17:03:56 #startmeeting murano 17:03:57 Meeting started Tue Dec 3 17:03:56 2013 UTC and is due to finish in 60 minutes. The chair is ativelkov. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:03:58 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:04:01 The meeting name has been set to 'murano' 17:05:13 So, let's follow the agenda at https://wiki.openstack.org/wiki/Meetings/MuranoAgenda 17:06:19 The only action item which we had was on sergmelikyan about the screencast for the release 0.3 17:06:53 Serge, do you have any progress on that? Did you speak to Nick Chase about it? 17:08:00 sergmelikyan, are you here? 17:08:19 ativelkov, as far as I know all slides are ready and video also almost done 17:08:49 so at the end of the week we will have screencast 17:08:52 Ok, seems like Serge is away, katyafervent thanks for update 17:09:11 #info slides for 0.3 presentation are ready, video is coming later this week 17:09:52 The question on elaboration with Nick still remains, but that can be answered only by Serge, probably 17:10:26 #topic Release 0.4 status 17:10:42 Let's move on: tsufiev, what are the news on 0.4? 17:11:17 1. Considering Metadata Repository, we're currently fixing bugs in Admin WebUI 17:12:00 2. Also two problems recently arised in 0.4 due to Advanced Networking implementation 17:12:10 But all functional for Metadata Repository usage from UI is totally implemented 17:12:43 #info all the functionality for Metadata Repository usage from UI is implemented 17:12:46 a. One is related to Security Groups and dynamically generated CIDRs 17:13:22 b. The second is related to MS SQL Cluster - its IPs validation and, again, dynamically generated CIDRs 17:13:42 I'm currently implementing fix for (b), planning to finish tomorrow 17:13:57 ativelkov: I think you can tell us more on (a) 17:14:08 #info: New major issues in 0.4: security group scopes do not match new generated subnet ciders, MS SQL Cluster IP address does not match is as well 17:14:22 that's all yet to be fixed in 0.4 17:14:33 Yes, this is also in progress, also is likely to be fixed by tomorrow 17:15:02 The fix itself is ready, I just have to update all he workflows to make use of the new steps which fix the problem 17:15:33 tsufiev: do we have bug reports on launchpad for these issues? 17:16:14 ativelkov: I think no, should make new tickets 17:16:25 Yes, please do that 17:16:33 ok 17:16:53 #action tsufiev to create launchpad issues to track bugs with security group cords and MS SQL cluster IP 17:16:59 And for now we need to fill up documentation with Murano Repository information 17:17:24 Same should probably go for advanced networking: we need to update the blueprint 17:17:29 I'll do it 17:17:43 I am started to work on api specification transtation to docbook 17:17:57 #action tsufiev to implement the subnet detection routing in the dashboard 17:18:19 #action ativelkov to fix the security groups cidd assignment 17:18:34 #action ativelkov to update the blueprint on advanced networking 17:18:58 #action katyafervent to transfer api spec to docbook 17:19:19 Anything else on 0.4? 17:20:04 we should document advanced networking also 17:20:29 I've just mentioned that, see above 17:20:31 there are some configuration details which are not obvious 17:20:36 ok 17:20:58 Ok, moving to the next topic 17:21:06 #topic Neutron LBaaS news 17:21:21 sergmelikyan: are you still away? 17:21:57 Probably yes. So, I'll give an update myself, as I know some details on the topic 17:22:56 As far as I know, sergmelikyan has implemented the LoadBalancer Murano Service using the Neutron LBaaS underneath, according to the proposal published earlier 17:23:33 The pull request can be found here: https://review.openstack.org/#/c/59705/ 17:23:50 There are some -1's there, so this is work in progress 17:24:22 #info elastic apps transition to OS::Neutron::LoadBalancer is in progress, code is being review 17:25:17 Also, the new service will require to have LB IP assigned by the user 17:26:15 And this faces the same problem as earlier mentioned issue with MS SQL cluster IP: we'll need to assign the IP which is located in the proper subnet, while the actual subnet cidd is not know at the time when the environment is being composed 17:27:16 So, this issue will be fixed by the same approach as we use for MS SQL (i.e. the fix which is being developed by tsufiev) 17:27:48 #info final implementation of LoadBalancer service depends on the fix by tsufiev 17:28:00 That's all I know on this topic ) 17:28:54 What's next? DSL and Metadata description? 17:29:27 igormarnat_: well, we don't have this in agenda for today ) Probably because we don't have any final vision yet 17:29:56 The last topic in agenda is "Security Groups issue" but we have actually covered it when discussed current 0.4 status 17:30:13 so, we can move to the open discussion :) 17:30:20 #topic Open Discussion 17:30:34 Ok, what is the status of DSL and Metadata description? :) 17:30:43 Any etherpad? 17:31:50 There is a first draft of the etherpad (https://etherpad.openstack.org/p/MuranoMetadata) but it is not complete yet, as the workflow definition language is still being discussed 17:32:45 We have planned an offline design session for the workflow definiiotn on this Thursday 17:33:06 The results will be reflected in this ether pad and will be suggested for further online discussion 17:33:46 Ok. Does it make sense to keep the workflow definition in separated etherpad,, given that this is not a metadata (which current etherpad is actually about)? 17:34:30 Actually, workflows are part of the metadata 17:34:51 and eventually they can be even exilicitly included into the metadata manifest file 17:35:17 We just propose to split them in sake of simplicity and readability, as they most likely will be long 17:35:39 Ok, so we have couple of days to complete the exercise and take a look at the draft of the DSL and Metadata proposal? 17:35:46 but in case of simple workflows, the developer may combine both app definitions and workflows in the same file - altogether with the manifest itself 17:35:55 Yes 17:38:04 #info DSL definition and workflow spec to be completed in several days 17:38:10 Any other questions? 17:38:59 If not, then seems we are done 17:39:31 Thanks for joining 17:39:41 #endmeeting