Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel-dev, 2014-06-19

| Channels | #fuel-dev index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:25 rmoe joined #fuel-dev
01:18 xarses joined #fuel-dev
04:16 vkozhukalov joined #fuel-dev
05:13 dshulyak_ joined #fuel-dev
05:47 salmon_ joined #fuel-dev
06:19 vkozhukalov joined #fuel-dev
06:29 dshulyak_ joined #fuel-dev
06:44 al_ex6 joined #fuel-dev
06:45 e0ne joined #fuel-dev
07:07 mrasskazov joined #fuel-dev
07:07 izinovik joined #fuel-dev
07:07 MiroslavAnashkin joined #fuel-dev
07:07 holser joined #fuel-dev
07:20 al_ex6 joined #fuel-dev
07:25 sanek_ joined #fuel-dev
07:27 mattymo joined #fuel-dev
07:28 AndreyDanin joined #fuel-dev
07:28 dteselkin joined #fuel-dev
07:28 jeremydei joined #fuel-dev
07:28 ogelbukh joined #fuel-dev
07:29 igormarnat joined #fuel-dev
07:39 sbartel joined #fuel-dev
07:43 mihgen bookwar: http still works. Did you change DNS record?
07:44 mihgen bookwar: I'm about srv11
07:50 bookwar yes, IT redirected srv11 domain name to srv08 yesterday
07:50 bookwar all the http links now
07:56 saibarspeis joined #fuel-dev
07:57 mihgen bookwar: excellent
08:05 ddmitriev joined #fuel-dev
08:05 e0ne joined #fuel-dev
08:23 e0ne joined #fuel-dev
08:26 Arminder- joined #fuel-dev
08:45 akasatkin Webinar: an introduction to the elk stack
08:45 akasatkin http://www.elasticsearch.org/webinars/introduction-elk-stack/
08:57 mihgen bogdando: hi, what's the status of https://blueprints.launchpad.net/fuel/+spec/nailgun-unified-object-model ? looks like it is the same as api versioning which was implemented
09:04 bogdando mihgen, hi. I believe this BP should be a base for rolling upgrades/downgrades updates/rollback automation efforts both for Fuel and OSt deployments
09:05 bogdando AFAIK, some ideas have been already taken from it by the Nailgun team (E. Lee and A. Kasatkin should know more)
09:06 bogdando So if there would be other common concerns, they should be put to that base BP as well.
09:06 mihgen evgeniyl: akasatkin: what would you say on it? ^^^
09:07 bogdando I believe that is a kind of an abstract interface class in OOP ;)
09:07 bogdando and we could derive from it anything we want for other dependend blueprints (at least it was the idea)
09:09 evgeniyl bogdando: we didn't implement this approach, what we have right now is versioned api (rest api and rpc), I think it's more then enough for our cases.
09:11 akasatkin mihgen: simplified scheme is used in updates now (smaller quantity of states). afaic, it can be expanded if we want to support automatic update/rollback.
09:11 e0ne_ joined #fuel-dev
09:14 bogdando akasatkin, evgeniyl Please correct me, then it is a time for automation comes, we could have borrowed some more extended states and upgrade script concerns from this BP, right?
09:15 bogdando the idea is to design such an upgrade scripts which could communicate with Nailgun via API and put/get its rolling state
09:15 bogdando that would allow us to postpone and ensure automation even for offline nodes
09:16 bogdando so I believe this BP should be tightly alligned with the automation user story for rolling upgrades and ko (I mean downgrades, updates, rollbacks stuff)
09:17 akasatkin bogdando: yes, this can be used for MOS update at least
09:21 lromagnoli joined #fuel-dev
09:25 mihgen dshulyak: do you know why we have so many failures here? https://fuel-jenkins.mirantis.com/job/stackforge-verify-fuel-web/
09:25 mihgen 45 min timeout
09:26 dshulyak mihgen: see python chat
09:26 mihgen let's discuss it here, other folks affected too
09:26 mihgen izinovik
09:26 mihgen looks like critical thing to fix. wondering how it got merged to master
09:30 dshulyak fuelclient tests in master are broken, i can reproduce it locally,i hope to update issue status soon
09:31 geekinut1h joined #fuel-dev
09:33 Loic____ joined #fuel-dev
09:39 e0ne joined #fuel-dev
09:41 ykotko joined #fuel-dev
10:29 vrovachev joined #fuel-dev
10:33 ykotko joined #fuel-dev
10:55 teran joined #fuel-dev
10:56 mihgen nurla: assigned you as QA for https://blueprints.launchpad.net/fuel/+spec/rabbitmq-cluster-controlled-by-pacemaker
10:57 mihgen holser: did we implement https://blueprints.launchpad.net/fuel/+spec/synched-memcached-for-ha ?
