08:04:55 <takahashi-tsc> #startmeeting tacker
08:04:55 <opendevmeet> Meeting started Mon Aug 25 08:04:55 2025 UTC and is due to finish in 60 minutes.  The chair is takahashi-tsc. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:04:55 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
08:04:55 <opendevmeet> The meeting name has been set to 'tacker'
08:05:10 <takahashi-tsc> #link https://etherpad.opendev.org/p/tacker-meeting
08:05:47 <takahashi-tsc> There seems to be 3 topics, but there is no update about some of them. anyway, let's check one by one.
08:06:08 <takahashi-tsc> Maintenance for tosca-parser and heat-translator
08:06:54 <takahashi-tsc> Maybe, there is still ongoing... Status of them seem to be -1. I'll review them when zuul fails are resolved.
08:06:57 <takahashi-tsc> Any comments?
08:07:57 <takahashi-tsc> good, then let's move to next topic.
08:08:34 <takahashi-tsc> Regarding Update for 2025.2 Flamingo Release, About the toggle key name for Cinder Volume Recreation in v1 Heal.
08:09:04 <takahashi-tsc> hi-koba: Did you discuss anything at IRC last week? sorry but I was not able to attend last week meeting.
08:10:12 <hi-koba> Yes, I talked about it last week.
08:11:44 <takahashi-tsc> Is there anything of discussion which are still unfinished?
08:12:46 <hi-koba> In conclusion, the toggle key name for the purpose already exists in the NFV standard, so I will try to modify the code as much as possible.
08:13:40 <hi-koba> There is nothing left to discuss today.
08:13:52 <takahashi-tsc> It will align to the NFV standard as much as possible, right?
08:15:37 <hi-koba> yes.
08:16:25 <takahashi-tsc> Thanks. If there is nothing left, could you please move this topic to the bottom? move to out of 1. Topics for the next meeting.
08:17:35 <hi-koba> Sorry. Understood.
08:18:14 <takahashi-tsc> Thanks
08:18:21 <takahashi-tsc> Then, let's move to the last topic.
08:18:34 <takahashi-tsc> Tosca-parser 2.13.0 Validation Checks Causing Failures for Tacker Packages
08:18:43 <takahashi-tsc> shivam: is there any update?
08:18:50 <shivam> yes
08:19:35 <shivam> This topic is the continuation of previously discussed "Zuul Errors For tacker Patches"
08:20:05 <shivam> Apart from other issues, some of the recent Zuul job failures in Tacker are due to stricter validation rules added in Tosca-parser new release 2.13.0.
08:20:38 <shivam> Tosca-parser patch which introduces new validation checks: https://review.opendev.org/c/openstack/tosca-parser/+/901762
08:21:24 <shivam> The new validation check assumes default occurrences as [1, UNBOUNDED] if not explicitly defined, which forces at least one requirement.
08:21:50 <shivam> Some Tacker package descriptor/definition files did not define occurrences, so they started failing validation.
08:22:46 <shivam> VduCp nodes were affected because in SOL001 v2.6.1 their requirements (virtual_link, virtual_binding) had no occurrences defined, while many Tacker packages rely on substitution mapping instead of defining a virtual_link directly.
08:23:41 <shivam> This mismatch caused validation failures as there is no virtual_link defined and by default, Tosca parser expects at least one virtual_link.
08:24:31 <shivam> On comparing with latest specification, SOL001 v5.1.1 explicitly defines occurrences [0,1] for these requirements, which aligns with current expectations from Tosca-parser.
08:25:11 <shivam> As a workaround, Tacker’s local copy of etsi_nfv_sol001_vnfd_types.yaml was updated to add occurrences [0,1] for virtual_link and virtual_binding, and other missing places.
08:25:38 <shivam> With this change, Tacker VNFDs now passes parser validation.
08:26:28 <shivam> However, this fix ideally belongs in the SOL001 v2.6.1 specification itself, since Tacker only references it. Hence, it must be fixed in the specification itself rather than changing anything in Tacker's referenced copy.
08:27:25 <shivam> Since this is really a specification gap in SOL001 v2.6.1, I would like to hear community opinions on our approach such as raising this issue formally to ETSI so that the specification itself is updated to reflect [0,1] occurrences for VduCp  requirements, ensuring alignment latest Tosca-parser updates
08:28:38 <shivam> Please refere the patch raised for fixing the zuul issues (including Tosca-parser issue as well) : https://review.opendev.org/c/openstack/tacker/+/956458
08:28:53 <shivam> Thank you.
08:29:42 <takahashi-tsc> Thank you. because the the error itself have already been fixed, so I think we can find the details by looking at the patch.
08:30:30 <takahashi-tsc> In my understanding, the discussion topic is, how to maintain the NFV spec fix in the latest version, which is not reflected to v2.6.1,
08:30:34 <takahashi-tsc> is it correct?
08:30:41 <shivam> yes
08:31:58 <takahashi-tsc> OK, I personally believe we need to discuss Tacker's strategy, but no need to rush.
08:32:36 <takahashi-tsc> Firstly, we should merge the bug fix patch ASAP, and discuss the maintenance strategy at later, for example, next PTG.
08:32:58 <takahashi-tsc> It is just my personal suggestion, so any other comments?
08:35:50 <takahashi-tsc> Any objection? If there is no objection, we leave this topics and proceed with only bug fix in this release.
08:38:07 <hi-koba> Is it possible to pin the version of the Tosca parser used to match the SOL version used by Tacker?
08:39:33 <shivam> Tosca-parser version 2.13.0
08:40:50 <takahashi-tsc> I think it is possible (need to check), but the problem is, this change in tosca-parser itself is *meaningful*, and if possible, SOL v2.6.1 should be fixed.
08:41:21 <takahashi-tsc> I heard that why SOL v2.6.1 is not fixed is, just a procedure matter of NFV.
08:41:53 <hi-koba> I see.
08:42:35 <takahashi-tsc> That is also why I said it will be discussed at IRC. To fix the current issue, any approach is OK. Shivam's approach, use old tosca-parser, both are OK.
08:42:55 <takahashi-tsc> But discussion topic is not only this issue, but regarding Tacker's policy.
08:43:15 <takahashi-tsc> s/IRC/PTG/
08:44:26 <takahashi-tsc> OK, shivam, do you have any comments on this discussion?
08:44:48 <shivam> no, its ok from side
08:45:03 <takahashi-tsc> Thanks, any other comments everyone?
08:45:45 <hi-koba> ok. It was a simple question. No objections.
08:46:06 <takahashi-tsc> Thanks.
08:47:39 <shivam> Thank you
08:48:01 <takahashi-tsc> OK, any other topics? if nothing I'd like to close the meeting.
08:50:13 <takahashi-tsc> OK, then let's close the meeting. Thank you, bye!
08:50:23 <shivam> ok. Thank you
08:50:43 <takahashi-tsc> #endmeeting