| *** banix has joined #openstack-kolla | 00:01 | |
| *** betheynyx has joined #openstack-kolla | 00:08 | |
| *** salv-orlando has joined #openstack-kolla | 00:12 | |
| *** sdake has quit IRC | 00:15 | |
| *** ravig has joined #openstack-kolla | 00:16 | |
| *** salv-orlando has quit IRC | 00:17 | |
| *** rstarmer_ has quit IRC | 00:19 | |
| *** daneyon has joined #openstack-kolla | 00:21 | |
| *** rstarmer has joined #openstack-kolla | 00:29 | |
| *** thumpba has quit IRC | 00:30 | |
| *** sdake has joined #openstack-kolla | 00:32 | |
| *** daneyon_ has joined #openstack-kolla | 00:39 | |
| *** sdake_ has joined #openstack-kolla | 00:42 | |
| *** daneyon has quit IRC | 00:42 | |
| *** sdake has quit IRC | 00:44 | |
| *** daneyon_ has quit IRC | 00:44 | |
| *** jasonsb has joined #openstack-kolla | 00:49 | |
| *** jasonsb has quit IRC | 00:54 | |
| *** rstarmer has quit IRC | 01:03 | |
| *** rstarmer has joined #openstack-kolla | 01:03 | |
| *** salv-orlando has joined #openstack-kolla | 01:13 | |
| *** salv-orlando has quit IRC | 01:18 | |
| *** rstarmer has quit IRC | 01:18 | |
| *** rstarmer has joined #openstack-kolla | 01:19 | |
| *** inc0 has joined #openstack-kolla | 01:27 | |
| *** ravig has quit IRC | 01:30 | |
| *** ravig has joined #openstack-kolla | 01:31 | |
| *** rstarmer_ has joined #openstack-kolla | 01:33 | |
| *** rstarmer has quit IRC | 01:33 | |
| *** ravig has quit IRC | 01:40 | |
| *** banix has quit IRC | 01:45 | |
| *** mdnadeem has quit IRC | 01:46 | |
| *** mdnadeem_ has joined #openstack-kolla | 01:48 | |
| *** ravig has joined #openstack-kolla | 01:48 | |
| *** rstarmer has joined #openstack-kolla | 01:48 | |
| *** rstarmer_ has quit IRC | 01:50 | |
| *** banix has joined #openstack-kolla | 01:50 | |
| *** rstarmer has quit IRC | 01:59 | |
| sbezverk | sdake ping | 02:03 |
|---|---|---|
| *** ravig has quit IRC | 02:10 | |
| *** salv-orlando has joined #openstack-kolla | 02:14 | |
| *** salv-orlando has quit IRC | 02:19 | |
| *** ravig has joined #openstack-kolla | 02:19 | |
| *** ravig has quit IRC | 02:29 | |
| *** ravig has joined #openstack-kolla | 02:30 | |
| *** inc0_ has joined #openstack-kolla | 02:45 | |
| *** inc0 has quit IRC | 02:49 | |
| *** jasonsb has joined #openstack-kolla | 02:51 | |
| *** jasonsb has quit IRC | 02:55 | |
| *** klint has joined #openstack-kolla | 02:56 | |
| *** coolsvap has joined #openstack-kolla | 02:59 | |
| *** rstarmer has joined #openstack-kolla | 03:01 | |
| *** rstarmer has quit IRC | 03:05 | |
| *** mdnadeem has joined #openstack-kolla | 03:07 | |
| *** mdnadeem_ has quit IRC | 03:10 | |
| *** inc0__ has joined #openstack-kolla | 03:11 | |
| *** betheynyx has quit IRC | 03:11 | |
| *** rstarmer has joined #openstack-kolla | 03:12 | |
| *** inc0_ has quit IRC | 03:14 | |
| *** salv-orlando has joined #openstack-kolla | 03:15 | |
| *** salv-orlando has quit IRC | 03:19 | |
| *** daneyon has joined #openstack-kolla | 03:22 | |
| *** daneyon has quit IRC | 03:27 | |
| *** inc0__ has quit IRC | 03:28 | |
| *** jpeeler has joined #openstack-kolla | 03:44 | |
| *** jpeeler has joined #openstack-kolla | 03:44 | |
| *** dave-mccowan has quit IRC | 04:01 | |
| sbezverk | sdake if you get this message, check this bug I hit, it applies to both master and stable/mitaka: https://bugs.launchpad.net/kolla/+bug/1577273 | 04:03 |
| openstack | Launchpad bug 1577273 in kolla "Failure of creating haproxy mysql user" [Undecided,New] - Assigned to Serguei Bezverkhi (sbezverk) | 04:03 |
| *** salv-orlando has joined #openstack-kolla | 04:16 | |
| *** salv-orlando has quit IRC | 04:20 | |
| *** rstarmer has quit IRC | 04:22 | |
| *** ravig has quit IRC | 04:32 | |
| sdake_ | evening | 04:45 |
| sdake_ | sbezverk i think my fix for removing kazoo should fix that sbezverk - could you try it? | 04:46 |
| *** manjeets has quit IRC | 04:47 | |
| *** manjeets has joined #openstack-kolla | 04:48 | |
| *** unicell has joined #openstack-kolla | 04:50 | |
| *** banix has quit IRC | 05:02 | |
| *** rstarmer has joined #openstack-kolla | 05:12 | |
| *** salv-orlando has joined #openstack-kolla | 05:16 | |
| *** rstarmer has quit IRC | 05:19 | |
| *** salv-orlando has quit IRC | 05:21 | |
| *** ssurana has joined #openstack-kolla | 05:21 | |
| *** daneyon has joined #openstack-kolla | 05:22 | |
| *** betheynyx has joined #openstack-kolla | 05:32 | |
| *** reddydodda has joined #openstack-kolla | 05:33 | |
| *** sdake_ has quit IRC | 05:47 | |
| *** Mr_Broken has joined #openstack-kolla | 05:48 | |
| *** sdake has joined #openstack-kolla | 05:52 | |
| *** jasonsb has joined #openstack-kolla | 05:53 | |
| *** ccesario_ has quit IRC | 05:55 | |
| *** ccesario_ has joined #openstack-kolla | 05:55 | |
| *** jasonsb has quit IRC | 05:58 | |
| *** salv-orlando has joined #openstack-kolla | 05:58 | |
| *** coolsvap has quit IRC | 06:26 | |
| *** coolsvap has joined #openstack-kolla | 06:30 | |
| *** mdnadeem_ has joined #openstack-kolla | 06:33 | |
| *** mdnadeem has quit IRC | 06:37 | |
| *** mdnadeem has joined #openstack-kolla | 06:41 | |
| *** mdnadeem_ has quit IRC | 06:43 | |
| *** fragatina has joined #openstack-kolla | 07:08 | |
| *** ssurana has quit IRC | 07:15 | |
| *** daneyon_ has joined #openstack-kolla | 07:19 | |
| *** daneyon has quit IRC | 07:22 | |
| *** salv-orlando has quit IRC | 07:28 | |
| *** fragatina has quit IRC | 07:46 | |
| *** rstarmer has joined #openstack-kolla | 07:58 | |
| *** dmk0202 has joined #openstack-kolla | 08:06 | |
| *** rstarmer has quit IRC | 08:13 | |
| *** gfidente has joined #openstack-kolla | 09:01 | |
| *** reddydodda has quit IRC | 09:37 | |
| *** fragatina has joined #openstack-kolla | 09:46 | |
| *** pbourke has quit IRC | 09:51 | |
| *** pbourke has joined #openstack-kolla | 09:51 | |
| *** fragatina has quit IRC | 09:52 | |
| *** reddydodda has joined #openstack-kolla | 09:53 | |
| openstackgerrit | Michal Rostecki proposed openstack/kolla: [WIP] Exclude the docker registry image from cleanup-images https://review.openstack.org/310677 | 09:59 |
| openstackgerrit | Michal Rostecki proposed openstack/kolla: [WIP] Exclude the docker registry image from cleanup-images https://review.openstack.org/310677 | 10:02 |
| *** ccesario_ has quit IRC | 10:16 | |
| *** betheynyx has quit IRC | 10:23 | |
| *** betheynyx_ has joined #openstack-kolla | 10:25 | |
| *** Serlex has joined #openstack-kolla | 10:28 | |
| *** coolsvap has quit IRC | 10:29 | |
| *** coolsvap has joined #openstack-kolla | 10:30 | |
| *** daneyon has joined #openstack-kolla | 10:35 | |
| *** Guest68119 has quit IRC | 10:37 | |
| *** zeus- has joined #openstack-kolla | 10:38 | |
| *** daneyon has quit IRC | 10:40 | |
| *** reddydodda has quit IRC | 10:42 | |
| *** fragatina has joined #openstack-kolla | 10:49 | |
| *** allen_gao has quit IRC | 10:49 | |
| *** Mr_Broken has quit IRC | 10:52 | |
| *** Mr_Broken has joined #openstack-kolla | 10:53 | |
| *** fragatina has quit IRC | 10:53 | |
| *** Mr_Broke_ has joined #openstack-kolla | 10:53 | |
| *** allen_gao has joined #openstack-kolla | 10:54 | |
| *** ccesario___ has joined #openstack-kolla | 10:54 | |
| *** rhallisey has joined #openstack-kolla | 10:54 | |
| *** Mr_Broken has quit IRC | 10:57 | |
| *** jasonsb has joined #openstack-kolla | 10:57 | |
| *** zeus- is now known as zeus | 10:58 | |
| *** zeus is now known as Guest25915 | 10:59 | |
| *** ccesario___ has quit IRC | 10:59 | |
| nihilifer | sdake: let me know when you'll awake | 10:59 |
| *** ccesario___ has joined #openstack-kolla | 10:59 | |
| *** jasonsb has quit IRC | 11:02 | |
| *** ccesario has joined #openstack-kolla | 11:07 | |
| *** reddydodda has joined #openstack-kolla | 11:19 | |