10:57 mihgen or there are some more items to complete?
11:05 e0ne joined #fuel-dev
11:09 holser mihgen, it’s already implemented
11:10 holser it’s in 5.0  < in production
11:10 mihgen holser: let's close it then...
11:14 Tatyanka_Leontov joined #fuel-dev
11:23 teran joined #fuel-dev
11:31 artem_panchenko joined #fuel-dev
11:36 mattymo alex_didenko, bogdando ping
11:36 mattymo got an erb syntax question question
11:36 mattymo oops only 1 question
11:37 mihgen holser: xenolog will you provide status on HA improvements during demo today? https://etherpad.openstack.org/p/fuel-demo-agenda
11:38 holser I do
11:40 mihgen holser: ok, thanks
11:43 mattymo holser, your -1 on https://review.openstack.org/#/c/98994/ is wrong :)
11:43 mattymo can you fix it?
11:43 mattymo dns is a local variable, not class scope
11:44 holser I didn’t give -1
11:44 holser I just commented
11:45 holser mattymo ^^^
11:51 teran_ joined #fuel-dev
11:55 mihgen folks https://bugs.launchpad.net/fuel/+bug/1323705 is customer-found. (vnc console works once in 3 attempts in HA mode)
11:55 mihgen is it really low-hanging-fruit?
11:56 mihgen we need to fix this in 5.0.1 and if possible to get fix earlier than later
11:57 bogdando mihgen, it should be according to https://bugs.launchpad.net/fuel/+bug/1323705/comments/7
11:59 mihgen sbog: are you working on this? ^^^
12:30 zynzel i hear from someone (dont remember who), that we will abandon non-HA setup in fuel
12:30 zynzel this is true?
12:52 scroiset_ zynzel: I don't think so .. never heard about it .. but don't trust me, I'm pretty newbie with fuel ;)
12:54 zynzel hello swann, bartosz here, we talk during recrutation of france office ;)
12:56 scroiset_ hi bartosz!
12:57 BillTheKat joined #fuel-dev
13:11 alexz joined #fuel-dev
13:34 teran joined #fuel-dev
13:38 Loic____ hello, I'm working on fuel customisazion. In previous version (4.1) I do some modifications on fuel UI (modifiy openstack.yaml and then use the manage.py script to sync the db). I want to know if there is a clean way to do something like thaht in 5.0 ? I see 5.0 use docker, can we modify only the ui image ? and reload it in docker ?
13:53 anteaya joined #fuel-dev
13:56 evgeniyl Loic____: you can login into the container and make changes right there, if you want to change nailgun use command `dockerctl shell nailgun`, if you want to change static files, they are in nginx container `dockerctl shell nginx`
13:57 teran joined #fuel-dev
13:57 evgeniyl Loic____: and yes, you can upload your own image with modified nailgun/ui/something else.
14:00 alexz joined #fuel-dev
14:02 alexz evgeniyl: don't forget after changes do " docker commit $cont.name" ?
14:03 alexz Can somebody share some links with status\docs\implementation  fuel+authentication? (fuel-api\fuel-web + ldap \keystone or etc)
14:03 evgeniyl alexz: as far as I know we use single container during all master node live cycle, we don't create new container each time when we restart service, as result I don't think that we need to commit changes in container.
14:04 evgeniyl alexz: are you sure that we need to commit container?
14:05 alexz evgeniyl: yep, but if node\container reboot, changes was destroyed, no? (first-looking into docker story)
14:05 alexz evgeniyl: nope :) its question) i can check in few min
14:06 evgeniyl alexz: I didn't check it, but I believe that containers are persistent.
14:07 alexz evgeniyl: yep, checked-files doesn't disappear . sorry
14:11 evgeniyl alexz: it's ok, very often docker users use `docker run` command to run some process, and in this case they just create new container each time, in fuel we create container with specific name, like fuel-core-5.0-nailgun, and we use this name to run the same container.
