16:02:09 <xarses> #startmeeting fuel
16:02:10 <openstack> Meeting started Thu Feb  4 16:02:09 2016 UTC and is due to finish in 60 minutes.  The chair is xarses. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:02:11 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:02:13 <openstack> The meeting name has been set to 'fuel'
16:02:13 <kozhukalov> hi
16:02:42 <xarses> #chair xarses
16:02:42 <xarses> Todays Agenda:
16:02:42 <openstack> Current chairs: xarses
16:02:42 <xarses> #link https://etherpad.openstack.org/p/fuel-weekly-meeting-agenda
16:02:42 <xarses> Who's here?
16:02:46 <maximov> hi
16:02:48 <dnikishov> hi
16:02:57 <vsakharov> hi
16:03:03 <ikalnitsky> o/
16:03:08 <mwhahaha> o/
16:03:12 <dklenov> \o
16:03:21 <akislitsky_> hi
16:03:38 <SheenaG> o/
16:03:41 <xarses> #topic Action items from last meeting
16:04:02 <xarses> ogelbukh will start moving configdb code over to openstack
16:04:59 <xarses> aglarendil raise rollback support in solar to the ML for further discussion
16:05:37 <xarses> moving on
16:06:07 <xarses> #topic Fuel 8.0 HCF Status (maximov, dpyzhov)
16:06:18 <maximov> Hi Folks
16:06:22 <monester> hi
16:06:36 <maximov> We had 7 High and 1 Crit bugs on fuel, we collected the current status of the bugs #link https://etherpad.openstack.org/p/fuel-bugs-triage-8-0
16:06:44 <maximov> highlights:
16:06:55 <maximov> I downgraded  https://bugs.launchpad.net/fuel/+bug/1541390 to High because it doesn’t affect deployment and according to triage rules
16:06:56 <openstack> Launchpad bug 1541390 in Fuel for OpenStack mitaka "bvt's diagnostic snapshot doesn't contain logs" [High,Triaged] - Assigned to Aleksandr Didenko (adidenko)
16:07:15 <sbog> hi
16:07:19 <maximov> QA agreed with this.
16:08:08 <maximov> https://bugs.launchpad.net/fuel/+bug/1536198 and  https://bugs.launchpad.net/fuel/+bug/1539576  were removed from 8.0
16:08:10 <openstack> Launchpad bug 1536198 in Fuel for OpenStack mitaka "After backup-reinstall-restore Fuel uses bootstrap with wrong ssh keys" [High,In progress] - Assigned to Dmitry Bilunov (dbilunov)
16:08:11 <openstack> Launchpad bug 1539576 in Fuel for OpenStack mitaka "fuel doesn't create diagnostic snapshot" [High,Incomplete] - Assigned to Fuel Library Team (fuel-library)
16:08:52 <maximov> https://bugs.launchpad.net/fuel/+bug/1541256  and https://bugs.launchpad.net/fuel/+bug/1541390 should be merged today (before HCF)
16:08:54 <openstack> Launchpad bug 1541256 in Fuel for OpenStack mitaka "[Ironic] Deployment failed: Missing a required parameter uids" [High,In progress] - Assigned to Maciej Kwiek (maciej-iai)
16:08:55 <openstack> Launchpad bug 1541390 in Fuel for OpenStack mitaka "bvt's diagnostic snapshot doesn't contain logs" [High,Triaged] - Assigned to Aleksandr Didenko (adidenko)
16:09:00 <maximov> questions ?
16:09:58 <xarses> thanks, non for me
16:10:05 <angdraug> are we officieally in HCF yet?
16:10:13 <maximov> nope.
16:10:19 <maximov> it should be declared today
16:10:23 <angdraug> ok
16:10:26 <angdraug> thanks
16:10:48 <maximov> moving on ?
16:10:48 <holser_> o/
16:10:56 <xarses> #topic CI team status (bookwar)
16:11:16 <bookwar_> hi all, several points i d note here
16:11:21 <bookwar_> 0) we have two active issues on ci.fuel-infra.org: a) ikutukov and romcheg are investigating the issue with pip and SSL https://bugs.launchpad.net/fuel/+bug/1541432 b) community iso builds were failing due to https://bugs.launchpad.net/fuel/+bug/1541899
16:11:23 <openstack> Launchpad bug 1541432 in Fuel for OpenStack "verify-python-fuelclient job failing" [Critical,Confirmed] - Assigned to Fuel Python Team (fuel-python)
16:11:24 <openstack> Launchpad bug 1541899 in Fuel for OpenStack "Problem with iso build on build1" [Undecided,New] - Assigned to Fuel build team (fuel-build)
16:11:41 <bookwar_> iso builds are more or less fixed we are waiting for new run
16:11:58 <angdraug> status Undecided?
