| *** prashantD has joined #tacker | 00:01 | |
| bobh_ | sripriya: I think I see it - "fixed_ip = []" and then "fixed_ip['ip_address']" - should probably be fixed_ip = {} I think | 00:01 |
|---|---|---|
| sripriya | bobh_: yes i was just looking into that code | 00:02 |
| bobh_ | sripriya: I thought they had tests for this - maybe not | 00:02 |
| sripriya | bobh_: yeah.. may be they are not testing out optional parameters | 00:03 |
| bobh_ | sripriya: I can file the bug and fix it in heat-translator - they should have a point release coming soon | 00:04 |
| *** prashantD_ has quit IRC | 00:05 | |
| sripriya | bobh_: sounds good and you can add a tag affects: tacker | 00:05 |
| bobh_ | sripriya: will do | 00:06 |
| *** uck has quit IRC | 00:09 | |
| *** santoshk has quit IRC | 00:13 | |
| openstackgerrit | Merged openstack/python-tackerclient: Cleanup copyright header https://review.openstack.org/297891 | 00:20 |
| openstackgerrit | Merged openstack/tacker: Remove unused attach/detach interface method. https://review.openstack.org/298630 | 00:21 |
| *** prashantD has quit IRC | 00:37 | |
| *** s3wong_ has joined #tacker | 00:42 | |
| *** s3wong has quit IRC | 00:42 | |
| *** sripriya has quit IRC | 00:43 | |
| *** s3wong_ has quit IRC | 00:44 | |
| *** s3wong has joined #tacker | 00:46 | |
| *** lhcheng has quit IRC | 00:48 | |
| openstackgerrit | Merged openstack/tacker: Updated from global requirements https://review.openstack.org/301396 | 00:52 |
| openstackgerrit | Merged openstack/tacker: Fix typos in Tacker files https://review.openstack.org/299749 | 00:53 |
| *** Tacker has joined #tacker | 01:32 | |
| *** Tacker has quit IRC | 01:35 | |
| *** rjankows has joined #tacker | 01:36 | |
| rjankows | Hello, my name is Richard Jankowski and I work at Ciena. I'm trying to get Tacker monitoring to work using some of the wiki documentation out there. I find that the docs are lacking some info. Is it appropriate to use this channel for asking questions? This is my first time using IRC. | 01:41 |
| s3wong | rjankows: this is the right channel | 01:43 |
| s3wong | rjankows: particularly for monitoring, you can ping bobh_ | 01:43 |
| rjankows | Thank you | 01:44 |
| bobh_ | Hi rjankows - did you have specific questions? | 01:44 |
| rjankows | Yes, there is one wiki that contains installation, getting started, and monitoring sections. It gives sample YAML syntax for basic "ping" monitoring. Is there a more comprehensive document that covers all possible syntax for the VNFD template? | 01:47 |
| *** sridhar_ram has quit IRC | 01:47 | |
| bobh_ | rjankows: We don't have a comprehensive document, mostly just sample templates in the devstack/samples directory, and also under the tests hierarchy | 01:48 |
| bobh_ | rjankows: we just added TOSCA support in the Mitaka so our focus will be on getting that documented and the old YAML format will be deprecated and probably not documented any more than it is now | 01:48 |
| bobh_ | rjankows: note that there are some known issues with monitoring in the TOSCA implementation that we are working to fix | 01:49 |
| rjankows | Ok...thanks. I've seen the devstack/samples directory. I'll check out "tests". Am using Liberty. Do you recommend moving to Mitaka? | 01:49 |
| *** s3wong has quit IRC | 01:50 | |
| bobh_ | rjankows: It depends on your needs. There is definitely more flexibility and options in the TOSCA format, and Mitaka is a big step forward in several other areas | 01:50 |
| *** prashantD has joined #tacker | 01:51 | |