14:17 e0ne joined #fuel-dev
14:17 nzrf_ joined #fuel-dev
14:21 vkozhukalov joined #fuel-dev
14:29 Loic____ Ok thx :) !
14:35 jobewan joined #fuel-dev
14:41 ruhe bookwar: hi! i received a bug report https://bugs.launchpad.net/fuel/+bug/1332099 ; it is assigned to mos-ceilometer group. can i re-assign it to a specific person? is there a procedure description we should follow?
14:54 vkozhukalov joined #fuel-dev
14:55 teran joined #fuel-dev
15:00 e0ne joined #fuel-dev
15:04 bookwar ruhe: hi, you can define the particular workflow for your team, but generally speaking on creation the bug is assigned to the team responsible for the topic. And when someone takes it to work on, he reassignes the bug to himself
15:05 ruhe bookwar: ok. thank you
15:07 casanch1 joined #fuel-dev
15:11 igormarnat ruhe: exactly, this is the process. You can assign it to specific engineer, tag will allow you to track the bug in the project. Group is assigned initially to be informed about the fact
15:15 bookwar igormarnat: and this bug should be in mos project
15:15 ruhe bookwar: can i get permission to re-assign tasks?
15:27 mihgen igormarnat: we need inform everyone on what we just agreed today… (where to file bug, assignee group, tags)
15:28 mihgen ideally I think we need to put this info on wiki and send email to everyone pointing to it
15:32 sbog left #fuel-dev
15:32 igormarnat mihgen: as per our discussion, I updated engineering process description with this information and distributed across my team. Feel free to update it and distribute across yours:)
15:33 mihgen igormarnat: can you forward then to fuel-core-team@
15:35 mihgen bookwar: is it https://review.openstack.org/#/c/86815/ failed because of CI?
15:35 igormarnat mihgen: Dude! I'll do it for you! :)
15:35 mihgen igormarnat: thanks a lot! :)
15:35 bookwar mihgen: yes, our Fuel CI for fuel-web repo was broken
15:36 nzrf_ joined #fuel-dev
15:36 bookwar it's fixed now, so i am going to retrigger failed tasks
15:40 vkramskikh hi guys
15:40 vkramskikh as for https://bugs.launchpad.net/fuel/+bug/1281838
15:40 vkramskikh so what we decide?
15:40 vkramskikh how do we fix this?
15:41 vkramskikh i see 2 viable options
15:41 mihgen brain461: ping
15:41 vkramskikh 1) we create a user for ostf
15:41 mihgen brain461: are you around?
15:41 vkramskikh 2) we ask a user for a password before running ostf. the user can enter the password if it is changed
15:42 teran joined #fuel-dev
15:42 mihgen vkramskikh: I think we need to run this question in ML...
15:42 mihgen doesn't seem to be trivial, many people have different opinions
15:43 vkramskikh mihgen: nice idea, will do
15:43 sbog joined #fuel-dev
15:43 teran_ joined #fuel-dev
15:44 mihgen vkramskikh: go to openstack-dev@
15:53 angdraug joined #fuel-dev
15:57 e0ne joined #fuel-dev
16:03 e0ne joined #fuel-dev
16:07 mattymo angdraug, whenever you get a moment: https://review.openstack.org/#/c/96776/3 and then https://review.openstack.org/#/c/98994/
16:07 mattymo angdraug, and https://review.openstack.org/#/c/98743/ you +2'd but I had -1 workflow
16:08 xarses joined #fuel-dev
16:20 geekinutah joined #fuel-dev
16:38 geekinut1h joined #fuel-dev
16:40 rmoe joined #fuel-dev
16:43 e0ne joined #fuel-dev
16:43 geekinutah joined #fuel-dev
16:55 geekinutah joined #fuel-dev
17:01 lromagnoli joined #fuel-dev
17:01 xarses xenolog: do we need to do anything to the corosync server config to support removing shadow, or for now do we just not worry about it?
17:02 xenolog no
17:02 xenolog but we should create cs_resources, locations, co_localions only while deploy primary-controller.
17:03 xarses xenolog: I figured as much
17:04 angdraug mattymo: merged 2 out of 3, need more +1's on the last one
17:04 xarses angdraug: which?
