13:02:17 <joehuang> #startmeeting tricircle 13:02:18 <openstack> Meeting started Wed Jul 6 13:02:17 2016 UTC and is due to finish in 60 minutes. The chair is joehuang. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:02:19 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:02:21 <openstack> The meeting name has been set to 'tricircle' 13:02:41 <joehuang> #topic rollcall 13:03:01 <joehuang> #info joehuang 13:03:05 <zhiyuan> #info zhiyuan 13:03:06 <Yipei_> #info Yipei 13:03:14 <huangdongfeng> #info dongfeng 13:04:25 <joehuang> hello Nova api 2.0 will be deprecated soon 13:04:52 <joehuang> so tricircle needs to call Nova 2.1 api too 13:06:16 <joehuang> hello, could you see the pacth in tricircle "bump to Nova 2.1"? 13:06:55 <Yipei_> yes, i saw it 13:06:55 <joehuang> only a little update is needed for https://review.openstack.org/#/c/324226/ 13:07:11 <Administrator_> hello 13:07:46 <Administrator_> :) 13:07:58 <joehuang> Administrator_, who are you? 13:08:00 <zhiyuan> could you change your nick name? 13:08:03 <joehuang> you can use /NICK to change to your name 13:08:35 <zhiyuan> shall we update the patch for zhengzhenyu, or send him a mail? 13:09:43 <ronghui01> hello 13:12:54 <joehuang> hello, my network connection was broken for a while 13:12:57 <joehuang> I contacted zhengyu, he is very busy, would someone update the commit message, then the patch could be merged soon 13:14:27 <ronghui01> i can try it 13:14:28 <joehuang> hello, can you see my message, I am afraid offline again 13:14:37 <joehuang> good! 13:14:43 <joehuang> hi, Ronghui 13:16:43 <ronghui01> i will update the commit message which from zhiyuan's suggest in https://review.openstack.org/#/c/324226/ 13:17:04 <joehuang> good, thank you ronghui. 13:17:19 <ronghui01> that's ok:) 13:17:22 <joehuang> #action bump to Nova 2.1 in Tricircle 13:17:59 <joehuang> there are some bugs reported in launch pad : https://bugs.launchpad.net/tricircle 13:18:53 <Yipei_> i take one, but still learn to how to fix it 13:19:30 <zhiyuan> ok, so you can assign the bug to you 13:23:26 <Yipei_> yes, i did. in #1562441, we need to use different ipv4 addresses belong to all kinds of addresses, respectively? 13:23:27 <zhiyuan> hi joe, maybe you can try irssi in our server 13:24:09 <joehuang> my link was broken again 13:24:30 <joehuang> to Zhiyuan, I remember this bug " #1566184" has been fixed in your patch? 13:24:38 <longxiongqiu> hi 13:24:47 <joehuang> hi xiongqiu 13:25:45 <zhiyuan> no, my patch has not changed the static bridige network CIDR yet 13:26:33 <joehuang> hello, this bug is "when getting client for top region, 'top' should be replaced with a constant" 13:26:55 <joehuang> ok 13:27:24 <zhiyuan> yes, i think that's another bug. 13:28:00 <zhiyuan> 1562441 is about the static CIDR 13:28:52 <joehuang> I think dongfeng can start from fixing some bugs which is easy to work on 13:30:11 <huangdongfeng> oh, I am beginning to learn the devstack and tricircle. but i can try with the help of Yipei. 13:30:44 <joehuang> good, how about the issue you shared in the mail-list 13:31:13 <huangdongfeng> zhiyuan helped me fix it. 13:31:33 <zhiyuan> oh yes, you can submit a patch to modify the configuration sample 13:31:40 <joehuang> good, you can submit a patch and fix it in the trunk 13:32:24 <huangdongfeng> ok, i can't log onto the openstack today, i will do as soon. 13:33:15 <joehuang> #action fix the local.conf.sample 13:33:59 <joehuang> there is one interesting patch in Neutron which will be helpful for L2 networking across OpenStack 13:34:02 <joehuang> https://review.openstack.org/#/c/282180/ 13:34:28 <joehuang> Add Unicast Flooding Vteps to Provider Network 13:35:05 <joehuang> This patch will be very helpful if the L2 network only needs to across very limited OpenStack instances 13:35:26 <joehuang> for it uses flooding for BUM 13:37:04 <joehuang> we can add this patch into watch list, and update cross pod l2 networking if needed 13:38:16 <zhiyuan> with this patch, host in OpenStack can directly connect to l2 gateway in the other OpenStack via VxLAN tunnel. in our previous solution, host needs to first connect to local l2 gateway, then to remote l2 gateway 13:39:24 <joehuang> yes, it provides the possibility to extend L2 network to remote site 13:40:53 <joehuang> the only issue is that if there are a list of remote VTEP, then a multicast will happen, if the local can learn which remote VTEP reponds, and send later patcket to this vtep only 13:41:22 <joehuang> then it will work in list of VTEPs 13:42:03 <joehuang> other wise, I am afraid that the packets will be replicated to multiple destination, and always, this is unecessary 13:43:20 <joehuang> let's keep watching on this patch. 13:43:23 <zhiyuan> it depends on how the patch is implemented 13:43:28 <zhiyuan> yes 13:43:39 <joehuang> agree 13:53:10 <zhiyuan> ij 13:53:22 <zhiyuan> jkjkdkkdkjkjkj 13:54:03 <Yipei_> Joe is offline 13:55:04 <zhiyuan_> it seems that joe lost connection again 13:55:30 <zhiyuan_> he asks me to end the meeting today 13:55:45 <joehuang> #endmeeting