15:00:48 #startmeeting monasca 15:00:49 Meeting started Wed Apr 11 15:00:48 2018 UTC and is due to finish in 60 minutes. The chair is witek. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:50 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:52 The meeting name has been set to 'monasca' 15:01:06 Hello everyone! 15:01:09 Hello 15:01:09 Hello 15:01:11 hello 15:01:18 hello 15:01:38 agenda: https://etherpad.openstack.org/p/monasca-team-meeting-agenda 15:01:55 Hello 15:02:04 #topic Vancouver Forum 15:02:38 hey all 15:02:49 sc: your stage 15:02:51 a disaster 15:02:59 hello 15:03:02 :( 15:03:16 OpenStack Foundation did not approved my travel request 15:03:38 I tried to get back to them but they replied there is no way to change their mind 15:04:06 I submitted a travel approval iternally but HPE announced we are not present at the Summit so... 15:04:15 I see 15:04:30 you proposed the topic for the Forum discussion 15:04:40 http://forumtopics.openstack.org/cfp/details/92 15:05:00 my manager is supporting me to convince his manager and ... I have 7 levels of approval :( 15:05:10 7. 15:05:11 Levels. 15:05:20 Wow, that's bad :-( 15:05:36 I hope this is going to stop at the first 2 15:06:14 I'll update you next meeting 15:06:39 anyone could take part to the forum? who is sure to go to the Vancouver? 15:06:57 mine is approved 15:07:10 great 15:07:16 grand 15:08:11 anyway, everyone is free to amend the proposal or ask me to amend, you know how to get in touch (sc@linux.it is my main email) 15:08:49 the deadline for submissions is Apr 15 15:10:20 witek: I'm not giving up 15:10:32 I'd be happy to help with the session, but don't want to do it alone 15:11:15 I'll be at the summit as well, so I can probably help, too. 15:11:33 cool, thanks 15:12:27 if you have any feedback to the proposal, please contact sc 15:12:32 I've got a lightning talk I should hold and don't know the time off the top of my head but I'm working on the assumption that there's no scheduling conflict for now :-) 15:13:06 for now not, the forum sessions are not accepted yet :) 15:13:28 so no schedule for them 15:13:28 thanks jgrassler 15:13:37 let's move on 15:13:47 #topic reviews 15:13:56 https://review.openstack.org/#/c/547379/ 15:14:14 I added this point 15:14:24 I did too 15:14:42 has anyone tested it in devstack? 15:14:51 this is old and waiting to be merged, it works 15:15:08 I can rebase and test it again 15:15:31 so, you've been installing in virtualenv? 15:16:40 the change itself looks good, but I haven't tested 15:16:44 I tested once on Helion so in virtualenv 15:17:08 ok, thanks 15:17:30 but then the environment got distroyed so I can test in again in the next few days 15:18:05 I have just given W+1 15:18:20 K 15:18:49 no need to test it twice 15:19:17 https://review.openstack.org/554304 15:20:11 I've seen devstack doesn't install monasca-agent with this change 15:20:14 I see the error 15:20:43 the problem is that if we merge, we won't be able to test other components with tempests 15:21:04 OK 15:22:18 good, that you're making progress on this 15:22:18 I don't understand where the - cames from but I'll sort out 15:23:06 any other comments on this change? 15:23:28 expect a new patch tonight (code is at home ;-)) 15:23:54 :) great, thanks for your contribution 15:24:21 #topic SNMP notification plugin 15:24:41 Thanks for your time. 15:24:44 haruki: it's yours 15:24:53 As sc discribed in here, https://blueprints.launchpad.net/monasca/+spec/snmp-notification 15:25:00 I assigned to me, I have the code but I missing the test 15:25:14 Many enterprises are deploying OpenStack into their IT environment along side other tools and need to integrate OpenStack management with `their own system management tools`. 15:25:53 I mean that I developed the plugin, it's running at a customer site but I'm unable to reverse engineer good unitests 15:26:58 oh, 15:26:59 sc: cool, could you publish the code to Gerrit? 15:27:01 cool 15:27:10 is it possible to see the code? 15:27:20 I aslso think so 15:27:20 mayeb we can help on unittests 15:27:24 witek: yes, expect a WIP commit ASAP 15:27:28 yes 15:27:47 I was ashamed to admit I need help 15:27:52 * sc red in face 15:28:01 sc: do you have any additional for this plugin? 