17:04 angdraug xarses: https://review.openstack.org/98994
17:05 angdraug would be nice if Artem could confirm that he tested that one, too
17:05 xarses what happens if we revert, who will run puppet on it to get it back to the state we need?
17:06 xarses or am I OT here?
17:06 xarses mattymo: ?
17:07 mattymo xarses, ? what?
17:07 mihgen angdraug: looks like ready for merge: https://review.openstack.org/#/c/96429/
17:07 mattymo angdraug, for 98994 I don't think artem tested it
17:08 xarses if I revert the container, is it to before, or after puppet runs? ie, will it be in the same state as after bootstrap_admin_node completes?
17:08 xarses also, can we get an alias for copy to cp?
17:08 mattymo same stat as after bootstrap_admin_node
17:08 mattymo state*
17:09 xarses ok, so If i use fuel-menu to change something and then revert the container, it would apply the new settings
17:09 xarses sweet
17:10 angdraug mattymo: indeed. merged
17:10 mattymo xarses, https://review.openstack.org/#/c/98743/ was just merged. if you have a newer iso in the last hour, then it will work right away
17:10 mihgen angdraug: thanks. Where do we get merged specs by the way?
17:10 mihgen what
17:10 mihgen 's the link?
17:10 mattymo but if you are using an older iso, you should copy /etc/asute.yaml to /etc/fuel/astute.yaml
17:12 angdraug mihgen: I don't have a link for that
17:12 mihgen vkozhukalov: you may be? ^^^
17:12 * xarses rips out cs_shadow joyfully
17:12 mihgen it should be on official openstack specs page...
17:13 angdraug xarses: don't be in a hurry to rip out cs_shadow. saviour it
17:13 xarses if it makes it work...
17:13 angdraug s/saviour/savour/
17:13 xarses oh, I'm enjoying it
17:13 xarses I've already lost a good nights sleep over it
17:15 angdraug mattymo: 98994 merged, small enough to go on two people reviewing the code
17:15 mattymo miracle!
17:15 angdraug I assume you tested it locally of course :D
17:15 mattymo angdraug, usually I have to throw heavy objects aroudn the office
17:15 mattymo yeah I did
17:15 mattymo I think
17:15 mattymo )
17:16 angdraug you know your fingers are on the fat side, right? ;p
17:16 mattymo they are quite similar to sausages
17:16 angdraug mihgen: what happened to the intro hangout?
17:16 mihgen angdraug: it's there
17:17 mihgen connect
17:17 mihgen just started
17:19 xarses mattymo: the cp alias would still be nice =)
17:19 mattymo xarses, patches welcome
17:19 xarses I reviewed it
17:19 xarses =P
17:20 xarses going to make you a full fledged docker tool suite
17:21 lromagnoli joined #fuel-dev
17:22 xarses ohh, of fun note xenolog mihgen if we only do cs_resource and location on primary controller, we will save a lot of time with deployment of additional controllers. should also make scale out easier
17:23 angdraug xarses: can we get away with that?
17:23 angdraug you keep promising nice stuff
17:29 e0ne joined #fuel-dev
17:34 xarses los
17:34 xarses lols even
17:34 xarses it runs now
17:34 bookwar I've set the temporary job at fuel ci to build fuel-specs on merge. The result, i.e. the current state of fuel-specs repo is now available here https://fuel-jenkins.mirantis.com/job/merged-fuel-specs/Fuel_Development_Specs_build_results/
17:34 * xarses does the no cs_shadow happy dance
17:35 * xarses stabs cs_shadow a dozen more times
17:35 bookwar it is temporary solution because we are waiting for specs.opentsack.org