16:12:21 <bookwar_> yes, that's new bug, we fixed the issue by server rebbot, bug is going to be triaged
16:12:56 <bookwar_> if there no questions about this: here are features currently in development:
16:13:02 <angdraug> what's the status of fuel ci for puppet-openstack?
16:13:06 <angdraug> ah )
16:13:08 <bookwar_> 1) External snapshots, export/import of system test environments
16:13:16 <bookwar_> #link https://wiki.openstack.org/wiki/Fuel/External_Snapshots
16:13:26 <bookwar_> this is an awesome wiki page everyone should read
16:13:38 <bookwar_> basic functionality is available, we are working on jenkins wrapper for this
16:13:43 <bookwar_> beta-testers are welcome
16:13:59 <bookwar_> 2) Dockerized test environments spec https://review.openstack.org/#/c/270799
16:14:07 <bookwar_> we have proof of concept, and docker files for fuel-web tests are available
16:14:13 <bookwar_> we are getting it integrated with internal docker registry
16:14:23 <bookwar_> 3) puppet-openstackhttps://blueprints.launchpad.net/fuel/+spec/deployment-tests-for-puppet-openstack
16:14:34 <bookwar_> noop job with master of puppet-openstack modules is available https://ci.fuel-infra.org/job/puppet-openstack-noop/
16:14:51 <bookwar_> it runs on a daily basis
16:14:58 <bookwar_> We need someone from puppet folks to start fixing existing fuel-library issues, there's no point in running more tests with the current state
16:15:19 <bookwar_> 4) we cover repositories other than fuel-library by deployment tests
16:15:25 <bookwar_> fuel-agent, python-fuelclient are covered
16:15:31 <bookwar_> https://review.fuel-infra.org/#/c/16102/ fuel-web is going to be enabled after HCF
16:15:35 <bookwar_> fuel-ostf and fuel-astute are in progress
16:15:50 <bookwar_> so that's it for now, questions? :)
16:16:07 <angdraug> maximov: can you help with (3)?
16:16:44 <maximov> i will read this bp after meeting
16:17:00 <maximov> we need to allocate resources
16:17:04 <maximov> for this I think
16:17:17 <angdraug> we need a puppet developer to get fuel-library noop tests to pass with puppet-openstack master branch
16:17:30 <angdraug> so ideally someone good at troubleshooting :)
16:17:55 <maximov> do we know how long would it take to fix all noops tests?
16:18:12 <maximov> does anybody try to estimate this work?
16:18:50 <angdraug> I don't think we can without having a puppet developer look at the current state of the results
16:19:24 <xarses> can we assign some to get this estimate?
16:19:30 <angdraug> given that it's passing with the versions of upstream that we currently have in Puppetfile shouldn't be anything major
16:20:04 <maximov> well, we will try to find resource but we already did allocation for 9.0
16:20:08 <angdraug> initial "get to green" part can be a few days or a couple of weeks at worst
16:20:20 <angdraug> but then we'll need to set up a duty to keep it green
16:20:57 <maximov> angdraug: ok, understood. yes we will add this to duty after we make them green
16:21:06 <dklenov> I agree with maximov that it is to be aligned with 9.0 resource allocation
16:21:11 <angdraug> whoever you allocated for Mitaka support is going to use this jenkins job
16:21:22 <dklenov> couple of weeks mean half of 9.0 dev time :)
16:21:31 <angdraug> I'm aware )
16:21:44 <angdraug> half of _remaining_ 9.0 dev time
16:21:45 <dklenov> it would definitely be easier to take this after 9.0 FF
16:21:54 <angdraug> and the only blocker feature in 9.0 is mitaka support
16:22:08 <angdraug> we can drop anything else, we can't drop that
16:22:09 <dklenov> nope
16:22:17 <dklenov> we have a number of them
16:22:24 <angdraug> SheenaG: ^
16:22:27 <dklenov> e.g. NFV items
16:22:38 <angdraug> the deal was, no other blockers in 9.0 except Mitaka
16:23:03 <angdraug> looks like we fell into priority inflation trap again
16:23:08 <maximov> yes, but mitaka support is not only fuel resources, we also have Ivan B. team
16:23:22 <maximov> they should work on mitaka support full time
16:23:41 <angdraug> yes, unfortunately Ivan is not in the meeting, please check with him
16:23:47 <maximov> angdraug: ok
16:23:50 <angdraug> it would make sense for his team to pick this up
16:24:06 <angdraug> degorenko: around? can you ping Ivan about this?