| openstackgerrit | Carlos Cesario proposed openstack/kolla: Fix radosgw 'rgw frontends' param https://review.openstack.org/311222 | 11:32 |
| *** thumpba has joined #openstack-kolla | 11:33 | |
| *** thumpba has quit IRC | 11:34 | |
| *** thumpba has joined #openstack-kolla | 11:43 | |
| openstackgerrit | Mauricio Lima proposed openstack/kolla: Remove unecessary blank lines https://review.openstack.org/311199 | 11:44 |
| *** coolsvap has quit IRC | 11:56 | |
| *** athomas_ has joined #openstack-kolla | 12:12 | |
| *** mlima has joined #openstack-kolla | 12:15 | |
| *** Mr_Broke_ has quit IRC | 12:16 | |
| *** Mr_Broken has joined #openstack-kolla | 12:17 | |
| *** Mr_Broken has quit IRC | 12:21 | |
| *** daneyon has joined #openstack-kolla | 12:24 | |
| *** Mr_Broken has joined #openstack-kolla | 12:26 | |
| *** daneyon has quit IRC | 12:28 | |
| *** mbound has joined #openstack-kolla | 12:30 | |
| openstackgerrit | Michal Rostecki proposed openstack/kolla: [WIP] Exclude the docker registry image from cleanup-images https://review.openstack.org/310677 | 12:37 |
| *** reddydodda has quit IRC | 12:46 | |
| *** athomas_ has quit IRC | 12:49 | |
| *** diogogmt has quit IRC | 12:52 | |
| *** dave-mccowan has joined #openstack-kolla | 12:53 | |
| *** diogogmt has joined #openstack-kolla | 12:57 | |
| *** jasonsb has joined #openstack-kolla | 13:00 | |
| *** sdake has quit IRC | 13:00 | |
| *** reddydodda has joined #openstack-kolla | 13:00 | |
| *** sdake has joined #openstack-kolla | 13:01 | |
| *** jtriley has joined #openstack-kolla | 13:02 | |
| *** klint has quit IRC | 13:02 | |
| *** sdake_ has joined #openstack-kolla | 13:04 | |
| *** jasonsb has quit IRC | 13:05 | |
| *** sdake has quit IRC | 13:06 | |
| *** thumpba has quit IRC | 13:10 | |
| *** thumpba has joined #openstack-kolla | 13:10 | |
| openstackgerrit | Steven Dake proposed openstack/kolla: Remove dependencies on kazoo and friends https://review.openstack.org/311632 | 13:13 |
| *** thumpba has quit IRC | 13:15 | |
| sdake_ | pbourke around? | 13:16 |
| *** thumpba has joined #openstack-kolla | 13:19 | |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla: Spec: Deploy Kolla images using Kubernetes https://review.openstack.org/304182 | 13:21 |
| *** banix has joined #openstack-kolla | 13:23 | |
| *** thumpba has quit IRC | 13:23 | |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla: Spec: Deploy Kolla images using Kubernetes https://review.openstack.org/304182 | 13:25 |
| *** thumpba_ has joined #openstack-kolla | 13:26 | |
| nihilifer | sdake_: around? | 13:32 |
| sdake_ | sup dude | 13:32 |
| openstackgerrit | Merged openstack/kolla: Remove the eth1 up and down manually in the rax-ord node https://review.openstack.org/310636 | 13:36 |
| *** rmart04 has joined #openstack-kolla | 13:38 | |
| openstackgerrit | Merged openstack/kolla: Use a random virtual_router_id in the keepalived https://review.openstack.org/304080 | 13:42 |
| *** dmsimard|afk is now known as dmsimard | 13:49 | |
| *** JoseMello has joined #openstack-kolla | 13:50 | |
| *** Guest25915 is now known as zeus` | 13:54 | |
| *** dwalsh has joined #openstack-kolla | 13:54 | |
| *** inc0 has joined #openstack-kolla | 13:58 | |
| *** fragatina has joined #openstack-kolla | 13:59 | |
| *** ravig has joined #openstack-kolla | 13:59 | |
| *** diogogmt has quit IRC | 14:03 | |
| *** fragatina has quit IRC | 14:03 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla: Addining ability to specify capabilities and security https://review.openstack.org/308447 | 14:04 |
| inc0 | o/ | 14:06 |
| *** sdake_ is now known as sdake | 14:11 | |
| *** rmart04 has quit IRC | 14:12 | |
| *** rmart04 has joined #openstack-kolla | 14:12 | |
| *** rmart04 has quit IRC | 14:13 | |
| sdake | can i get another core reviewer to ack https://review.openstack.org/#/c/290238/9 | 14:17 |
| patchbot | sdake: patch 290238 - kolla - Handle multi Ethernets case in the gate | 14:17 |
| sdake | hey inc0 | 14:17 |
| sdake | hey sbezverk | 14:17 |
| sdake | hey all :) | 14:17 |
| sbezverk | hey sdake | 14:18 |
| sdake | if your a core reviewer please vote on the general "lets get rolling on k8s" | 14:19 |
| sdake | btw, if we do have a split repo, it will have to be named kolla-k8s becuase kolla-kubernetes wont work in pypi for some reaon | 14:20 |
| inc0 | kolla-k8s is good | 14:20 |
| sdake | either someone registered it already (my guess) or kubernetes is not allowed or its too long who knows ;) | 14:20 |
| sdake | i like andreas's approach though - fit filter allows us to split anythign later down the road | 14:20 |
| sdake | and maintain all history | 14:20 |
| sdake | but anyway vote on the lets get the ball rollign on k8s, then we can sort out the repo | 14:20 |
| sdake | so far it looks split pretty even | 14:21 |
| inc0 | ehh, I'd rather have first split then merged, we can make history wirks | 14:21 |
| sbezverk | sdake I could not test your patch yet as I have a presentation at noon my time.. I made deployment working by little hack. | 14:21 |
| sdake | andreas is an expert in git | 14:21 |
| sdake | get him to say thato n the mailign list :) | 14:21 |
| inc0 | my point is, we can do both with history | 14:21 |
| sdake | sbezverk which patch | 14:21 |
| sdake | i disagree, a split then merge would lose history | 14:21 |
| sbezverk | sdake you mentioned kazoo | 14:21 |
| sdake | sbezverk oh that - ya that works | 14:22 |
| sdake | i tested it ;) | 14:22 |
| sdake | need a rubber stamp on https://review.openstack.org/#/c/290238/9 plz | 14:22 |
| patchbot | sdake: patch 290238 - kolla - Handle multi Ethernets case in the gate | 14:22 |
| sbezverk | sdake so you saw the same issue right? | 14:22 |
| sdake | the gate is seriously busted - if this doesn't fix it we can fix it some othe rway | 14:22 |
| sdake | sbezverk which issue | 14:22 |
| inc0 | sdake, http://stackoverflow.com/questions/13040958/merge-two-git-repositories-without-breaking-file-history | 14:23 |
| sdake | brain fry - going to have to give me somethign to work with :) | 14:23 |
| sdake | inc0 if you read that full thraed, the essence is "yes you can merge repos but lose history" | 14:24 |
| sbezverk | sdake https://bugs.launchpad.net/kolla/+bug/1577273 | 14:24 |
| openstack | Launchpad bug 1577273 in kolla "Failure of creating haproxy mysql user" [Critical,New] - Assigned to Serguei Bezverkhi (sbezverk) | 14:24 |
| sdake | look at the footnotes | 14:24 |
| sdake | sbezverk oh - ya it is speculatory if my fix fixes that problem | 14:24 |
| sdake | which demo are you giving if you dont mind me asking, nfvi? | 14:25 |
| inc0 | sdake, I'll try it locally | 14:25 |
| sbezverk | sdake yes nfv vt | 14:25 |
| sdake | bot hdims and ajeager have said its irreversible in private conversations | 14:25 |
| sdake | anyway up to the community to decide :) | 14:26 |
| sdake | atm vote is equal but many folks are unrepresented | 14:26 |
| inc0 | first I personally don't want bunch of cores in kolla (that I don't know) | 14:26 |
| sdake | inc0 that wont be the case | 14:26 |
| inc0 | second I don't want people to think k8s is ready | 14:26 |
| sdake | the new folks will go in kolla-k8s-core | 14:26 |
| rhallisey | sdake, ya but they = kolla core | 14:26 |
| inc0 | if they checkout kolla, everything there works | 14:27 |
| sdake | and be promoted to kolla-core as they show interest in the general repo | 14:27 |
| sdake | rhallisey this is where training and lecturing come in :) | 14:27 |
| inc0 | I think we need more than interest | 14:27 |
| rhallisey | there's no difference between kolla-k8-core and kolla-core | 14:27 |
| sdake | they also do not have policy voting rights | 14:27 |
| sdake | that is the difference | 14:27 |