17:45 xarses angdraug: http://paste.openstack.org/show/84514/
17:46 xarses repeatable =))
17:52 angdraug don't think that pastie all came through
17:53 e0ne joined #fuel-dev
18:10 lromagnoli joined #fuel-dev
18:10 rmoe joined #fuel-dev
18:10 xarses joined #fuel-dev
18:10 angdraug joined #fuel-dev
18:10 sbog joined #fuel-dev
18:10 casanch1 joined #fuel-dev
18:10 jobewan joined #fuel-dev
18:10 anteaya joined #fuel-dev
18:10 ykotko joined #fuel-dev
18:10 Arminder- joined #fuel-dev
18:10 saibarspeis joined #fuel-dev
18:10 igormarnat joined #fuel-dev
18:10 ogelbukh joined #fuel-dev
18:10 jeremydei joined #fuel-dev
18:10 dteselkin joined #fuel-dev
18:10 AndreyDanin joined #fuel-dev
18:10 mattymo joined #fuel-dev
18:10 sanek_ joined #fuel-dev
18:10 holser joined #fuel-dev
18:10 MiroslavAnashkin joined #fuel-dev
18:10 izinovik joined #fuel-dev
18:10 mrasskazov joined #fuel-dev
18:10 salmon_ joined #fuel-dev
18:10 bogdando joined #fuel-dev
18:10 iyozhikov joined #fuel-dev
18:10 YorikSar joined #fuel-dev
18:10 rshon joined #fuel-dev
18:10 sbanka joined #fuel-dev
18:10 dilyin joined #fuel-dev
18:10 enikanorov_ joined #fuel-dev
18:10 bas___ joined #fuel-dev
18:10 zynzel joined #fuel-dev
18:10 dmitryme joined #fuel-dev
18:10 ChanServ joined #fuel-dev
18:10 aignatov joined #fuel-dev
18:10 Bomfunk_ joined #fuel-dev
18:10 mirrorbox joined #fuel-dev
18:10 ruhe joined #fuel-dev
18:10 christopheraedo joined #fuel-dev
18:10 evgeniyl joined #fuel-dev
18:10 aedocw joined #fuel-dev
18:10 meow-nofer_ joined #fuel-dev
18:10 enikanorov joined #fuel-dev
18:10 alex_didenko joined #fuel-dev
18:10 dpyzhov joined #fuel-dev
18:10 scroiset_ joined #fuel-dev
18:10 aglarendil joined #fuel-dev
18:10 higgins joined #fuel-dev
18:10 openstackstatus joined #fuel-dev
18:10 katyafervent_awa joined #fuel-dev
18:10 xenolog joined #fuel-dev
18:10 vkramskikh joined #fuel-dev
18:10 monester joined #fuel-dev
18:10 evg joined #fuel-dev
18:10 kat_pimenova joined #fuel-dev
18:10 jkirnosova joined #fuel-dev
18:10 apalkina joined #fuel-dev
18:10 akislitsky joined #fuel-dev
18:10 MorAle joined #fuel-dev
18:10 meow-nofer joined #fuel-dev
18:10 mihgen joined #fuel-dev
18:10 dburmistrov joined #fuel-dev
18:10 rvyalov joined #fuel-dev
18:10 nurla joined #fuel-dev
18:10 justif joined #fuel-dev
18:11 angdraug <4 minutes?
18:11 xarses for neutron ha, yes
18:11 xarses not all of neutron
18:12 mihgen xarses: what if compare With debug?
18:13 xarses its an extra near 1k lines, and takes maybe 10-20 sec more
18:16 lromagnoli joined #fuel-dev
18:16 rmoe joined #fuel-dev
18:16 xarses joined #fuel-dev
18:16 angdraug joined #fuel-dev
18:16 sbog joined #fuel-dev
18:16 casanch1 joined #fuel-dev
18:16 jobewan joined #fuel-dev
18:16 anteaya joined #fuel-dev
18:16 ykotko joined #fuel-dev
18:16 Arminder- joined #fuel-dev
18:16 saibarspeis joined #fuel-dev
18:16 igormarnat joined #fuel-dev
18:16 ogelbukh joined #fuel-dev
18:16 jeremydei joined #fuel-dev
18:16 dteselkin joined #fuel-dev
18:16 AndreyDanin joined #fuel-dev
18:16 mattymo joined #fuel-dev
18:16 sanek_ joined #fuel-dev
18:16 holser joined #fuel-dev
18:16 MiroslavAnashkin joined #fuel-dev
18:16 izinovik joined #fuel-dev
18:16 mrasskazov joined #fuel-dev
18:16 salmon_ joined #fuel-dev
18:16 bogdando joined #fuel-dev
18:16 iyozhikov joined #fuel-dev
18:16 YorikSar joined #fuel-dev