| bobh_ | rjankows: it might be a bit "bleeding edge" - it will take a couple of weeks for all of the bugs to show up | 01:51 |
| rjankows | Thanks...I'll stick with Liberty for now...just trying to demonstrate ping-based monitoring... | 01:52 |
| bobh_ | rjankows: it works in Liberty using the YAML format - let me know if you have questions or run into problems | 01:53 |
| rjankows | Final question for the night: When I launch a heat template, the spawned VM has an associated security group. When I create a VNF, the spawned VM does not. Is there a way to specify security group in the VNFD? | 01:55 |
| *** tbh has joined #tacker | 01:56 | |
| bobh_ | rjankows: I don't think we support security group in the YAML format. I'm not sure it's in the basic TOSCA format either - we'll have to look at adding support for it in the Newton cycle | 02:14 |
| bobh_ | rjankows: I confirmed it's not supported in either format. We can add it to the TOSCA format pretty easily, though I'm not sure exactly what the property name would be in the TOSCA format. | 02:20 |
| rjankows | Thanks! | 02:22 |
| *** tbh has quit IRC | 02:31 | |
| *** rjankows has quit IRC | 02:40 | |
| *** prashantD has quit IRC | 02:48 | |
| *** lhcheng has joined #tacker | 02:52 | |
| *** gongysh_ has joined #tacker | 03:14 | |
| *** bobh_ has quit IRC | 03:50 | |
| *** gongysh_ has quit IRC | 03:51 | |
| *** tbh has joined #tacker | 04:15 | |
| *** Shahid_ has joined #tacker | 04:58 | |
| *** lhcheng has quit IRC | 06:10 | |
| *** Shahid_ has quit IRC | 07:13 | |
| *** lhcheng has joined #tacker | 07:43 | |
| *** mbound has joined #tacker | 07:54 | |
| *** lhcheng has quit IRC | 07:54 | |
| *** mbound has quit IRC | 08:00 | |
| *** zeih has joined #tacker | 08:05 | |
| *** mbound has joined #tacker | 08:06 | |
| *** tbh has quit IRC | 08:19 | |
| *** Shahid_ has joined #tacker | 08:24 | |
| *** tbh has joined #tacker | 08:32 | |
| *** Shahid_ has quit IRC | 08:37 | |
| *** Shahid_ has joined #tacker | 08:51 | |
| *** Liuqing has joined #tacker | 08:54 | |
| *** yifei has quit IRC | 08:54 | |
| *** yifei has joined #tacker | 09:02 | |
| *** tbh has quit IRC | 09:06 | |
| *** zeih has quit IRC | 09:07 | |
| *** tbh has joined #tacker | 09:19 | |
| *** Liuqing has quit IRC | 09:21 | |
| *** Liuqing has joined #tacker | 09:57 | |
| *** zeih has joined #tacker | 10:38 | |
| *** janki has joined #tacker | 10:43 | |
| *** janki has quit IRC | 10:44 | |
| *** mbound has quit IRC | 11:24 | |
| *** mbound has joined #tacker | 11:39 | |
| *** mbound has quit IRC | 11:39 | |
| *** mbound has joined #tacker | 11:40 | |
| *** mbound has quit IRC | 11:40 | |
| *** mbound has joined #tacker | 12:11 | |
| *** tbh has quit IRC | 12:34 | |
| *** zeih has quit IRC | 12:37 | |
| *** zeih has joined #tacker | 12:39 | |
| *** zeih has quit IRC | 12:47 | |
| *** igordcard has joined #tacker | 13:01 | |
| *** igordcard has quit IRC | 13:02 | |
| *** igordcard has joined #tacker | 13:03 | |
| *** igordcard has quit IRC | 13:21 | |
| *** dkushwaha has quit IRC | 13:21 | |
| *** mbound has quit IRC | 13:21 | |
| *** Liuqing has quit IRC | 13:21 | |
| *** vishwanathj has quit IRC | 13:21 | |
| *** yifei has quit IRC | 13:21 | |
| *** trozet has quit IRC | 13:21 | |
| *** openstackgerrit has quit IRC | 13:21 | |
| *** Shahid_ has quit IRC | 13:21 | |
| *** gongysh has quit IRC | 13:22 | |
| *** zeih has joined #tacker | 13:24 | |
| *** igordcard has joined #tacker | 13:24 | |
| *** mbound has joined #tacker | 13:24 | |
| *** Liuqing has joined #tacker | 13:24 | |
| *** yifei has joined #tacker | 13:24 | |
| *** Shahid_ has joined #tacker | 13:24 | |