| inc0 | every core should have equal chances/requirements to join the eam | 14:27 |
| rhallisey | how they not have policy voting rights? | 14:27 |
| rhallisey | they belong to the kolla repo | 14:28 |
| sdake | they may not commit to any directory other then kubernetes | 14:28 |
| sdake | *ONLY* people in kolla-core have policy voting rights | 14:28 |
| *** thumpba_ has quit IRC | 14:28 | |
| *** Mr_Broken has quit IRC | 14:28 | |
| sdake | just as it was with kolla-k8s-core | 14:28 |
| sdake | rather kolla-mesos-core | 14:28 |
| inc0 | sdake, this will be trust-based and I'm 100% sure someone will break it, even if untintentionally | 14:28 |
| rhallisey | except that was a differnt repo | 14:28 |
| openstackgerrit | Merged openstack/kolla: Handle multi Ethernets case in the gate https://review.openstack.org/290238 | 14:28 |
| sdake | if they do unintentionally their change will be reverted | 14:28 |
| *** thumpba has joined #openstack-kolla | 14:29 | |
| *** Mr_Broken has joined #openstack-kolla | 14:29 | |
| inc0 | also major raise on core team will make people think that we're less stabke | 14:29 |
| inc0 | I would... | 14:29 |
| sdake | if they do so intentionallly and it becomes a pattern they will be kicked from the kolla-k8s-core reviewer team | 14:29 |
| rhallisey | I'd rather let that community manage itself until it's ready | 14:29 |
| inc0 | we did kolla-mesos, we should follow pattern in kolla0k8s | 14:29 |
| sdake | they are added to a separate group | 14:29 |
| sdake | yes but that pattern was flawed | 14:29 |
| inc0 | also it will be hell to figure out who does reviews on main repo | 14:30 |
| sdake | what i'd like to see is people from kolla-mesos chime in on what they think went wrong/right | 14:30 |
| inc0 | so we can actually promote them for core status | 14:30 |
| sdake | no, there are two seprate acl groups in gerrit | 14:30 |
| inc0 | sdake, kolla-mesos Angus and Michal did | 14:30 |
| sdake | promotion is easy | 14:30 |
| sdake | angus got promoted | 14:30 |
| inc0 | yes, once he earned it | 14:30 |
| sdake | there was another dude in kolla-mesos-core that was never promoted | 14:30 |
| sdake | agree he earned it | 14:30 |
| inc0 | granted, for Angus we give him a bit of a slack, but because both you and me and few others knows him | 14:31 |
| sdake | i am promoting same model here | 14:31 |
| rhallisey | I don't think it's the same | 14:31 |
| inc0 | let's wait for vote to finish, so far split of repos wins | 14:31 |
| inc0 | but let's wait for other cores to chime in | 14:32 |
| inc0 | also I honestly think that having k8s in main repo will make people deploy it and potentially hurt | 14:33 |
| inc0 | and in case k8s fails, which it might (I hope not, but well), it would make it harder to remove | 14:33 |
| sdake | so far its equal in my count :) | 14:33 |
| rhallisey | I think it's catastrophic in the main repo. All so we don't potentially lose some git histy | 14:33 |
| *** Mr_Broken has quit IRC | 14:33 | |
| sdake | inc0 i raised that with a warning in the tool | 14:33 |
| sdake | that isn't the main reason | 14:33 |
| sdake | the main reason is to have one integrated solution | 14:34 |
| sdake | wtb kolla-mesos reviewers to chime in | 14:34 |
| rhallisey | ok, so what about other projects? | 14:34 |
| inc0 | now with nihilifer vote is 3v2 for separate repo | 14:34 |
| sdake | nihilifer can you weigh in plz | 14:34 |
| inc0 | few minutes ago was 3v1 | 14:34 |
| *** salv-orlando has joined #openstack-kolla | 14:34 | |
| rhallisey | we can't integrate every project into kolla | 14:34 |
| sdake | rhallisey agree, but we aren't talking about future proofing this | 14:35 |
| rhallisey | this is the first class argument all over again | 14:35 |
| sdake | this is for one case | 14:35 |
| sdake | yes i want k8s to be first class | 14:35 |
| inc0 | sdake, we all do..when it's ready | 14:35 |
| sdake | britt, sdake, nihilifer = 3 | 14:35 |
| rhallisey | what about tripleo/chef/puppet | 14:35 |
| inc0 | britt is not a core | 14:35 |
| rhallisey | they all get denoted to 2nd class | 14:35 |
| inc0 | also dims voted for separate repo if non-cores also vote | 14:36 |
| sdake | britt was on the list of people invited to vote because he was interested in participated | 14:36 |
| sdake | ok 4/3 then ;) | 14:36 |
| inc0 | so did dims | 14:36 |
| rhallisey | plus this vote is not a merge now and vote to split later | 14:36 |
| *** v1k0d3n has joined #openstack-kolla | 14:36 | |
| rhallisey | your talking about merging the tools | 14:36 |
| sdake | everyone on that list has a say | 14:36 |
| sdake | what we do later is a separate and orthogonal discussion | 14:36 |
| sdake | vikram iirc was in favor of same repo | 14:37 |
| sdake | 4/4 :) | 14:37 |
| sdake | equal | 14:37 |
| nihilifer | but what's the problem with having all the things (even tripleo/puppet/chef) in one repo? i see no real problem to solve. things like "it's better to keep ecosystems in separate repo" are no problems for me | 14:37 |
| Lyncos | python-memcache package is missing in Cinder and glance ubuntu containers..... | 14:37 |
| *** mag009_ has joined #openstack-kolla | 14:37 | |
| Lyncos | is this adressed somewhere ? | 14:37 |
| nihilifer | on the other hand, we indeed had technical problems when we separated kolla-mesos | 14:37 |
| sdake | Lyncos probably not, ubuntu binary needs some love - could you file a bug | 14:37 |
| mag009_ | mornin all | 14:37 |
| sdake | hey mag :) | 14:37 |
| sdake | welcome to the party | 14:37 |
| Lyncos | sdake will file a bug | 14:37 |
| rhallisey | nihilifer, that's not the problem. The problem is that all the things won't be willing to join the main repo | 14:37 |
| *** ravig has quit IRC | 14:38 | |
| sdake | Lyncos thanks | 14:38 |
| rhallisey | and since they arn't willing the get demoted | 14:38 |
| sdake | and ify ou want ot fix it we will take tha t too ;) | 14:38 |
| inc0 | nihilifer, idea is to merge repos when k8s becomes stable | 14:38 |
| inc0 | one idea at least | 14:38 |
| nihilifer | rhallisey: how we can know whether all the things will will to join the main repo if they don't exist yet? | 14:38 |
| inc0 | problem with one repo is that it will be hell to manage with first commits | 14:38 |
| inc0 | these have to be fast paced | 14:38 |
| sdake | inc0 little faith goes a long long way :) | 14:39 |
| nihilifer | inc0: yea, so in the first months we will begin development with copied files, copied code and with technical debt | 14:39 |
| sdake | ok its day 1 of th evote | 14:39 |
| nihilifer | and then "solve it" | 14:39 |
| sdake | lets not argue it into death | 14:39 |
| sdake | until we have wide community weighin | 14:39 |
| v1k0d3n | hey guys. morning. | 14:39 |
| sdake | nihilifer right on dude i agree | 14:39 |
| sdake | hey v1k0d3n | 14:39 |
| rhallisey | nihilifer, tripleo has a stable repo and community. Their case is to consume. Chef was a similar scenario | 14:39 |
| nihilifer | "later solving" rarely happens, just saying | 14:40 |
| sdake | tbh i dont know if chef will consume or not | 14:40 |
| inc0 | sdake, it might be my polish untrusting soul of an eastern european, but I want people to earn our trust before deciding about future of our project | 14:40 |
| sdake | its unclear what tripleo's plan is :) | 14:40 |
| sdake | they decide the future of k8s directory ONLY | 14:40 |
| v1k0d3n | wanted to get back into kolla, and check it out. seems like i keep running into issues. does anyone have a few minutes to direct message for a bit, so i can make sure i'm reading through docs correcty? | 14:40 |
| sdake | and possibly tests directory | 14:40 |
| sdake | v1k0d3n i'm here for you bro if you throw me some vikoden :) | 14:40 |