18:16 rshon joined #fuel-dev
18:16 sbanka joined #fuel-dev
18:16 dilyin joined #fuel-dev
18:16 enikanorov_ joined #fuel-dev
18:16 bas___ joined #fuel-dev
18:16 zynzel joined #fuel-dev
18:16 dmitryme joined #fuel-dev
18:16 ChanServ joined #fuel-dev
18:16 aignatov joined #fuel-dev
18:16 Bomfunk_ joined #fuel-dev
18:16 mirrorbox joined #fuel-dev
18:16 ruhe joined #fuel-dev
18:16 christopheraedo joined #fuel-dev
18:16 evgeniyl joined #fuel-dev
18:16 aedocw joined #fuel-dev
18:16 meow-nofer_ joined #fuel-dev
18:16 enikanorov joined #fuel-dev
18:16 alex_didenko joined #fuel-dev
18:16 dpyzhov joined #fuel-dev
18:16 scroiset_ joined #fuel-dev
18:16 aglarendil joined #fuel-dev
18:16 higgins joined #fuel-dev
18:16 openstackstatus joined #fuel-dev
18:16 katyafervent_awa joined #fuel-dev
18:16 xenolog joined #fuel-dev
18:16 vkramskikh joined #fuel-dev
18:16 monester joined #fuel-dev
18:16 evg joined #fuel-dev
18:16 kat_pimenova joined #fuel-dev
18:16 jkirnosova joined #fuel-dev
18:16 apalkina joined #fuel-dev
18:16 akislitsky joined #fuel-dev
18:16 MorAle joined #fuel-dev
18:16 meow-nofer joined #fuel-dev
18:16 mihgen joined #fuel-dev
18:16 dburmistrov joined #fuel-dev
18:16 rvyalov joined #fuel-dev
18:16 nurla joined #fuel-dev
18:16 justif joined #fuel-dev
18:16 bookwar joined #fuel-dev
18:22 xarses mihgen: timed readl 3m40.412s  with debug
18:23 xarses so +1 sec =)
18:24 mihgen ))
18:27 xarses mihgen: ok so HA is straightened out (as long as we go with removing cs_shadow) going to start plugging in the rest of the wires =)
18:35 teran joined #fuel-dev
18:35 angdraug bookwar: is http://docs.mirantis.com/fuel-dev/ auto-updated by jenkins?
18:36 a_teem joined #fuel-dev
18:37 mihgen angdraug: it should be
18:38 mihgen xarses: you mean you is able to run neutron ha with upstream puppet module and ml2?
18:38 angdraug doesn't look like this got merged: https://review.openstack.org/#/c/100354/5/docs/develop/env.rst
18:40 xarses no ml2, just our ha composed on top of monolithic ovs its ~= to what we have now. I'm plugging the rest in so we can use the whole upstream module instead of the hacks i had against the agents to let me do the HA testing
18:40 xarses I was worried that if we couldn't get HA to work then there was no point in plugging the rest of the module in
18:40 mihgen xarses: cool
18:42 * mihgen anyone wants to break fuel? go try! http://fuel-demo.mirantis.com:8000/
18:42 angdraug ^ err, merged>built
18:43 xarses mihgen: boo, i thought it would at least be real vms
18:43 xarses fake-ui can't join my bot army
18:43 bookwar angdraug: should have been build, but wasn't
18:45 mihgen haha nope, real vms  - it's still slow
18:45 mihgen here you can run deployment in 10 sec :)
18:48 sbog mihgen: how it deploying so fast? ) It some kind of trick?
18:49 mihgen sbog: it's has a special rocket fuel ;)
18:57 casanch1 mihgen: ok, thanks
18:57 artem_panchenko joined #fuel-dev
19:05 vkozhukalov joined #fuel-dev
19:08 dshulyak_ joined #fuel-dev
19:08 xarses casanch1: welcome to fuel
19:08 xarses Where are you working out of?
19:13 dilyin NAMES
19:15 teran joined #fuel-dev
19:19 dilyin xarses: hello, they say you have some questions about pacemaker?
19:19 xarses dilyin: yes, was fixed by removing cs_shadow
19:20 xarses otherwise, I couldn't get the order to work
19:20 dilyin have you seen my doc about pacemaker improvements?