15:28:05 perfect, that's very helpful for us 15:28:09 additional requirements 15:28:29 yes 15:28:56 depending on pysnmp and ASN.1 python modules needs to be compiled 15:29:14 pysnmp is in g-r 15:30:01 pyasn1==0.1.9 15:30:03 pysmi==0.0.7 15:30:05 pysnmp==4.3.2 15:30:07 pysnmp-mibs==0.1.6 15:30:09 yesYes 15:30:11 yes 15:30:33 means there are some requirements? 15:30:51 this are the main requirments I added (blame the fine IRC clinet fo the last two lines) 15:31:42 in term of python packages the plugin depends on, yes 15:32:11 sc: thx 15:32:26 then if you need to run integration tests that another story, a complex one 15:32:36 I did integration tests 15:33:10 if you have any feedback to the story: https://storyboard.openstack.org/#!/story/2001822 15:33:15 very welcome 15:33:17 For the integration tests, you might start by documenting the details of how you tested 15:34:40 sc: thanks for your input, I think it's a good starting point 15:34:47 I'm so grad if sc add comment in the story: https://storyboard.openstack.org/#!/story/2001822 15:35:11 haruki: I'll do ASAP, I just had time to assign to me 15:35:42 haruki: thanks for the interest in this functionality, I was about to abandon the idea 15:36:23 your welcome:) 15:37:04 #topic InfluxDB 1.5.1 15:37:53 nseyvet: you've proposed this one, right? 15:38:56 yes 15:39:10 I noticed an upgrade to influx 1.3.8 today. 15:39:29 And Influx recently release 1.5.1. 15:39:49 From release notes and documentation, https://docs.influxdata.com/influxdb/v1.5/administration/upgrading/, "Starting with the InfluxDB 1.5 release, enabling Time Series Index (TSI) is recommended for all customers." 15:40:05 High level description: https://www.influxdata.com/blog/path-1-billion-time-series-influxdb-high-cardinality-indexing-ready-testing/ 15:40:26 Basically, this index is available since 1.3.x. But it is now production ready. 15:40:56 have you done or planning any testing with this? 15:40:59 We have been internally running with 1.4.3 15:41:25 No plan or test with this so far. 15:41:51 but we want into the limits of TS in the default influxdb indexing engine (in-memory) 15:42:01 so, we probably will try. 15:42:24 from a Monasca perspective, what is preventing an upgrade to latest 1.4.x? 15:43:35 we haven't considered it yet 15:44:03 we've had bad experience with using new InfluxDB versions previously 15:44:17 but it might have changed 15:44:20 1.4.3 is old 15:45:04 if you've tested and using it, I think we could upgrade 15:45:21 Up to you. We are running 1.4.3. 15:46:28 feel free to propose the change 15:47:21 I mentionned it in https://review.openstack.org/#/c/558944/ 15:47:22 btw. I think InfluxDB logs are not configured correctly and not displayed for zuul jobs 15:47:33 in devstack 15:47:39 thought that it was sufficient as a comment. 15:47:45 but that could be the subject for another change 15:48:39 what do ppl think of this tsi1 index? 15:49:21 seems to be the solution for your limitations 15:50:01 so, we are the first to notice this problem? 15:50:36 we haven't hit the limits 15:51:52 what settings do you use by default? Because the default of 1 million is quite small. 15:52:48 I think we use default, but would have to check 15:53:35 any other voices? 15:55:07 nseyvet: thanks for the reference about TSI 15:55:32 witek: np 15:56:15 please propose the upgrade change 15:56:48 if you would like to work on improving InfluxDB logging in devstack, that would be great as well 15:57:16 #topic Monasca Board 15:57:26 https://storyboard.openstack.org/#!/board/60 15:57:36 I have made some small status updates 15:57:42 and added few tasks 15:58:07 Chuck's change adding Python 3 support to monasca-ui has been merged 15:58:58 Adrian's and Ashwin's changes added to Doing 15:59:44 please remember about adding Task references to your changes, so that we can better track the progress 16:00:00 I have to finish now 16:00:05 thank you all 16:00:11 and see you next week 16:00:13 bye 16:00:14 bye, bye 16:00:19 Bye 16:00:23 thanks, bye 16:00:24 thanks! 16:00:30 thanks a lot, see you 16:00:41 #endmeeting