13:02:36 #startmeeting 13:02:36 Meeting started Wed Apr 1 13:02:36 2015 UTC. The chair is alinefm. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:02:36 Useful Commands: #action #agreed #help #info #idea #link #topic. 13:02:36 #meetingname scrum 13:02:36 The meeting name has been set to 'scrum' 13:03:06 #info Agenda 1) Status 2) Kimchi 1.5 Plan 3) Open discussion 13:03:07 anything else? 13:05:37 ok.. so let's get started 13:05:49 #topic Status 13:05:49 #info Please provide your status using the #info command: #info 13:06:44 #info ziviani Sent a V3 of a patch ([PATCH V3] Number of CPUs and Physical Memory) 13:06:45 #info ziviani Sent a RFC about serial console tool in Kimchi. 13:06:45 #info ziviani I'm currently testing PCI hotplug in X86_64 and PPC64. 13:06:47 #info vianac sent RFC for hotplug a new disk, hotplug a network interface, add suspend/resume, change ref_cnt; sent another revision of fix for #461; 13:06:47 #info alinefm worked for 1.4.1 release: update package, changelog, github main page, etc 13:07:11 #info alinefm sent RFC to make template defaults configurable 13:07:45 #info royce invested about using puppet to setup kimchi, tested the UI bugs on ipad, started writing patch for logical pool for existed VG 13:09:07 royce, could you send a RFC about creating a storage pool from an existing VG? 13:09:19 so we can get agreement on API before you start working 13:09:26 just to make it easier 13:09:58 I thought it is a small feature, so planned to send the patch with API spec... 13:11:11 I will send it tomorrow 13:11:31 royce, thanks 13:11:42 anything else? 13:12:56 #topic Kimchi 1.5 Plan 13:13:06 I have created the schedule in the github wiki: https://github.com/kimchi-project/kimchi/wiki/planning-1.5 13:13:14 please check the dates in the link 13:13:51 I will create the todo page later today after reading all the RFC on ML and check the open issues 13:14:04 any questions? doubts? suggestion? 13:16:00 vianac, royce, ziviani, danielhb, ^ 13:16:12 alinefm, no 13:16:21 alinefm, I have none 13:17:00 #topic Open Discussion 13:17:07 what do you want to discuss today? 13:19:34 vianac, any update about the needinfo issues? 13:20:29 alinefm, no, any new issue with that status 13:21:20 ok 13:21:51 vianac, royce, danielhb, ziviani, if you don't have anything else we can finish the meeting for today 13:22:12 alinefm, did you had the time to look at that github issue I talked about yesterday? 13:22:22 I sent a RFC about using puppet to set up kimchi environment, because I find every release it is painful for me to set up a complete test env 13:23:36 alinefm, I believe that Kimchi shouldn't bother resolving host name to IP if qemu can't handle hostnames. It creates a whole new different issue on Kimchi side. I'd rather ask people to download the ISO or update their qemu than to rely on the randomness we have today with the remote streaming 13:25:14 danielhb, just read it 13:25:34 danielhb, removing it now we cause a regression for fedora users 13:25:49 alinefm, it works in F21 13:26:15 danielhb, qemu can resolve hostnames on F21? 13:26:35 alinefm, vianac said that he can't reproduce the issue in his F21. vianac ? 13:27:01 danielhb, are you on f21? 13:27:35 alinefm, danielhb does it work in RHEL 6.x? or is this feature not available in older distros? 13:27:38 alinefm, and even if F21 is bugged, I don't consider it a regression because now Kimchi has the option to download the ISO. And, again, unless Kimchi can guarantee that the remote server will resolve the URL correctly, the feature is flawed 13:28:45 lagarcia, RHEL 6.X does not have support for iso streaming on qemu 13:28:54 alinefm, ok 13:29:44 danielhb, I agree but I suggest to find a way to improve the hostname-IP resolution Kimchi is doing before removing this 13:30:25 danielhb, for example, I have tested long time ago, the resolution failed in the original URL redirects to other one 13:30:47 I think we first need to get the final URL and then resolve the hostname 13:31:00 so it may work in all cases 13:31:12 *but* we need to test and investigate it is a good solution 13:32:12 alinefm, I believe it will fail the same way as it does now. If you insert an URL like http://n.n.n.n:80/my_link there is no guarantee that the webserver will redirect you to http://nameserver.com/my_link 13:32:36 alinefm, we can't control how the remote server resolves the incoming URL 13:33:06 alinefm, but I do not oppose testing other ways before ditching this design 13:33:53 brb 13:34:48 danielhb, we need also to confirm all the current qemu versions still have this issues on resolving hostnames 13:34:58 otherwise we can remove it IMO 13:36:21 royce, about the complete test env, vianac is using a script to do that on his vms 13:37:59 danielhb, alinefm: on Fedora 21, that issue doesn't exist because its qemu-io version can handle DNS names, so we don't need to convert anything 13:40:45 I will test it on RHEL7, opensuse and ubuntu 13:41:01 probably the qemu versions there is also working now 13:41:08 any other topic for today? 13:42:00 I tried to write script too, but found set up like iscsi server and ldap needs some intervention. So I planned to switch to puppet 13:43:02 oh.. vianac's script is simpler than that 13:43:16 it is just to run "make check" inside the vm 13:43:49 Oh, it's like jenkins, right? 13:44:27 I just meant to make user easier to set up a sand box to play with kimchi's all features 13:45:01 got it 13:48:41 any other topic? 13:49:33 no 13:51:53 thanks everyone for joining! 13:51:56 #endmeeting