16:24:15 <xarses> we need to move along, can we create an action on someone
16:24:16 <SheenaG> What's up angdraug?  Sorry I'm multitasking in another meeting
16:24:43 <angdraug> please action maximov to coordinate with other teams and find a volunteer for puppet-openstack noop job
16:24:49 <degorenko> angdraug, hey, yes, i'm here
16:24:58 <maximov> ok
16:25:28 <angdraug> degorenko: please coordinate with maximov on ^
16:25:28 <xarses> #topic Build team status (kozhukalov)
16:25:44 <kozhukalov> 1) Docker removal:
16:25:44 <kozhukalov> Spec has been merged https://review.openstack.org/#/c/248814/
16:25:44 <kozhukalov> Implementation patches are on review (announced in mos-dev) but still not final patch set, still testing them
16:25:44 <kozhukalov> Master node deployment works, but bvt fails later on cluster deployment
16:25:44 <kozhukalov> Preliminary agreement with Fuel team is to merge this feature on Monday 02/08/2016 (will announce)
16:25:45 <kozhukalov> 2) Centos bootstrap:
16:25:45 <kozhukalov> Announced in openstack-dev
16:25:46 <kozhukalov> Spec has been merged https://review.openstack.org/#/c/273159/
16:25:46 <kozhukalov> S. Kulanov is working on implementation, gonna be merged till the end of iteration 7
16:25:47 <kozhukalov> 3) Provisioning/deployment separation
16:25:47 <kozhukalov> Spec is on review https://review.openstack.org/#/c/254270/
16:25:48 <kozhukalov> Got feedback from python, library and upgrade teams
16:25:52 <xarses> #action maximov to coordinate with other teams and find a volunteer for puppet-openstack noop job
16:26:04 <kozhukalov> that is it. questions?
16:27:33 <ogelbukh> do we test centos and ubuntu bootstrap images compat?
16:28:19 <ogelbukh> like I reboot node provisioned in centos into ubuntu bootstrap
16:28:24 <kozhukalov> i am not sure what you mean
16:28:50 <ogelbukh> will metadata change or smthn?
16:28:55 <kozhukalov> ubuntu bootstrap can not be compatible with centos one
16:29:01 <dklenov> ogelbukh: you can do it in 8.0
16:29:06 <kozhukalov> it is only compatible with software we use
16:29:14 <ogelbukh> ifaces numbering/order
16:29:20 <ogelbukh> hw detection
16:29:27 <ogelbukh> that kind of stuff
16:29:40 <kozhukalov> ubuntu bootstrap runs newer kernel and has better hardware support
16:29:59 <dklenov> plus you can include any custom driver into it
16:30:10 <ogelbukh> so I can get some hw I didn't have before?
16:30:19 <kozhukalov> both ubuntu and centos support persistent interface naming
16:30:24 <dklenov> yes
16:30:35 <ogelbukh> k
16:31:15 <angdraug> kozhukalov: we need to improve our replacement for version.yaml
16:31:18 <kozhukalov> any other q?
16:31:38 <kozhukalov> what do you mean?
16:31:45 <angdraug> the current package changelogs summary is not 100% reliable (some packages don't have sha1)
16:31:52 <angdraug> and it has to be machine readable
16:32:05 <kozhukalov> it is a matter of package changelog
16:32:08 <xarses> angdraug: +1
16:32:15 <angdraug> that's the problem
16:32:26 <angdraug> unlike deb, rpm changelog is freeform
16:32:40 <angdraug> and even deb format doesn't guarantee presense of sha1 in the changelog
16:32:44 <kozhukalov> we have merged to patches in fuelclient and network-checker that "fix" this changelog related "issue"
16:33:05 <xarses> but we need this information for reproducing environments, if its missing we likely cant
16:33:17 <angdraug> it has to be bullet proof, a file that you can pass as-is as a jenkins job parameter that allows one to re-create exact same Fuel master node
16:33:26 <kozhukalov> once again, stop identifying the content of the master node by sha sums
16:33:32 <kozhukalov> use package versions
16:33:50 <xarses> kozhukalov: you propose that all package versions live forever then?
16:34:07 <kozhukalov> we don't need to re-create exact copy of the iso
16:34:11 <angdraug> if that's the case, we have to disable 30-day time limit on perestroika snapshots
16:34:24 <kozhukalov> you can not control on the build stage the content of the master node
16:34:29 <angdraug> kozhukalov: not of the iso, of the env where a bug can be reproduced
16:34:35 <xarses> and that somehow know how which version is present in a given env
16:35:23 <SergK> changelog for fuel packages now in form: http://paste.openstack.org/show/485984/
16:35:26 <angdraug> lets move this to ML, xarses: can you start a thread on openstack dev?