| * sdake needs some anestaphine from the medlock | 14:41 | |
| v1k0d3n | rhallisey: maybe you would be best, since i'm joining the party since i'm going to get involved with the kolla-k8s portion. | 14:41 |
| v1k0d3n | lol sorry sdake i'm all out. :) | 14:41 |
| rhallisey | v1k0d3n, oh hey! | 14:41 |
| v1k0d3n | gotta love those aggressive gamertag handles. :) | 14:41 |
| sdake | v1k0d3n its a-ok i was just jking :) | 14:41 |
| rhallisey | v1k0d3n, sure | 14:42 |
| inc0 | v1k0d3n, issue is - we want to start working on kolla-k8s | 14:42 |
| Lyncos | sdake I submitted the bug :-) now have to figure out how to submit a patch | 14:42 |
| inc0 | but discussion is about whether we make dir in main kolla tree or separate kolla-k8s repo | 14:42 |
| sdake | Lyncos moment i have a link for you | 14:43 |
| Lyncos | Ok thanks | 14:43 |
| sdake | Lyncos http://docs.openstack.org/infra/manual/developers.html | 14:44 |
| sdake | need thorough review of https://review.openstack.org/#/c/311632/4 | 14:45 |
| patchbot | sdake: patch 311632 - kolla - Remove dependencies on kazoo and friends | 14:45 |
| v1k0d3n | inc0: agreed...me too, definitely. | 14:45 |
| inc0 | please, put a mail into openstack-dev then | 14:45 |
| inc0 | that's how we vote | 14:46 |
| sdake | v1k0d3n yo uhavce an action on the ml - all people that were interested were asked to vote whether they want split repo or unified repo | 14:46 |
| sdake | please read the entire thread before making your choice | 14:46 |
| sdake | iirc your name was on the list of poeple interested | 14:47 |
| sdake | btw, i got my office half assembled ;) | 14:48 |
| sdake | my lab mahcines are still not hooked up | 14:48 |
| sdake | which iw ill do today | 14:48 |
| sdake | then i'll have 10gig multinode rolling yay :) | 14:48 |
| sdake | rhallisey would you mind atleast voting +1 on the proposal to actually do kolla k8s support ;) | 14:49 |
| rhallisey | I'm +1. I wouldn't vote against my own spec | 14:50 |
| *** MarMat has joined #openstack-kolla | 14:50 | |
| *** ayoung has joined #openstack-kolla | 14:50 | |
| *** salv-orlando has quit IRC | 14:53 | |
| sdake | rhallisey cool can you put that on the ml ;) | 14:55 |
| sdake | because as it stands now you sort of look -1 ;) | 14:55 |
| sdake | just put the caveat in that you prefer a split repo but your +1 for the idea of k8s dev | 14:56 |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla: Addining ability to specify capabilities and security https://review.openstack.org/308447 | 14:56 |
| *** thumpba has quit IRC | 14:59 | |
| sdake | hey folks i just spoke with infra, we can have two groups, one with +2 ability but kolla-core would retain +w rights on the repo | 15:00 |
| *** dwalsh has quit IRC | 15:01 | |
| inc0 | sdake, but according to your proposal k8s will be *in* our repo | 15:01 |
| inc0 | so either there will be no k8s specific cores or kolla will have lot's of cores | 15:02 |
| sdake | no, there will be a kolla-k8s-core group | 15:04 |
| *** ravig has joined #openstack-kolla | 15:04 | |
| sdake | these people can +2 patches, but not workfow them is one option | 15:04 |
| sdake | or +2 and +w patches | 15:04 |
| sdake | remember git revert is an easy solution to bad actors :) | 15:05 |
| *** mlima has quit IRC | 15:05 | |
| inc0 | not so easy if there would be lots of reverts | 15:05 |
| sdake | dont let paranoia rule :) | 15:05 |
| sdake | we will have training session :) | 15:05 |
| *** daneyon has joined #openstack-kolla | 15:06 | |
| sdake | the core team knows who the core team is | 15:06 |
| sdake | anyone know which tz mlima is in | 15:06 |
| *** |GonZo2000| has joined #openstack-kolla | 15:10 | |
| *** daneyon has quit IRC | 15:10 | |
| *** severion has joined #openstack-kolla | 15:12 | |
| *** thumpba has joined #openstack-kolla | 15:12 | |
| nihilifer | idea of having only +2 sounds good for kolla-k8s-core | 15:13 |
| *** dwalsh has joined #openstack-kolla | 15:14 | |
| *** diogogmt has joined #openstack-kolla | 15:15 | |
| *** cfarquhar has joined #openstack-kolla | 15:18 | |
| *** cfarquhar has joined #openstack-kolla | 15:18 | |
| *** harmw_ has joined #openstack-kolla | 15:18 | |
| ccesario | afternoon guys :) | 15:19 |
| ccesario | sdake, GMT -03:00 or -02:00 | 15:20 |
| *** mlima has joined #openstack-kolla | 15:20 | |
| *** betheynyx has joined #openstack-kolla | 15:21 | |
| *** betheynyx_ has quit IRC | 15:21 | |
| *** GonZo2000 has quit IRC | 15:21 | |
| *** harmw has quit IRC | 15:21 | |
| *** v1k0d3n has quit IRC | 15:21 | |
| *** harmw has joined #openstack-kolla | 15:21 | |
| *** cfarquhar_ has quit IRC | 15:21 | |
| *** jrist has quit IRC | 15:21 | |
| *** harmw has quit IRC | 15:21 | |
| *** ravig has quit IRC | 15:21 | |
| mlima | sdake, ? | 15:23 |
| openstackgerrit | Travis McPeak proposed openstack/kolla: Adding a few #nosec tags to prepare for Bandit usage https://review.openstack.org/310869 | 15:31 |
| openstackgerrit | Stephane Boisvert proposed openstack/kolla: Closes-Bug: 1577422 https://review.openstack.org/311765 | 15:33 |
| openstack | bug 1577422 in kolla "python-memcache is missing from cinder and glance containers" [Undecided,New] https://launchpad.net/bugs/1577422 | 15:33 |
| *** fragatina has joined #openstack-kolla | 15:34 | |
| *** ravig has joined #openstack-kolla | 15:36 | |
| *** salv-orlando has joined #openstack-kolla | 15:36 | |
| timss | maybe I misunderstood something, but regarding preserving git history, maybe this is of interest? http://stackoverflow.com/questions/1365541/how-to-move-files-from-one-git-repo-to-another-not-a-clone-preserving-history/11426261#11426261 | 15:37 |
| *** jrist has joined #openstack-kolla | 15:37 | |
| timss | it can be pretty cool (unless there are lots of conflicts) | 15:38 |
| *** thumpba_ has joined #openstack-kolla | 15:40 | |
| inc0 | timss, no conflicts ever as it will be separate directory | 15:40 |
| timss | sure, I was just talking in general for that approach | 15:41 |
| rhallisey | timss, neat | 15:41 |
| Lyncos | sdake: I did submit my patch... can you just please confirm that I did it the right way ? | 15:41 |
| sdake | Lyncos link? | 15:41 |
| timss | https://github.com/vimperator/vimperator.vim/commits/master | 15:41 |
| timss | example where I used it | 15:41 |
| Lyncos | https://review.openstack.org/311765 | 15:41 |
| *** fragatina has quit IRC | 15:41 | |
| *** thumpba has quit IRC | 15:41 | |
| inc0 | Lyncos, you need commit message;) | 15:42 |
| Lyncos | damn | 15:42 |
| inc0 | a title and some short desc | 15:42 |
| *** fragatina has joined #openstack-kolla | 15:42 | |
| inc0 | also tag is Closes-Bug: #xxx | 15:42 |
| inc0 | you forgot # | 15:42 |
| Lyncos | ok I'll try to figure out how to change my commit :-) | 15:42 |
| inc0 | this will make your bug in launchpad turn into inprog automagically | 15:42 |
| inc0 | Lyncos, git commit --amend | 15:42 |
| inc0 | if we'd ask you to change something in files, git commit --amend as well | 15:43 |
| inc0 | git add files_changed, git commit --amend | 15:43 |
| sdake | Lyncos your commit message is wrong but otherwise looks good :) | 15:43 |
| Lyncos | Ok fixing this | 15:44 |
| sdake | Lyncos serguei has a question int he bug tracker | 15:44 |
| sdake | i have same question | 15:44 |
| sdake | https://bugs.launchpad.net/kolla/+bug/1577422 | 15:44 |
| openstack | Launchpad bug 1577422 in kolla "python-memcache is missing from cinder and glance containers" [Undecided,New] | 15:44 |
| sdake | Lyncos what is your launchpad id plz | 15:45 |
| Lyncos | Launchpad Id: | 15:45 |
| Lyncos | stephane-boisvert | 15:45 |
| *** haplo37 has joined #openstack-kolla | 15:47 | |
