07:59:58 #startmeeting Functest weekly meeting 09 Oct. 2018 07:59:58 Meeting started Tue Oct 9 07:59:58 2018 UTC. The chair is ollivier. Information about MeetBot at http://wiki.debian.org/MeetBot. 07:59:58 Useful Commands: #action #agreed #help #info #idea #link #topic. 07:59:58 The meeting name has been set to 'functest_weekly_meeting_09_oct__2018' 08:00:01 Hello 08:00:06 #info Cédric ollivier 08:00:14 #info Juha Kosonen 08:00:30 #info Viktor Tikkanen 08:01:53 Delia is not connected. Let's wait few minutes. 08:03:23 #info Delia Popescu 08:03:25 Hi! 08:03:47 Hello! 08:04:25 Hello 08:04:46 #info Action point follow-up 08:04:51 #undo 08:04:51 Removing item from minutes: 08:04:57 #topic Action point follow-up 08:05:15 #info Intel NFV CI test-cases is ready for reviews 08:06:02 It seems quite good. I will check how the patches on OpenStack side are rated. 08:06:34 #link https://gerrit.opnfv.org/gerrit/#/c/61151/ 08:07:16 #link https://gerrit.opnfv.org/gerrit/gitweb?p=functest.git;a=blob;f=docker/smoke/Dockerfile;h=c35d0ca27967c0c84b34feedecc4612189dd4e4c;hb=0a009fc2abb3eeb9d7240853ae8d5643e3722d2f#l10 08:07:36 #info LFN team asks for modifying our tox.ini + doc trees 08:07:47 #link https://gerrit.opnfv.org/gerrit/#/c/62523/ 08:08:15 I think both LFN packages (theme and conf) should be improved before merge (conf.py is rated at -330/10) 08:08:50 And if merge, the overall jjb verification would raise timeout (30 min) 08:09:03 I just sent an email to LFN 08:09:45 #link https://build.opnfv.org/ci/job/functest-verify-unit-tests-and-docs-master/1778/console 08:10:52 #info requirements have been synchronized in most Features integrated in Functest 08:11:09 be free to review https://gerrit.opnfv.org/gerrit/#/c/63317/ 08:11:21 (FDS doesn't participate to Gambia) 08:13:49 #topic Functest 08:14:09 #info Fraser last Apex run was successful 08:14:32 #link https://build.opnfv.org/ci/view/functest/job/functest-apex-baremetal-daily-fraser/294/console 08:14:38 the branch seems over 08:15:07 #info Gambia 08:15:27 Fuel (both architectures) starts running Functest gambia containers (unsuccessfully) 08:16:06 Sorry the last runs are much better 08:16:35 #link https://build.opnfv.org/ci/view/functest/job/functest-fuel-armband-baremetal-daily-gambia/lastFailedBuild/console 08:16:42 At least for Healthcheck 08:17:23 delia_: it seems related to fuel arm64, no? 08:17:55 ollivier: whell, it depends on the error 08:18:22 ollivier: on what you sent yes it is 08:18:26 I will look into it 08:19:17 delia_: thank you. 08:19:20 #link https://gerrit.opnfv.org/gerrit/#/c/63319/ 08:19:42 #info new IMG_PROPERTIES env var to reduce config patches 08:19:50 delia_: Thank you for that patches 08:20:00 #link https://gerrit.opnfv.org/gerrit/#/c/61913/ 08:20:03 ollivier: any time 08:20:30 ollivier; patch to fix sent :( 08:20:41 We may apply the same rules for Flavor extra specs? 08:21:06 delia_: the patch is about releng. Functest is working well ;) 08:21:24 ollivier: I say we can 08:21:56 Can I action you about that ? 08:22:14 sure 08:22:37 #action Delia ease configuring extra specs via env vars 08:22:38 Thank you 08:23:18 #info Rally: floating ip tests have been integrated 08:23:25 juhak: Thank you 08:23:33 It helped us to highligh issues in Orange Openlab 08:23:56 that's great 08:24:17 I should be noted that the rally test doesn't wait the floating ip to be active (few sec in normal cases). 08:24:35 #link https://gerrit.opnfv.org/gerrit/#/c/63099/ 08:24:45 do we backport it till gambia? 08:25:09 I would say yes. 08:25:15 I think we should 08:26:08 agree 08:26:09 juhak: would you mind backporting the patch? 08:26:33 already on progress 08:26:39 thank you 08:27:43 I would like to share with you one rally exception 08:27:59 juhak: maybe you're already aware of it. 08:28:12 If shared networks already exist, rally creates a new shared network 08:28:39 It may change a little bit the bahavior (ie. vm booting with network or not in case of Cinder tests) 08:29:32 Then tempest.conf in res dir is not a full copy of the executed one (fixed_network_name is modified during rally execution) 08:30:02 I think the network creation depends on test case 08:30:11 what case you are referring? 08:30:30 few sec 08:31:54 tempest.api.volume.test_volumes_actions.VolumesActionsTest.test_attach_detach_volume_to_instance[compute,id-fff42874-7db5-4487-a8e1-ddda5fb5288d,smoke] 08:32:36 If one shared network already exists, then fixed_network_name is set and the vm is connected to the new network. 08:32:57 ah, tempest cases executed by Rally 08:32:59 If not, no network is detected and the vm is not connected to any network 08:33:13 It works here in both cases 08:33:35 It may raise side effects vs Contrail what I'm checking 08:33:59 juhak: yes. 08:35:02 ok, your analysis make sense 08:35:06 Nothing wrong. But I would like to share this very specific case where the tempest.conf saved is not the right one. It may happen when testing OpenStack in production. 08:35:57 thanks, this is good to know 08:36:09 #info rally_full sometimes fails when testing Cinder in Openlab (in progress) 08:36:39 Rally seems fine. I'm writing a new task to raise the bugs. 08:38:28 Nothing else from my side. I cleant several JIRA tickets for the next release. Only docs have to be updated (+ few backports) 08:38:50 #topic AoB 08:39:48 I forgot the new bug opened by CristinaPauna 08:39:52 #link https://jira.opnfv.org/browse/FUNCTEST-1010 08:40:14 I will add the missing condition in jjbs. 08:40:18 Anything else? 08:40:29 I'll be "out of office" this and next week 08:41:43 Thank you for sharing that. It's not related to OpenSource summit 08:42:43 Then let's close the meeting. Thank you. Have a nice week! 08:42:48 #endmeeting