or anything
13:30:22 <alinefm> it only will be on source code for devel matters
13:30:26 <YuXin> wenwang, have you tried whether the sample can work if load it locally?
13:30:57 <alinefm> YuXin, I did =)
13:31:16 <YuXin> ok
13:31:27 <YuXin> so the sample will be writen this way
13:31:46 <YuXin> 1. it will be placed in a folder in the source
13:31:58 <YuXin> 2. it reference to kimchi css/js files with relative patch
13:32:13 <wenwang> YuXin: I have tried saperately, it works. But within kimchi, there might be a little problem running independently. But can work in kimchi project
13:32:14 <YuXin> 3. it can work if load locally with browser
13:32:39 <YuXin> alinefm, it works inside kimchi is ok, right?
13:32:41 <alinefm> YuXin, that is the way wenwang did the samples I tested
13:33:08 <YuXin> I mean placed inside kimcih source
13:33:25 <alinefm> there is a directory for it ui/js/widgets/samples/.
13:33:31 <alinefm> in the next branch
13:34:14 <YuXin> then only work when loading the sample html from this directory is ok, right?
13:34:21 <alinefm> sure
13:34:28 <alinefm> sounds pretty good for me
13:34:57 <YuXin> ok
13:35:19 <alinefm> anything else for today?
13:35:36 <YuXin> I send a mail with 3 identified chart library
13:36:02 <YuXin> I did not get which one is strickly better than others
13:36:21 <YuXin> so would like team to take a look see whether any idea
13:36:27 <alinefm> sure
13:36:35 <alinefm> is there some online tool to test them?
13:36:51 <alinefm> (I haven't checked the email yet)
13:36:51 <YuXin> 3 web sites for them
13:36:58 <YuXin> there is is demo
13:37:03 <YuXin> doc about API
13:38:08 <YuXin> those charts are fully customizable
13:38:38 <alinefm> great!
13:38:47 <alinefm> I will do some tests later today and send my comments to ML
13:38:48 <YuXin> for example, line chart, the axis, tick, line boldness, color, grid backgroud, tooltip, all can be customized
13:39:22 <YuXin> take a look at its API, there is a big list of options for customizatoin
13:39:40 <alinefm> will do
13:40:02 <YuXin> I will prototype a line chart with ChartJS tomorrow follow new kimchi UI design
13:40:41 <alinefm> ok
13:40:45 <YuXin> ok
13:41:14 <YuXin> all for me today
13:41:54 <alinefm> vianac, royce, wenwang, anything else from your side?
13:43:07 <wenwang> nope from me
13:43:13 <royce> me neither
13:43:34 <alinefm> ok...
13:43:40 <alinefm> thanks everyone for joining!
13:43:44 <alinefm> #endmeeting