| Lyncos | I think my comment is better now | 15:50 |
| Lyncos | also answered your questions | 15:50 |
| *** rstarmer has joined #openstack-kolla | 15:51 | |
| *** mbound has quit IRC | 15:52 | |
| *** rstarmer has quit IRC | 15:55 | |
| *** rstarmer has joined #openstack-kolla | 15:56 | |
| dims | sdake : inc0 : rhallisey : multiple groups on same repo gets too complicated (esp so many new folks!). i assumed that eventually we'll have a base repo (may be even publish code to pypi) and ansible/ kubernetes/ would consume that deliverable and do things specific to their environments. | 15:56 |
| *** ozialien10 has joined #openstack-kolla | 15:58 | |
| jpeeler | so basically, you're saying a separate repo to maintain access permissions makes sense? | 15:58 |
| *** dmk0202 has quit IRC | 15:59 | |
| dims | jpeeler : my opinion. yes | 15:59 |
| *** JoseMello has quit IRC | 15:59 | |
| sdake | so acls are one thing | 15:59 |
| sdake | but we can enforce acl's via a social contract | 15:59 |
| sdake | i am thinking back to kolla-mesos and some of the reasons that failed | 15:59 |
| sdake | and it had alot to do with nihilfer's email | 15:59 |
| dims | jpeeler : though i think that once we hit the deadline that sdake mentioned (july 15?) then we should have one single group | 16:00 |
| sdake | there were basically two camps | 16:00 |
| sdake | deadline pulled out of ass btw :) | 16:00 |
| dims | :) | 16:00 |
| sdake | just a proposal to get poeple thinking | 16:00 |
| sdake | mesos folsk reviewed kolal repo | 16:00 |
| sdake | kolla core didn't review mesos code base | 16:01 |
| *** rstarmer has quit IRC | 16:01 | |
| sdake | which i found disturbing and not what i thought would make the projects successful | 16:01 |
| sdake | Lyncos your patch doesn't need to do that work for centos | 16:01 |
| sdake | its an ubuntu binary only problem | 16:01 |
| *** unicell has quit IRC | 16:01 | |
| Lyncos | yeah I just figured it out | 16:01 |
| Lyncos | I guess I'll amend again | 16:01 |
| *** ssurana has joined #openstack-kolla | 16:03 | |
| dims | sdake : true. we have that problem in say nova + nova-vmware/hyperv driver etc as well. one repo does not help | 16:03 |
| sdake | if we fail a 3rd time on k8s + kolla i'm afraid people will lose confidence in our ability to execute | 16:03 |
| sdake | ya but nova is bigger scale | 16:03 |
| dims | fair point sdake | 16:03 |
| Lyncos | Sorry for my noobness but do I have to 'push' when I ammend ? | 16:03 |
| sdake | we are talking two coe's here | 16:03 |
| sdake | Lyncos type git review | 16:04 |
| sdake | and gerrit takes care of the git push stuff | 16:04 |
| rhallisey | sdake, disagree | 16:04 |
| sdake | Lyncos we are here to train :) | 16:04 |
| dims | sdake : am perfectly willing to go with majority opinion :) | 16:04 |
| openstackgerrit | Stephane Boisvert proposed openstack/kolla: Add python-memcache package on cinder / glance Adding python-memcache package to cinder-base and glance ubuntu docker images https://review.openstack.org/311765 | 16:04 |
| rhallisey | sdake, what if kolla-ansible and kolla-mesos were sep repos | 16:04 |
| rhallisey | that was the real problem | 16:04 |
| Lyncos | sdake thanks for your comprehension we used to be a SVN shop :-) | 16:05 |
| dims | rhallisey : right, that was my long term view | 16:05 |
| sdake | rhallisey that is a different way of looing at the problem indeed | 16:05 |
| Lyncos | sdake: I think I fixed my commit It looks better now | 16:05 |
| rhallisey | the repos should've been split to allow for kolla-mesos to have a fair shot at it | 16:05 |
| *** ssurana has quit IRC | 16:05 | |
| sdake | should have but capacity problems limited our ability to execute that | 16:05 |
| sdake | we have the same capacity problems on splitting the repo now | 16:05 |
| sdake | i tink that would have slowed down our ansible implementation tremendously as well | 16:06 |
| openstackgerrit | Stephane Boisvert proposed openstack/kolla: Add python-memcache package on cinder / glance Adding python-memcache package to cinder-base and glance ubuntu docker images https://review.openstack.org/311765 | 16:06 |
| rhallisey | sdake, ^ you nailed it right there | 16:06 |
| rhallisey | as to why mesos didn't work | 16:06 |
| rhallisey | because anisble was a first class citizen | 16:06 |
| dims | and is still going to be first class whether we do one repo or two | 16:07 |
| rhallisey | dims, correct, but ansible is far more developed now | 16:07 |
| dims | rhallisey : ah ok | 16:08 |
| rhallisey | dims, plus the k8s community is a lot larger | 16:08 |
| *** salv-orlando has quit IRC | 16:08 | |
| *** ssurana has joined #openstack-kolla | 16:08 | |
| rhallisey | like 10x larger | 16:08 |
| sdake | nailed which | 16:08 |
| sdake | that splitting the repos would have slowed down commits? | 16:08 |
| sdake | instead of committing to one repo we would have commit to two | 16:09 |
| sdake | our gating is all based aroudn one repo and not two | 16:09 |
| rhallisey | sdake, it would've slowed down ansible commits because we wanted to develop ansible | 16:09 |
| sdake | it takes 2 months to split the repos and get the ci going again | 16:09 |
| sdake | i dont want to wait | 16:09 |
| sdake | it would have slowed down the docker file dev as well | 16:09 |
| sdake | backports of a fully split repo as mentioend before is a manual process and irrevisible | 16:10 |
| dims | sdake : so another option...we have a feature branch and provide ACL access to k8s group to the branch only | 16:10 |
| dims | sdake : and have sync every week from master to branch or when necessary | 16:11 |
| openstackgerrit | Stephane Boisvert proposed openstack/kolla: Add python-memcache package on cinder / glance https://review.openstack.org/311765 | 16:11 |
| dims | we could even do two way sync's | 16:11 |
| sdake | dims how would we sync? | 16:11 |
| sdake | (I am a git noob :) | 16:11 |
| dims | sdake : there's a way to do it. we used it for pika driver and zmq driver in oslo.messaging | 16:12 |
| dims | let me dig up details | 16:12 |
| jpeeler | dims: but then you'd have to audit the merges from k8s, right? | 16:12 |
| rhallisey | dims, at that point why not just make a new repo? | 16:12 |
| dims | rhallisey : going from 2 repos to one is difficult. merging feature branch is easy | 16:12 |
| dims | jpeeler : if we decide k8s is viable then we audit+merge | 16:13 |
| dims | sdake : http://docs.openstack.org/infra/manual/drivers.html | 16:14 |
| rhallisey | how about merging two communities into one? I'd say that's harder than merging two repos. | 16:14 |
| sdake | that is why we have one community | 16:15 |
| dims | rhallisey : right, we have a deadline say (newton-2) at which point we decide to pull feature branch into master and make it as one ACL | 16:15 |
| sdake | dims what type of rocket science is required to commit to a feature branch? | 16:15 |
| dims | until then folks on branch can pull stuff from master on demand and we have one repo to review for all | 16:15 |
| sdake | (is it hard) | 16:15 |
| dims | sdake : nope. just an entry in .gitreview | 16:16 |
| sdake | one repo one community - sounds good | 16:16 |
| rhallisey | except there won't be one community.. | 16:16 |
| dims | example https://github.com/openstack/oslo.messaging/blob/feature/pika/.gitreview#L5 | 16:16 |
| rhallisey | even in one repo.. | 16:16 |
| sdake | dims i'd have tot trust you have read through everyone's concerns and understand them all and this meets the requirements people have set out | 16:16 |
| sdake | we promote people that are active in kolla-k8s-core to kolla-core once htey have earned it | 16:17 |
| sdake | everyone else gets removed from +2/+W | 16:17 |