| *** trozet has joined #tacker | 13:24 | |
| *** vishwanathj has joined #tacker | 13:24 | |
| *** openstackgerrit has joined #tacker | 13:24 | |
| *** gongysh has joined #tacker | 13:24 | |
| *** dkushwaha has joined #tacker | 13:24 | |
| *** trozet has quit IRC | 13:37 | |
| *** trozet has joined #tacker | 13:37 | |
| *** bobh_ has joined #tacker | 13:47 | |
| *** mbound has quit IRC | 13:55 | |
| *** Shahid_ has quit IRC | 13:59 | |
| *** mbound has joined #tacker | 14:15 | |
| *** zeih has quit IRC | 14:52 | |
| *** zeih has joined #tacker | 14:53 | |
| *** zeih has quit IRC | 15:26 | |
| *** Liuqing has quit IRC | 15:38 | |
| *** bharath has joined #tacker | 15:52 | |
| *** sripriya has joined #tacker | 15:52 | |
| *** sripriya has quit IRC | 16:03 | |
| *** lhcheng has joined #tacker | 16:03 | |
| *** bharath has quit IRC | 16:16 | |
| *** janki_chhatbar has joined #tacker | 16:20 | |
| *** tbh has joined #tacker | 16:41 | |
| *** mbound has quit IRC | 16:48 | |
| *** brucet has joined #tacker | 16:53 | |
| *** sripriya has joined #tacker | 16:55 | |
| *** sridhar_ram has joined #tacker | 16:57 | |
| *** s3wong has joined #tacker | 16:59 | |
| *** prashantD has joined #tacker | 17:06 | |
| *** santoshk has joined #tacker | 17:17 | |
| *** sripriya has quit IRC | 18:02 | |
| *** s3wong has quit IRC | 18:04 | |
| *** janki_chhatbar has quit IRC | 18:13 | |
| *** tbh has quit IRC | 18:17 | |
| *** Janki has joined #tacker | 18:22 | |
| *** Janki has quit IRC | 18:23 | |
| *** brucet has quit IRC | 18:23 | |
| *** sripriya has joined #tacker | 18:35 | |
| sripriya | sridhar_ram: bobh_: can you please help review https://review.openstack.org/#/c/299124/. i would like to cherry pick this to Mitaka release | 18:54 |
| bobh_ | sripriya: will do | 18:54 |
| sripriya | bobh_: thanks | 18:55 |
| *** bobh_ has quit IRC | 19:01 | |
| trozet | sridhar_ram: ping? | 19:40 |
| openstackgerrit | Sripriya Seetharam proposed openstack/tacker-horizon: VNF name and VIM name should be optional https://review.openstack.org/301882 | 19:41 |
| trozet | sripriya: ping? | 19:45 |
| sripriya | trozet: pong | 19:45 |
| trozet | sripriya: hi. Looking through tacker code for vnfm extension, why is there no vnfd-file listed as an attribute? | 19:46 |
| sripriya | trozet: the file content gets punted under attributes with 'vnfd' key befor hitting the API | 19:49 |
| sripriya | trozet: https://github.com/openstack/python-tackerclient/blob/master/tackerclient/tacker/v1_0/vm/vnfd.py#L66 | 19:49 |
| trozet | sripriya: ah so its put into attributes | 19:50 |
| trozet | sripriya: typ | 19:50 |
| trozet | sripriya: ty* | 19:50 |
| sripriya | trozet: yeah that was the design earlier and we retained it, with new apis/extensions we want to get out of punting all params under attributes and call them out individually | 19:51 |
| trozet | sripriya: makes sense, just wanted to mirror it for the VNFFGD | 19:51 |
| *** bobh_ has joined #tacker | 20:01 | |
| *** bobh_ has quit IRC | 20:07 | |
| *** s3wong has joined #tacker | 20:12 | |
| trozet | sridhar_ram, sripriya, s3wong: do you understand how the substitution_mappings in the tosca template work? I don't get how it helps | 20:15 |
| s3wong | trozet: I don't :-( | 20:18 |
| s3wong | trozet: from OASIS, it says: | 20:18 |
| s3wong | "In short, the substitution_mappings section provides the following information: | 20:18 |
| s3wong | 1. It defines what node templates, i.e. node templates of which type, can be substituted by the topology template. | 20:18 |