19:20 xarses the cs_order and or cs_colocate would inevatabley run before one of the cs_resource defines
19:20 xarses yes
19:21 xarses looks nice
19:21 xarses also holser and xenolog said that they are removing cs_shadow anyway for galera
19:21 dilyin i'm suggesting to get rid of these shadows one way or another
19:22 xarses https://review.openstack.org/#/c/95764/
19:22 xarses dilyin: I agree, simply removing it fixed my ordering problem
19:22 holser xarses, not only for galera
19:22 holser for RabbitMQ OCF also
19:22 holser as we keep some values in CIB, and update them
19:22 dilyin yes, and in our current code base there is not a single place where they are actually needed
19:22 xarses holser: =) yes I see ceilometer and neutron in that commit too
19:24 dilyin they were introduced long ago by vkuklin to make adding constraints and primitives be done in a single transaction. but we don't have a situaltion when not having transactions could actually harm
19:25 bookwar angdraug: i've fixed http://docs.mirantis.com/fuel-dev/ for now, but it is still unclear why CMS polling didn't work properly. I'll research later
19:25 dilyin ordering problems you are having is that "unnecessary complexity" I was writing about
19:25 angdraug bookwar: thanks!
19:25 xarses dilyin: yes, It was painful to try and get the ordering to work
19:26 dilyin xarses: join the club of asymmetric cluster scheme supporters)
19:27 xarses dilyin: from the manifest side, is there anything I should do now to make it easier to support
19:27 xarses since we might not get cs_shadow to work in neutron again if we merge this upstream
19:28 xarses (by this upstream i mean merge the upstream puppet neutron into fuel)
19:28 e0ne joined #fuel-dev
19:28 dilyin i belive there is no shadow support
19:29 dilyin the only corosync module implemetation that bothered about them is ours
19:29 xarses dilyin: it appears to be in upstream puppetlabs-corosync now, but was recently added
19:30 xarses I tried using the module and it didn't make it any easier
19:30 xarses I did barely find a blog post that even talked about what shadow is used for
19:31 dilyin puppetlabs' modules is very far from being useful
19:31 casanch1_ joined #fuel-dev
19:36 e0ne joined #fuel-dev
19:37 igormarnat mihgen: Can I deploy env prepared at http://fuel-demo.mirantis.com:8000/?
19:41 e0ne joined #fuel-dev
19:42 mihgen igormarnat: you want to deploy it on your machine? if so, yes, follow http://docs.mirantis.com/fuel-dev/develop/nailgun/development/env.html#running-nailgun-in-fake-mode
19:42 mihgen file a bug if it doesn't work :)
19:44 igormarnat Uhm, I just want to press "Deploy changes" on Bacon
19:48 mihgen igormarnat: just do it
19:49 e0ne joined #fuel-dev
19:49 igormarnat mihgen: doesn't work. It requires controller, and there are no nodes available for that
19:50 mihgen igormarnat: remove one of the existing envs then first
19:50 mihgen you will get a few free nodes
19:50 mihgen add and deploy (but remove offline first)
19:51 mihgen env with offline node will go to error saying that it can't deploy on offline nodes :)
19:51 igormarnat Having done some reconfiguration
19:51 igormarnat Got an err message
19:51 igormarnat Number of OSD nodes (0) cannot be less than the Ceph object replication factor (2). Please either assign ceph-osd role to more nodes, or reduce Ceph replication factor in the Settings tab.
19:51 igormarnat Where can I take my prizes?
19:51 mihgen yep it's valid message ;)
19:52 igormarnat Wait, man
19:52 igormarnat No prizes yet?
19:52 mihgen no prizes yet) you get your prize when you break an env with exception internally)
19:52 igormarnat Ok, let me see ...
19:55 igormarnat Am I supposed to be able to open Horizon at http://172.16.0.2/
19:55 igormarnat After the deployment?
19:55 igormarnat This webpage is not available
19:57 mihgen of course not if it doesn't deploy anything ;)
19:57 e0ne joined #fuel-dev
19:58 mihgen igormarnat: or you think I'm magician and can make openstack env in 10sec?)
19:58 igormarnat Good catch. I almost believed it:)
20:08 casanch1 joined #fuel-dev
20:16 nzrf_ joined #fuel-dev
20:34 casanch1 joined #fuel-dev
20:44 angdraug joined #fuel-dev
21:12 geekinutah joined #fuel-dev
21:22 geekinutah joined #fuel-dev
21:37 geekinutah joined #fuel-dev
22:32 lromagnoli joined #fuel-dev

| Channels | #fuel-dev index | Today | | Search | Google Search | Plain-Text | summary