| dims | sdake : that won't change mental model of people who are currently ansible only to review k8s code | 16:17 |
| dims | sdake : right | 16:17 |
| sdake | dims, they might take a peek :) | 16:18 |
| sdake | if its in one gerrit review queue | 16:18 |
| dims | sdake : u surely hope so :) | 16:18 |
| dims | s/u/i/ | 16:18 |
| sdake | that is what i am after | 16:18 |
| dims | it's just like people looking at stable/ branches | 16:18 |
| dims | we can setup jobs that run on feature branches too | 16:19 |
| *** sacharya has joined #openstack-kolla | 16:21 | |
| *** haplo37 has quit IRC | 16:24 | |
| dims | one more. for the things nihilifer mentioned, the idea would be to first submit patches to master and then pull them into feature branch | 16:25 |
| dims | so that will take care of making sure folks are working on master mostly and doing only minimal things in the branch | 16:26 |
| *** ravig has joined #openstack-kolla | 16:29 | |
| *** banix has quit IRC | 16:30 | |
| inc0 | dims, but that's kina non-standard way in OS | 16:31 |
| inc0 | possible, non standard | 16:31 |
| *** haplo37 has joined #openstack-kolla | 16:31 | |
| inc0 | feature branches aren't something gerrit handles in openstack | 16:31 |
| inc0 | we don't want to be infra's own snowflake | 16:31 |
| dims | inc0 : we used for pika and zmq drivers without much problems. its up to you all :) the idea is that we are trying to bootstrap both people and code in a "safe" space and then have a good way to either merge code into master or split into a separate repo. i pointed it out as feature branches are fully supported currently by infra folks AFAIK | 16:35 |
| inc0 | ok, fair enough...who will have +2+w power on feature branch? | 16:36 |
| *** severion has quit IRC | 16:36 | |
| inc0 | can we specify separate group for that? | 16:36 |
| openstackgerrit | Merged openstack/kolla: Remove dependencies on kazoo and friends https://review.openstack.org/311632 | 16:36 |
| dims | inc0 : yes, we can specify a group for feature branch | 16:36 |
| inc0 | ok...so essentially it's lightweight separate repo | 16:37 |
| dims | right | 16:37 |
| inc0 | not horrible solution | 16:37 |
| *** v1k0d3n has joined #openstack-kolla | 16:38 | |
| dims | let me confirm the ACL question with some folks | 16:38 |
| sdake | rhallisey this lightweight second repo - thoughts? | 16:43 |
| dims | got confirmation from infra that a separate group for feature branch is possible | 16:44 |
| inc0 | cons of this is it will in fact lower exposure of k8s to people | 16:45 |
| inc0 | if you don't know what to look for, it's unlikely you'll find it | 16:45 |
| sdake | inc0 not if we weekly periodically merge | 16:46 |
| sdake | but i'm all about simplicity nad one repo gives that :) | 16:46 |
| inc0 | sdake, if we merge periodically we might as well skip feature branch as well | 16:47 |
| inc0 | as code people will merge in branch will end up in kolla master as well | 16:47 |
| sdake | inc0 that was dims idea :) | 16:47 |
| inc0 | that would solve lack of trust I have tho;) | 16:48 |
| sdake | trust until given reason not to ;) | 16:48 |
| dims | haha | 16:48 |
| dims | inc0 : so i wanted current core to bless sync from branch to master. but folks on k8s group could pull from master whenever they want to | 16:49 |
| inc0 | dims that makes sense | 16:49 |
| inc0 | but I still think this won't fix problems...I want people to feel in control of their code | 16:50 |
| inc0 | I don't want to have core team and half core team | 16:50 |
| inc0 | (I would be pissed if I'd be a half-core;)) | 16:50 |
| dims | inc0 : they have complete freedom in the branch :) | 16:50 |
| inc0 | better give them their own repo and powers of +2 in it's full glory | 16:50 |
| dims | it's all a tradeoff :) | 16:50 |
| dims | all 3 options are valid in one way or another | 16:50 |
| rhallisey | idk if I really like the branch. It will cause merging of the CLI + ansible config | 16:51 |
| rhallisey | which means down the road is a 0% of a split if the cores want it | 16:52 |
| rhallisey | I see that split gives is the most flexibility | 16:52 |
| rhallisey | and if we merge we lose some git history, even though there will not have been a release yet | 16:52 |
| rhallisey | http://stackoverflow.com/questions/1365541/how-to-move-files-from-one-git-repo-to-another-not-a-clone-preserving-history/11426261#11426261 | 16:53 |
| jpeeler | i'm pretty sure the git history will work either way. two repos -> one, one -> two repos | 16:53 |
| rhallisey | earlier from timss | 16:53 |
| inc0 | rhallisey, I don't think we will | 16:53 |
| inc0 | there are ways | 16:54 |
| sdake | it took me two ways to get one history for stable/liberty | 16:54 |
| sdake | just keep that metric in mind | 16:54 |
| sdake | ways/weeks | 16:54 |
| inc0 | sdake, we revmaped whole branch | 16:55 |
| inc0 | this is easier | 16:55 |
| rhallisey | so then what's the argument against here? k8s is 10x the size of mesos & ansible is stable | 16:55 |
| rhallisey | the mesos arguments from earlier don't apply here | 16:55 |
| inc0 | and for git magic, liberty was easy too | 16:55 |
| inc0 | it was politics that took week | 16:55 |
| inc0 | s | 16:55 |
| sdake | the k8s code base will be small | 16:56 |
| sdake | probably 3-4kloc | 16:56 |
| *** ssurana has quit IRC | 16:58 | |
| inc0 | sdake, I don't think so | 16:59 |
| inc0 | we are creating new models of HA | 16:59 |
| inc0 | unlike whatever we have now | 16:59 |
| sdake | k8s in the past was just a bunch of yaml | 16:59 |
| *** ssurana has joined #openstack-kolla | 17:00 | |
| inc0 | if we do PoC quality | 17:00 |
| *** stvnoyes has joined #openstack-kolla | 17:00 | |
| inc0 | if you want to include fencing of mariadb, ceph stuff | 17:00 |
| inc0 | it'll grow | 17:00 |
| sdake | we dont have fencing now ;) | 17:00 |
| inc0 | galera does | 17:01 |
| inc0 | we're not running galera on k8s | 17:01 |
| *** gfidente has quit IRC | 17:01 | |
| sdake | right | 17:01 |
| *** banix has joined #openstack-kolla | 17:02 | |
| sdake | rhallisey can you send out a link to your specification and get people reviewing and commenting on it plz | 17:02 |
| sdake | we need to get that finished so people know what the implementation willl look like | 17:02 |
| inc0 | so we need to solve problems, that'll take time, work and some clever scripts I bet | 17:02 |
| rhallisey | sdake, I'm adding a name to the list one moment | 17:03 |
| inc0 | in fact, I think my first contribution to kolla-k8s will be a spec with HA model | 17:03 |
| inc0 | we need to figure out this one before we start implementing hard stuff | 17:03 |
| sdake | can we just have one spec | 17:03 |
| sdake | plz | 17:03 |
| rhallisey | inc0, I mentioned your comments about fencing in the spec | 17:03 |
| *** unicell has joined #openstack-kolla | 17:04 | |
| rhallisey | inc0, I can co author you, but I don't know if you can vote on it | 17:04 |
| rhallisey | if I do that | 17:04 |
| *** rajathagasthya has joined #openstack-kolla | 17:04 | |
| sdake | inc0 can you ack https://review.openstack.org/#/c/311570/ | 17:04 |
| patchbot | sdake: patch 311570 - kolla-mesos - Retire the kolla-mesos repository | 17:04 |
| inc0 | rhallisey, let's just get started and have this discussion later | 17:04 |
| rhallisey | roger | 17:05 |
| inc0 | we have basics figured out, it's crazy netsplits that we need to do | 17:05 |
| rhallisey | https://review.openstack.org/#/c/304182/4 | 17:05 |
| patchbot | rhallisey: patch 304182 - kolla - Spec: Deploy Kolla images using Kubernetes | 17:05 |
| rhallisey | ^ spec | 17:06 |
| rhallisey | that's good for another round of comments | 17:06 |
