08:01:58 #startmeeting Functest weekly meeting 20 Feb. 2018 08:01:58 Meeting started Tue Feb 20 08:01:58 2018 UTC. The chair is ollivier. Information about MeetBot at http://wiki.debian.org/MeetBot. 08:01:58 Useful Commands: #action #agreed #help #info #idea #link #topic. 08:01:58 The meeting name has been set to 'functest_weekly_meeting_20_feb__2018' 08:02:04 #info Cédric Ollivier 08:02:19 #info Juha Kosonen 08:02:28 #info Jose Lausuch 08:03:02 juhak: jose_lausuch: hi 08:03:11 hi 08:03:48 Hi 08:03:57 let's start? maybe other will enter. Linda and Serena should be on holidays. 08:03:58 #info Delia Popescu 08:04:13 depo: hi 08:04:19 ok 08:04:22 hi 08:04:31 #topic action point follow-up 08:04:47 #info Fix the config of fixed_network_name in tempest" was merged 08:05:01 #info rebaca team updated juju_epc to leverage on snaps instead of openstack_utils 08:05:39 #info many JIRA tickets about code improvement were closed 08:05:45 #link https://jira.opnfv.org/browse/FUNCTEST-931 08:05:52 #link https://jira.opnfv.org/browse/FUNCTEST-932 08:05:59 #link https://jira.opnfv.org/browse/FUNCTEST-933 08:06:20 #info mainly refstack_defcore(under investigation) and rally_sanity(under investigation) have to fixed 08:06:36 #info the second Kubernetes testcase is still under review https://gerrit.opnfv.org/gerrit/51635 08:07:03 be free to review it. It seems fine. 08:07:13 #info we are still waiting for a full synchonization between functest and functest-xtesting (we can't push all changes due to rules vs authors) 08:07:27 #info a new proposal rewriting our Jenkins jobs is ready for reviews https://gerrit.opnfv.org/gerrit/#/c/51853/ 08:07:48 depo: I think you updated it to tkae Fatih's comments into account, no? 08:08:37 It's the best time to merge simply because several installers are stopped (Daisy, Compass...) 08:09:12 I will do my best to review it soon. I would like to first finish vnfs and refstack. 08:09:56 yes, I updated after Fatih's comments 08:10:06 thank you 08:10:47 #topic Code improvement 08:10:59 #info last CONST.__getattribute__() are being removed 08:11:13 Could you please review https://gerrit.opnfv.org/gerrit/#/c/52327/ ? 08:11:27 all env vars are now read via env.get() and set directly via os.environ instead of CONST 08:11:54 we are updating SFC as well to remove the backward compatibility 08:12:13 https://gerrit.opnfv.org/gerrit/#/c/52243/ 08:12:31 I'll review it 08:12:52 but looks good with getattr(CONST, 08:13:44 Thx. It's trivial. There is currently a misunderstanding between logging rules and string operations between python2 and python3. 08:14:36 I think we could apply the rules for config.py and offer a new method for third-parties. 08:15:05 Mainly SFC and SDNVPN are leveraged on Const and Functest utils for that. 08:15:58 I would prefer a clear API than reusing internal methods as Python badly allows. 08:16:54 #info our code has been rated at 9.06/10 (most errors are located in deprecated modules) 08:17:02 That's very good 08:17:27 only orchestra_openims.py and orchestra_clearwaterims.py still leverages on openstack_utils instead of snaps. 08:18:27 nothing clear about possible actions on this modules by the original authors. I'm afraid that we have to update them (or remove them) 08:19:27 another git repository could have been considered too. 08:20:57 #topic Functest 08:21:10 #info only EXTERNAL_NETWORK is now required 08:21:30 The wiki page has been updated as well (https://wiki.opnfv.org/pages/viewpage.action?pageId=13211751) 08:21:50 the env var you mean 08:22:39 yes. I could have precised it. Thank you 08:22:41 https://git.opnfv.org/functest/tree/functest/utils/env.py 08:23:27 I se 08:23:28 I see 08:23:46 is that also included in the jjobs? 08:23:59 adding EXTERNAL_NETWORK to the docker env vars? 08:24:21 The jjobs have been updated to override all values anyway. 08:24:35 It's more or less designed for end users. 08:24:53 ok 08:25:05 We are still waiting for XCI community. 08:25:15 https://gerrit.opnfv.org/gerrit/51947 08:25:25 https://gerrit.opnfv.org/gerrit/51985 08:25:48 (the gate seems down) 08:26:29 We could also detect EXTERNAL_NETWORK but for the time being it's very good. 08:27:15 #info we now publish a report for tempest testcases 08:27:25 #link https://jira.opnfv.org/browse/FUNCTEST-934 08:27:55 thx 08:28:17 #info a new testcase integrating patrole should be published soon 08:29:02 We may refactor a little bit TempestCommon as run() should be fully overidden for a simple post config operation. 08:31:07 juhak: Do you agree? 08:31:31 there is always room for refactoring :) 08:32:24 great. There are at least two new tempest plugins: Patrole, neutron (Trunk) 08:33:12 #info a new patchset is being developed on refstack 08:33:24 Mainly to automatically download defcore.txt, to clean instance variables, to stop creating file in /usr/lib... 08:33:53 We could have considered using rally to run refstack's tests list instead. 08:34:42 refstack is failing but tempest works. 08:35:22 maybe several tests are part of tempest_full. 08:35:57 #info cloudify_ims, vyos_vrouter and juju_epc seems working 08:36:32 #link https://jira.opnfv.org/browse/FUNCTEST-935 08:36:41 #link https://jira.opnfv.org/browse/FUNCTEST-936 08:36:51 #links https://jira.opnfv.org/browse/FUNCTEST-937 08:36:57 #undo 08:36:57 Removing item from minutes: 08:37:00 #link https://jira.opnfv.org/browse/FUNCTEST-937 08:37:39 But they should work vs OPNFV installer as all of them are defining the default domain names. 08:38:55 Let's see the next CI runs. 08:39:14 #topic AoB 08:39:41 #info two presentations of Orange's ONAP integration chain (leveraging on functest-core/Xtesting) has been done by Morgan 08:41:16 We are mainly running functest containers to validate the infrastructure and integrating the ONAP's robot files into a new container based on functest-core. 08:41:48 https://git.opnfv.org/functest/tree/functest/core/robotframework.py 08:41:58 It works very well :) 08:42:58 Let me know if you need more details. I know that Jose was part of the second meeting too. 08:43:30 Another topic you would like to address? Else we can close the meeting a litllte bit earlier :) 08:43:46 links to presentations available? 08:44:33 Sure. I will broadcast them after the meeting. 08:44:47 thanks 08:47:13 juhak: have you spent time on the possible issue on live migration? I know you have been working (at least) on tempest (+ pylint runs on rally) 08:48:09 not yet... I'll try to check that this week 08:48:55 thank you. 08:50:03 Then we can close the meeting a little bit earlier. 08:50:07 Thank you 08:50:13 #endmeeting