16:35:36 <kozhukalov> seems this place is not the best for this discussion
16:35:39 <xarses> angdraug: sure
16:36:22 <xarses> #action xarses will bring up viability of version information (was version.yaml) for environment triage
16:36:26 <xarses> #topic Agenda for design summit in Austin (angdraug) https://etherpad.openstack.org/p/fuel-austin-agenda
16:36:34 <kozhukalov> if changelog does not contain any info about sha sums, it is not a bug
16:36:41 <angdraug> folks, lets start brainstorming ideas for design sessions and workshops that we want to schedule for Austin
16:36:41 <angdraug> #link https://etherpad.openstack.org/p/fuel-austin-agenda
16:36:41 <angdraug> if you have a feature planned for N/10.0 that is not yet fully defined and designed and could benefit from public discussion, please propose it in the etherpad
16:36:41 <angdraug> at the very least, I think we should abundantly cover LCM use cases
16:36:41 <angdraug> I've added it as a first proposal, we need a volunteer to lead this topic
16:36:43 <angdraug> </report>
16:37:26 <angdraug> questions?
16:37:35 <angdraug> if not, please post ideas in the etherpad
16:38:10 <xarses> ++
16:38:15 <dklenov> are we going to cover solar-related items?
16:38:20 <angdraug> yes
16:38:22 <xarses> we should
16:38:40 <ogelbukh> we need couple of specs then
16:39:00 <ogelbukh> didn't find any bps on LP
16:39:03 <angdraug> more than a couple I suspect
16:39:12 <ogelbukh> ;)
16:39:15 <angdraug> that's what the design summit is for )
16:39:28 <xarses> and something about upgrades =)
16:39:40 <ogelbukh> absolutely
16:40:07 <xarses> moving on then
16:40:07 <ogelbukh> the whole api for that
16:40:21 <xarses> #topic Telco Team Status (vsakharov)
16:40:34 <vsakharov> Hello folks
16:40:41 <vsakharov> This is update from Telco team:
16:40:41 <vsakharov> - majority of engineers worked on the main activity - fixing bugs for 8.0 release.
16:40:41 <vsakharov> - started the NFV design of 9.0 features.
16:41:42 <vsakharov> Any questions?
16:41:54 <xarses> how are we doing with the NFV features?
16:42:06 <xarses> we don't have alot of time left for 9 development
16:42:09 <dklenov> design specs are on review
16:42:25 <dklenov> implementation is started for some items
16:42:34 <vsakharov> Also we start implementation of base items of each feature.
16:42:53 <dklenov> xarses: yes, that's true
16:43:14 <dklenov> implementation will start full-speed with declaring HCF
16:43:31 <xarses> thanks
16:43:31 <dklenov> as part of engineers were involved into bug fixing
16:43:52 <xarses> #topic Mixed Team Status (pzhurba)
16:43:59 <pzhurba> Hi
16:44:10 <pzhurba> Mixed team was working on bugfixing and reviews.
16:44:19 <pzhurba> All bugs connected to HCF 8.0 was merged
16:44:30 <pzhurba> Finished work on https://blueprints.launchpad.net/fuel/+spec/openstack-config-change
16:44:40 <pzhurba> that is all
16:44:57 <pzhurba> question?
16:45:20 <angdraug> plans for 9.0?
16:45:46 <pzhurba> Now not clear
16:45:53 <pzhurba> multipass probaly
16:46:04 <pzhurba> multipass support probaly
16:46:20 <xarses> multipath?
16:46:38 <dklenov> yes
16:46:40 <pzhurba> yes
16:46:47 <pzhurba> :)
16:46:51 <angdraug> ok, thanks )
16:46:55 <dklenov> plus some work on manifests idempotency AFAIK
16:47:13 <pzhurba> yes
16:47:33 <xarses> #topic Fuel nonroot feature status (dnikishov)
16:47:38 <dnikishov> Hello
16:47:42 <dnikishov> Current implementation status: got nailgun/fuel-agent parts working, now moved on to library: refactoring openstack::auth_file and investigating ceph puppet module
16:47:52 <dnikishov> I have also updated the spec: https://review.openstack.org/#/c/243340/; still need more reviews.
16:48:03 <dnikishov> Would also appreciate Infra and QA folks reviewing it as new update contains details on CI and QA changes.
16:48:18 <dnikishov> any questions?
