====================================
#opendaylight-meeting: kernel-weekly
====================================


Meeting started by vorburger at 17:06:50 UTC.  The full logs are
available at
http://meetings.opendaylight.org/opendaylight-meeting/2019/kernel_weekly/opendaylight-meeting-kernel_weekly.2019-01-22-17.06.log.html
.



Meeting summary
---------------

* NETCONF Test Tool removal  (vorburger, 17:07:45)
  * jmorvay asks if this tool can be removed  (vorburger, 17:07:57)
  * jmorvay explains what this does.. testing tool is a NETCONF server
    which simulates like a switch or router  (vorburger, 17:09:02)
  * it has an option when you start it to specific if you want to use
    the real MD-SAL data store or its local in-memory List (this is not
    the MD-SAL in-memory TestDataBroker)  (vorburger, 17:10:29)
  * jmorvay propose to keep the md-sal backed one, but ditch that fake
    data store  (vorburger, 17:11:32)
  * rovarga says that the fake data store is useful for performance
    testing netconf protocol  (vorburger, 17:12:26)
  * vorburger suggests asking if using the MD-SAL in-memory
    TestDataBroker could be an option here; rovarga says no, because
    that's still too heavy  (vorburger, 17:14:20)
  * jmorvay reason for wanting to remove it is maintenance burden,
    rovarga says not much maintenance is needed  (vorburger, 17:15:09)
  * jmorvay reason was JAXB removal, rovarga says it's OK to keep it in
    the test tool  (vorburger, 17:16:18)
  * LINK:
    https://github.com/opendaylight/netconf/tree/master/netconf/tools/netconf-testtool
    (vorburger, 17:18:49)

* YANG 1.1 action support in NETCONF  (vorburger, 17:20:43)
  * Martin Sandberg says Ericsson is looking for this, and has had some
    email exchange with jmorvay, there is a JIRA  (vorburger, 17:21:30)
  * LINK: jmorvay says there is some basic support but more work is
    needed, shouldn't be that hard for non-clustered, for clustering
    "also needs some proxy clustered services"   (vorburger, 17:22:23)
  * LINK: https://git.opendaylight.org/gerrit/#/c/74745/  (jmorvay,
    17:24:09)
  * LINK: https://lists.opendaylight.org/pipermail/netconf-dev/
    (vorburger, 17:26:52)

* Code Freeze in 2 days  (vorburger, 17:28:18)
  * skitt clarifies it's been moved from in 2 days to next week
    (vorburger, 17:28:36)
  * rovarga says distribution is currently broken, skitt merging netconf
    right now, and it should be OK after that  (vorburger, 17:29:08)
  * TSC call on Thursday needs to discuss this  (vorburger, 17:29:26)
  * rovarga says it should be OK tomorrow  (vorburger, 17:30:06)
  * jmorvay says his netconf patches could wait for Sodium if he has to,
    rovarga says it would be great to still get it into Neon
    (vorburger, 17:31:22)
  * vorburger asks what the timeline and plan for branching and then
    opening master as Sodium will be  (vorburger, 17:34:58)
  * rovarga says that there should be just 1 day of interruption, so
    we'll have stable/neon (for blockers only) and master open for
    Sodium the day after  (vorburger, 17:36:12)
  * small request: if someone using multi-patch job in a regular basis
    can review this patch:  (LuisGomez, 17:36:50)
  * LINK: : https://git.opendaylight.org/gerrit/#/c/79586/  (LuisGomez,
    17:36:54)
  * the TSC call on Thursday needs to make the decision for the code
    freeze cut off, community majority will probably want to post pone
    in  (vorburger, 17:36:54)

* Multi Patch Job  (vorburger, 17:38:19)
  * LuisGomez requesting skitt rovarga may be to review
    https://git.opendaylight.org/gerrit/#/c/79586/  (vorburger,
    17:38:28)

* Attendees  (vorburger, 17:42:38)
  * vorburger skitt LuisGomez TomP rovarga Martin Sandberg jmorvay
    (vorburger, 17:43:06)

* Gerrit API  (vorburger, 17:43:19)
  * skitt asks for interest in a "bot" which could auto-merge jobs
    (vorburger, 17:46:57)
  * LuisGomez says this could be a project for an intern (but still
    needs someone to manage it)  (vorburger, 17:47:13)
  * vorburger points to this opendaylight-bot, which never went this far
    (to actually "control" Gerrit via API), that was the dream
    (vorburger, 17:47:44)
  * vorburger says there are of course existing bot systems using in
    OpenStack and Kubernetes and elsewhere which perhaps could be
    re-used in OpenDaylight  (vorburger, 17:48:12)
  * https://zuul-ci.org ? LuisGomez says LF already said No to this a
    while ago  (vorburger, 17:50:07)

* Cookies  (vorburger, 17:52:44)



Meeting ended at 17:52:50 UTC.



People present (lines said)
---------------------------

* vorburger (39)
* rovarga (9)
* odl_meetbot (4)
* jmorvay (2)
* LuisGomez (2)
* skitt (1)



Generated by `MeetBot`_ 0.1.4