| openstackgerrit | Merged openstack/kolla-mesos: Retire the kolla-mesos repository https://review.openstack.org/311570 | 17:06 |
| *** fragatina has quit IRC | 17:08 | |
| sdake | dims quick question https://github.com/openstack/kolla-mesos | 17:09 |
| sdake | i missed the dot files | 17:09 |
| sdake | i am going to submit a new review to remove them | 17:09 |
| sdake | should I change the readme to say git checkout HEAD^2 | 17:10 |
| dims | sdake : sure. that or the SHA either one should be ok | 17:11 |
| openstackgerrit | Steven Dake proposed openstack/kolla-mesos: Final removal of files beginning with dot https://review.openstack.org/311791 | 17:13 |
| sdake | inc0 jpeeler rhallisey can you guys rubber stamp that one plz :) | 17:13 |
| sdake | one thing about seaprate repos, reitrement is easier ;) | 17:14 |
| openstackgerrit | Merged openstack/kolla-mesos: Final removal of files beginning with dot https://review.openstack.org/311791 | 17:14 |
| sdake | thanks ;) | 17:14 |
| dims | lol | 17:15 |
| dims | dang i missed a easy review to pad my stats | 17:15 |
| openstackgerrit | Steven Dake proposed openstack/kolla-mesos: Fix typos in readme.rst https://review.openstack.org/311792 | 17:16 |
| sdake | one more | 17:16 |
| jpeeler | the readme says where the kubernetes dev will take place | 17:16 |
| * sdake groans | 17:16 | |
| dims | lol | 17:16 |
| jpeeler | which hasn't been decided. and there were several misspellings in the readme :( | 17:16 |
| sdake | jpeeler i am going to leave the acls in palce | 17:16 |
| sdake | until that is decided | 17:16 |
| sdake | and make whatever we decide | 17:16 |
| sdake | in the readme | 17:16 |
| sdake | then remove the acls as is documented in the retirement document | 17:16 |
| sdake | fair enough? | 17:17 |
| sdake | see https://review.openstack.org/#/c/311574/ | 17:17 |
| patchbot | sdake: patch 311574 - openstack-infra/project-config - Part two of kolla-mesos retirement | 17:17 |
| jpeeler | what editor do you use sdake ? | 17:17 |
| sdake | notice worklow -1 | 17:17 |
| sdake | jpeeler vi | 17:17 |
| jpeeler | i wonder if there's a way to add spell check to that | 17:18 |
| sdake | it would be helpful for rst files only | 17:18 |
| sdake | if you find a way let me know | 17:18 |
| sdake | that way I dont keep having typos in my commit messages :) | 17:18 |
| *** ccesario___ has quit IRC | 17:19 | |
| rhallisey | dims, lol | 17:19 |
| sdake | i did that retirement on a plane :) | 17:19 |
| *** ccesario___ has joined #openstack-kolla | 17:19 | |
| sdake | wit hlots of liquor | 17:19 |
| sdake | dims i dont think your stats need paddding :) | 17:21 |
| sdake | last cycle I was top 1% of all commits for openstack for the lifetime of the project | 17:21 |
| dims | sdake : lol | 17:21 |
| sdake | this cycle top 1.5% :( | 17:21 |
| sdake | my typos are actually to pad stats | 17:22 |
| sdake | to get into the top 10 :) | 17:22 |
| * dims starts a new email thread "Gaming stackalytics - How to do it right" :) | 17:22 | |
| sdake | hey, atleast i am not changing "openstack" to "OpenStack" across all the projects | 17:22 |
| jpeeler | haha | 17:22 |
| * sdake shakes head at that dude | 17:23 | |
| jpeeler | honestly though if people want to do project wide spelling fixes, go for it. at least it wasn't a one project drive by commit to get an ATC | 17:24 |
| *** ccesario___ has quit IRC | 17:30 | |
| sdake | jpeeler agree - although even drive by atc commits typically improve the code or I -1 em :) | 17:32 |
| sdake | why is my heater swet to 80 in summertime | 17:32 |
| dims | related to prev discussion but another data point... if we say ansible is advanced, how long would it take to split that into a separate repo? | 17:32 |
| sdake | i dont get this whole nest thing | 17:32 |
| sdake | it just doesn't work | 17:32 |
| sdake | dims i think 4-6 weeks | 17:32 |
| *** fragatina has joined #openstack-kolla | 17:32 | |
| sdake | to split both efforts from repo 6-8 weeks | 17:33 |
| rhallisey | and that's a code freeze | 17:33 |
| *** fragatina has joined #openstack-kolla | 17:33 | |
| sdake | ya its a cvode freeze | 17:33 |
| sdake | it locks up the repo for some time | 17:33 |
| dims | rhallisey sdake : mainly CI? | 17:33 |
| sdake | dims ci integration project config etc | 17:34 |
| timss | I actually noticed a couple of typos in the doc when reading through it earler. Is a commit with only 1-2 character fixes ok? | 17:34 |
| rhallisey | timss, ya | 17:34 |
| sdake | timss if there is an error fix it | 17:34 |
| sdake | doc fixes dont' need bug ids | 17:34 |
| sdake | we dont backport docs because afaik they can't be versioned | 17:34 |
| rhallisey | say we were merged and we want to split 6-8 weeks of a code freeze on an entire project O.o | 17:34 |
| sdake | the reason we require bug ids is to deal with backports | 17:34 |
| timss | all right, I'll try to take a look at it later, no experience with gerrit or openstack development in general other than lurking | 17:35 |
| sdake | timss √ | 17:35 |
| sdake | https://wiki.openstack.org/wiki/How_To_Contribute | 17:35 |
| dims | rhallisey : so no way to start a new repo, get CI to work on new repo while continuing work to proceed in the old repo in parallel until we cut over | 17:36 |
| sdake | dims interesting idea - hadn't thought of that | 17:36 |
| sdake | it would require a repo rename | 17:36 |
| sdake | which fungi says infra doesn't do anymore | 17:36 |
| sdake | but they might cut us some slack on that special case | 17:36 |
| rhallisey | hm | 17:37 |
| timss | as for spell checking in vi; it's builtin, as well as dictionary functionality if you use vim. There are plugins which make it easier to use though, such as vim-lexical: https://github.com/reedes/vim-lexical | 17:37 |
| dims | y, we can ask for exception | 17:37 |
| sdake | right dims | 17:37 |
| rhallisey | dims, does that mean were always playing catchup though? | 17:37 |
| timss | examples for only using with certian filetypes are included in its docuemntation | 17:37 |
| rhallisey | say only 2 people work on the split | 17:37 |
| timss | sdake: thanks for the link | 17:37 |
| dims | rhallisey : once the CI is up on the new repo, we cut over immediately and copy over stuff we need from the other repo | 17:38 |
| dims | that's all the options i can think of... head hurts now :) | 17:38 |
| sdake | ya my brain was melted by thursday of alst week | 17:38 |
| sdake | face melted by friday :) | 17:38 |
| sdake | next time we only doing a half day contributor meetup ;) | 17:38 |
| *** inc0 has quit IRC | 17:39 | |
| sdake | midcycle, ansible offices or wesford? | 17:39 |
| *** ccesario___ has joined #openstack-kolla | 17:39 | |
| *** Serlex has quit IRC | 17:39 | |
| *** inc0 has joined #openstack-kolla | 17:39 | |
| sdake | rdu woudl be good too at rht facilities | 17:39 |
| rhallisey | :) | 17:39 |
| * sdake not a fan of westford | 17:39 | |
| rhallisey | sdake, dude come to westford! | 17:40 |
| * dims starts his car | 17:40 | |
| rhallisey | ya dims is nearby! | 17:40 |
| sdake | the red aht offices there are beutiful | 17:40 |
| sdake | really modern ad great grass | 17:40 |
| sdake | (to walk on, not smoke :) | 17:40 |
| inc0 | grass...I've heard that it is a thing...not in Texas tho | 17:40 |
| sdake | i have grass in my backyard | 17:41 |
| sdake | and weeds | 17:41 |
| dims | if it was winter we could go snow tubing next door...right rhallisey ? | 17:41 |
| rhallisey | dims, it' | 17:41 |
| rhallisey | it's not as much fun as you think | 17:41 |
| rhallisey | :) | 17:41 |
| rhallisey | it's decent | 17:41 |
| dims | rhallisey : y, brought the kids ones there. long lines :) | 17:43 |
| rhallisey | it's just hard to argue it's a mountain | 17:44 |