| s3wong | 2. It defines how capabilities of the substituted node (or the capabilities defined by the node type of the substituted node template, respectively) are bound to capabilities of node templates defined in the topology template. | 20:18 |
| s3wong | 3. It defines how requirements of the substituted node (or the requirements defined by the node type of the substituted node template, respectively) are bound to requirements of node templates defined in the topology template." | 20:18 |
| s3wong | http://docs.oasis-open.org/tosca/TOSCA-Simple-Profile-YAML/v1.0/csprd01/TOSCA-Simple-Profile-YAML-v1.0-csprd01.html | 20:19 |
| s3wong | (don't think that helps too much :-) ) | 20:19 |
| trozet | s3wong: http://docs.oasis-open.org/tosca/tosca-nfv/v1.0/tosca-nfv-v1.0.pdf section 8.1 | 20:19 |
| trozet | s3wong: im thinking the only thing it does is present: node_type: tosca.nodes.nfv.VNF.VNF2 as an external type, which VNFFG NFP is allowed to reference | 20:21 |
| trozet | s3wong: I guess the CP itself is supposed to be internal to the VNF and VNFFG is not supposed to reference it directly... thats the only thing I can think of | 20:21 |
| s3wong | trozet: from the link I sent above (describing simple profile), it is used to deploy things that are outside of what this template defines | 20:23 |
| *** bobh_ has joined #tacker | 20:23 | |
| s3wong | trozet: for example, the hosting stack has a bunch of tiers of apps to deploy, one of which is database, and the database template is defined somewhere else | 20:24 |
| s3wong | trozet: bobh_ just joined, we can ask him also :-) | 20:24 |
| trozet | s3wong: bobh_ works on the TOSCA stuff? | 20:24 |
| s3wong | trozet: yes | 20:24 |
| trozet | cool | 20:24 |
| trozet | hi bobh_ | 20:24 |
| bobh_ | trozet: hello | 20:25 |
| trozet | bobh_: I'm going through the TOSCA spec, and I'm not understanding why in the Forwarding Path requirements it points to: | 20:25 |
| s3wong | trozet: in the sample you provided via NFV profile, what I can see is that the VNF2 detail is defined in another template, so you don't need to define VNF2's specific stuff in this template | 20:25 |
| trozet | -forwarder: VNF1, capability: forwarder1 #CP11 (which then points to a substitution_mapping) | 20:25 |
| trozet | s3wong: what other template? | 20:26 |
| bobh_ | trozet: I haven't found any examples in the spec that make a lot of sense | 20:26 |
| bobh_ | trozet: they seem to assume a lot of things that are not obvious to the reader | 20:26 |
| s3wong | trozet: the template which defines tosca.nodes.nfv.VNF.VNF2 | 20:26 |
| trozet | bobh_: what I don't get is why not just directly reference the CP in the NFP | 20:28 |
| trozet | bobh_: it's capability is listed as forwarder, so you can assume it has external connectivity | 20:29 |
| trozet | s3wong: can you tell me which page you are looking at? | 20:29 |
| bobh_ | trozet: which section are you looking at? | 20:29 |
| s3wong | trozet: 16 | 20:29 |
| trozet | bobh_: http://docs.oasis-open.org/tosca/tosca-nfv/v1.0/tosca-nfv-v1.0.pdf page 35 where it shows Forwarding path3 | 20:30 |
| trozet | look at requirements | 20:30 |
| s3wong | trozet: need to step out for a moment. Interviewing someone. Be back in 45 minutes :-) | 20:31 |
| trozet | s3wong: yeah I'm saying why even include that substitution_mappings block, on page 17 CP21 is already listed as a Forwarder, just add its VirtualLink there | 20:31 |
| trozet | bobh_: so you see where it lists the forwarder as VNF1, capability:forwarder1 for example ? | 20:32 |
| *** sripriya_ has joined #tacker | 20:33 | |
| bobh_ | yes | 20:33 |