16:48:21 <xarses> dnikishov: what about #link https://blueprints.launchpad.net/fuel/+spec/fuel-nonroot-master-node
16:48:36 <xarses> it has no spec still
16:49:05 <dnikishov> I'm not sure I'll have enough time for master node as well
16:49:19 <dnikishov> so this is going to 10
16:49:44 <dnikishov> and IMO it's too early for the spec for this
16:50:34 <xarses> ok
16:51:01 <xarses> #topic UI Team status (vkramskikh)
16:51:05 <vkramskikh> Hi! This week we've fixed a few High bugs for 8.0, and currently there are no UI bugs for 8.0.
16:51:06 <vkramskikh> As for 9.0 features:
16:51:06 <vkramskikh> 1) https://blueprints.launchpad.net/fuel/+spec/converge-to-eslint-config-openstack - implemented
16:51:06 <vkramskikh> 2) https://blueprints.launchpad.net/fuel/+spec/remove-vendor-code - removal started: https://review.openstack.org/#/c/276279/ ; as for downstream repo, there is still no CI and we have some access issues, but they will be probably resolved by the time we merge code removal and sync the repos
16:51:08 <vkramskikh> 3) https://blueprints.launchpad.net/fuel/+spec/redesign-of-node-roles-panel - good progress, most likely will land next week
16:51:11 <vkramskikh> 4) https://blueprints.launchpad.net/fuel/+spec/network-requirements-popup - good progress, most likely will land this or next week
16:51:14 <vkramskikh> 5) https://blueprints.launchpad.net/fuel/+spec/node-display-ip-address - good progress, most likely will land this or next week
16:51:17 <vkramskikh> 6) Ability to show all network groups - part of https://blueprints.launchpad.net/fuel/+spec/multirack-in-fuel-ui which wasn't done in 8.0 - good progress, no ETA yet
16:51:20 <vkramskikh> 7) https://blueprints.launchpad.net/fuel/+spec/allow-choosing-nodes-for-provisioning-and-deployment - design is in progress
16:51:23 <vkramskikh> 8) https://blueprints.launchpad.net/fuel/+spec/separate-fuel-ui-repo - slow progress, waiting for review of https://review.fuel-infra.org/#/c/16627/
16:51:26 <vkramskikh> 9) NFV stuff - waiting for backend and design finalization/negotiation
16:51:28 <vkramskikh> Questions?
16:52:26 <angdraug> xarses: you want to bring up qemu checkboxes?
16:52:39 <xarses> ya
16:52:56 <xarses> what about https://bugs.launchpad.net/fuel/+bug/1536836
16:52:58 <openstack> Launchpad bug 1536836 in Fuel for OpenStack "KVM/QEMU button combined in wizard and separate in settings" [High,In progress] - Assigned to Andrew Woodward (xarses)
16:53:22 <vkramskikh> this is partners bug, but we can take care of it if you want
16:53:35 <vkramskikh> most of work here is work in puppet manifests
16:54:06 <vkramskikh> and openstack.yaml, no UI code modifications required
16:54:22 <xarses> I proposed last night with the help of mwhahaha to add auto feature, but there is still discussion on  qemu vs kvm
16:54:52 <xarses> holser_: ?
16:55:19 <holser_> I’m not sure if it’s good to have it in 8.0
16:55:19 <angdraug> I think this is a major UX risk that we should address in 8.0
16:55:26 <holser_> it’s not
16:55:54 <xarses> then we need to put the wizard in 8.0 back to what it is in 7.0, two choices
16:56:01 <angdraug> if the current behaviour will lead every other operator to erroneously disable kvm in production, it is
16:56:15 <holser_> yeah, that’s what I think
16:56:35 <holser_> for instance I have environment with opterons and xeons
16:56:43 <holser_> and I want to have live migration
16:57:12 <holser_> that’s a big risk thus I wanted to discuss it publically
16:57:44 <holser_> moving back to 7.0 wizard seems to be better option
16:57:59 <xarses> thats a whole other story IMO, we would need to support not enabling all the cpu flags
16:58:14 <holser_> yep!
16:59:05 <xarses> I still don't see the operator taking this into consideration when they build their cloud
16:59:24 <holser_> Should we update our documentation?
16:59:25 <xarses> ok, I will propose back to 7.0 style in UI
16:59:31 <holser_> that can be fixed that way
16:59:49 <xarses> and we can discuss 9.0 plans on the ML
17:00:05 <holser_> sure
17:00:19 <xarses> #action xarses to create thread over QEMU/KVM on ML
17:00:23 <xarses> thanks everyone
17:00:29 <xarses> #endmeeting