| trozet | bobh_: so the only place i see forwarder1 defined is if you look at page 16 in subsitution_mappings | 20:35 |
| trozet | bobh_: it shows forwarder1: [CP21, Forwarder], and in the example that is the external CP for VNF2 | 20:35 |
| *** prashantD has quit IRC | 20:36 | |
| bobh_ | trozet: right - bound to VDU2 | 20:36 |
| *** sripriya has quit IRC | 20:36 | |
| *** santoshk has quit IRC | 20:36 | |
| trozet | bobh_: so I don't get why use this substitution_mapping, when you could just list in the forwarding path, the CP21 itself | 20:37 |
| bobh_ | trozet: I suspect that the forwarding path is intended to be described at the VNF level, since it is a collection of VNFs that provide a network service | 20:38 |
| trozet | bobh_: so they dont want to expose CP21 as its an internal VNF port, so they expose the mapping? | 20:38 |
| bobh_ | trozet: The underlying implementation maps to CP21, but it wouldn't necessarily be known at the time that the service was described | 20:38 |
| bobh_ | trozet: That's my guess | 20:38 |
| trozet | bobh_: its a good guess I think | 20:39 |
| *** sridhar_ram has quit IRC | 20:39 | |
| trozet | bobh_: so for the VNFFG extension, would it be bad if I allow just listing the CP, along as the capability type is forwarder | 20:40 |
| trozet | bobh: we dont have a NS extension yet, so I am directly relying on VNFDs and their names... | 20:40 |
| bobh_ | trozet: I think at this point it's valid - one of our tasks as we implement is to figure out what makes sense and what doesn't | 20:41 |
| bobh_ | trozet: so if the CP makes sense we should do it until someone explains why it doesn't | 20:41 |
| trozet | bobh_: OK. In the NS template, they describe VNFs abstractly like you mentioned | 20:41 |
| trozet | bobh_: but I am not implementing that part, so when someone lists the VNFs that should be in the VNFFG, I simply check to make sure VNFDs of those names have been created | 20:42 |
| trozet | bobh_: so it is not abstract for now, but definitely can change it when someone implements an NS extension | 20:42 |
| trozet | so in that sense I rely that the VNFD exists to be able to create a VNFFG | 20:43 |
| bobh_ | trozet: I think that's reasonably for a first step | 20:46 |
| bobh_ | s/reasonably/reasonable/ | 20:46 |
| trozet | bobh_: cool, thanks | 20:49 |
| bobh_ | trozet: np | 20:56 |
| *** lhcheng has quit IRC | 21:06 | |
| *** sripriya__ has joined #tacker | 21:10 | |
| *** lhcheng has joined #tacker | 21:11 | |
| *** sripriya_ has quit IRC | 21:14 | |
| *** sripriya has joined #tacker | 21:33 | |
| *** sripriya__ has quit IRC | 21:35 | |
| openstackgerrit | Tim Rozet proposed openstack/tacker-specs: Adds Tacker VNFFG/SFC spec https://review.openstack.org/292196 | 21:51 |
| trozet | s3wong: please review ^ | 21:53 |
| s3wong | trozet: sure | 21:53 |
| *** bobh_ has quit IRC | 22:08 | |
| *** prashantD has joined #tacker | 22:46 | |
| *** prashantD_ has joined #tacker | 22:47 | |
| *** prashantD has quit IRC | 22:50 | |
| openstackgerrit | Sripriya Seetharam proposed openstack/tacker-horizon: VNF Catalog should handle user input vnfd name https://review.openstack.org/219048 | 23:07 |
| *** sridhar_ram has joined #tacker | 23:36 | |
| * sridhar_ram just realized his irc-client hanged many hrs ago, now catching up some nice discussions... | 23:37 | |
| sridhar_ram | trozet: For read on substituition_mapping please refer to http://docs.oasis-open.org/tosca/TOSCA-Simple-Profile-YAML/v1.0/csprd01/TOSCA-Simple-Profile-YAML-v1.0-csprd01.html#_Toc430015632 | 23:44 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!