| dims | :) | 17:44 |
| rhallisey | :D | 17:44 |
| openstackgerrit | Travis McPeak proposed openstack/kolla: Adding a few #nosec tags to prepare for Bandit usage https://review.openstack.org/310869 | 17:46 |
| *** thumpba_ has quit IRC | 17:47 | |
| *** daneyon has joined #openstack-kolla | 17:48 | |
| *** thumpba has joined #openstack-kolla | 17:51 | |
| *** inc0_ has joined #openstack-kolla | 17:52 | |
| *** inc0 has quit IRC | 17:52 | |
| *** daneyon has quit IRC | 17:53 | |
| openstackgerrit | Travis McPeak proposed openstack/kolla: Adding a few #nosec tags to prepare for Bandit usage https://review.openstack.org/310869 | 17:53 |
| *** ccesario___ has quit IRC | 17:54 | |
| openstackgerrit | Travis McPeak proposed openstack/kolla: Adding a few #nosec tags to prepare for Bandit usage https://review.openstack.org/310869 | 18:04 |
| *** fragatina has quit IRC | 18:13 | |
| *** fragatina has joined #openstack-kolla | 18:14 | |
| *** fragatina has quit IRC | 18:19 | |
| *** Mr_Broken has joined #openstack-kolla | 18:22 | |
| *** Mr_Broken has quit IRC | 18:26 | |
| *** inc0_ has quit IRC | 18:29 | |
| *** mkoderer__ has quit IRC | 18:32 | |
| *** ravig has quit IRC | 18:39 | |
| *** fragatina has joined #openstack-kolla | 18:40 | |
| *** ravig has joined #openstack-kolla | 18:41 | |
| *** ravig has quit IRC | 18:43 | |
| ccesario | Lyncos, I just try simulate the problem reported... and you are right.... !! the memcache is missing in ubuntu binary | 18:44 |
| Lyncos | ccesario... I'm also testing neutron.. it seems that it also need it | 18:45 |
| Lyncos | (not part of initial bug report) | 18:45 |
| ccesario | Lyncos, cool! | 18:45 |
| Lyncos | now I'm trying to figure out how to refresh my images .. as it seems to always pull the old one | 18:46 |
| *** rajathagasthya has quit IRC | 18:49 | |
| *** banix has quit IRC | 18:51 | |
| *** banix has joined #openstack-kolla | 18:51 | |
| *** inc0 has joined #openstack-kolla | 18:52 | |
| inc0 | sdake, | 19:13 |
| inc0 | we need to merge liberty patch | 19:13 |
| inc0 | to stable/liberty | 19:13 |
| inc0 | we don't have stable/liberty now | 19:13 |
| *** ravig has joined #openstack-kolla | 19:18 | |
| sdake | inc0 justwoek up from nap | 19:18 |
| sdake | did people test it? | 19:18 |
| sdake | i am ok with merging butnot ok with testing unless there is enough testing | 19:18 |
| *** ravig has quit IRC | 19:18 | |
| inc0 | people, did you test it? | 19:20 |
| inc0 | if you didn't, then do test it | 19:20 |
| sdake | iu tested centos source works | 19:21 |
| sdake | i tested centos binary fails | 19:21 |
| sdake | i dont know why centos binary doesn'twork | 19:21 |
| *** ravig has joined #openstack-kolla | 19:21 | |
| sdake | i also tested thatsome services have 127.0.0.1 as their ip address which is wrong | 19:21 |
| inc0 | sdake, there is nothing in liberty that would make this happen, not that I know of in any case | 19:22 |
| inc0 | centos binary - no clue, I don't speak centos | 19:22 |
| sdake | it has to do with how auth is done | 19:23 |
| sdake | it was fixed in neutron 7.0.3 | 19:23 |
| sdake | centos binary hsa 7.0.3 | 19:23 |
| sdake | yet centos binary doesn't appear to have the correct change | 19:23 |
| sdake | or alternately iwas testing the wrong stuff :) | 19:23 |
| sdake | i was akind of tired | 19:23 |
| sdake | i can get back to it today | 19:23 |
| sdake | the 127.0.0.1 was around [vnc] in nova | 19:24 |
| sdake | which changed | 19:24 |
| sdake | in liberty to mitaka | 19:24 |
| inc0 | if it's in nova, we can't really do much about it | 19:25 |
| inc0 | guys please test it out | 19:25 |
| inc0 | it's important | 19:25 |
| mlima | inc0, tell me how can i help | 19:25 |
| mlima | i can test ubuntu source here :) | 19:26 |
| inc0 | mlima, https://review.openstack.org/#/c/308390/ | 19:26 |
| patchbot | inc0: patch 308390 - kolla (stable/liberty) - WIP: Changes needed to deploy liberty with mitaka ... | 19:26 |
| sdake | if someone could test someo f hte newer services too that would rock | 19:26 |
| sdake | becausei only tested comptue kit | 19:26 |
| sdake | but i am feeding myself atm because i barely ate last week | 19:27 |
| sdake | except blowout dinners :) | 19:27 |
| mlima | multinode? single node? | 19:27 |
| sdake | multinode woudl be best | 19:28 |
| inc0 | all needs working | 19:28 |
| inc0 | needs to work* | 19:28 |
| sdake | i need to get my lab put tgoether today before i can begin testing | 19:29 |
| sdake | my office reorgdisconnected dmy server i test on | 19:29 |
| mlima | what is the deadline? | 19:29 |
| *** ravig has quit IRC | 19:29 | |
| *** ravig has joined #openstack-kolla | 19:29 | |
| mlima | the test is build, deploy and launch some instance? | 19:30 |
| sdake | yup | 19:30 |
| sdake | e need the new services tested too | 19:30 |
| mlima | just it? | 19:30 |
| *** absubram has joined #openstack-kolla | 19:30 | |
| mlima | new services? | 19:30 |
| sdake | also the versions used in the tarball references are out of date | 19:30 |
| sdake | that needs fixing | 19:30 |
| sdake | i fixed it for nova and neutron | 19:31 |
| sdake | but then had to get on a plane without snakes | 19:31 |
| *** MarMat_ has joined #openstack-kolla | 19:31 | |
| mlima | ok | 19:31 |
| mlima | what is the deadline? sdake | 19:32 |
| inc0 | sdakes on the plane | 19:32 |
| inc0 | mlima, it would be cool to make it work before summit | 19:32 |
| *** huikang has joined #openstack-kolla | 19:32 | |
| inc0 | in Austin | 19:33 |
| sdake | inc0 what i'm getting at is if you could fix some of the problems i just outlined that would be great - while I get my lab reassembled | 19:33 |
| inc0 | will do | 19:33 |
| mlima | I need to set up my multinode environment, but i can test single node | 19:34 |
| mlima | :/ | 19:34 |
| mlima | and tomorrow i test multinode | 19:34 |
| *** MarMat has quit IRC | 19:34 | |
| ccesario | Lyncos, ping ? | 19:38 |
| Lyncos | yep | 19:39 |
| ccesario | Lyncos, neutron needs memcache too | 19:39 |
| ccesario | http://paste.openstack.org/show/495908/ | 19:39 |
| Lyncos | want me to add it up to my original bug report ? | 19:39 |
| ccesario | could you please update your ps? | 19:40 |
| Lyncos | ps ? | 19:40 |
| inc0 | patchset Lyncos | 19:40 |
| ccesario | patch set | 19:40 |
| ccesario | https://review.openstack.org/#/c/311765/ | 19:41 |
| patchbot | ccesario: patch 311765 - kolla - Add python-memcache package on cinder / glance | 19:41 |
| Lyncos | yes sure | 19:41 |
| Lyncos | just give me some time so I can test everything ... trying to setup a better test environmnet | 19:41 |
| Lyncos | I'm also trying to figure out why memcache is not added to glance even if i modified my docker file | 19:42 |
| ccesario | Ok, i 'll update the launchpad bug id | 19:42 |
| ccesario | Lyncos, did you rebuild it ? --no-cache option ? | 19:42 |
| Lyncos | yeah | 19:42 |
| Lyncos | but I can't see apt getting memcache | 19:42 |
| Lyncos | I probably did something wierd | 19:43 |
| sdake | Lyncos are you using a registry and forgot to --push? | 19:43 |
| Lyncos | I'm always --push | 19:43 |
| Lyncos | and I also deleted images from my 'nodes' | 19:43 |
| Lyncos | and set docker_image_pull: always in global.yml | 19:43 |
| *** rajathagasthya has joined #openstack-kolla | 19:44 | |
| Lyncos | I'm re-building everything again just to make sure | 19:44 |
| ccesario | right | 19:49 |
| Lyncos | Is it normal that git review hangs ? | 19:51 |
| Lyncos | nvm | 19:53 |
| Lyncos | my bad :-) | 19:53 |
| openstackgerrit | Stephane Boisvert proposed openstack/kolla: Add python-memcache package cinder,glance,neutron https://review.openstack.org/311765 | 19:54 |
| *** absubram has quit IRC | 19:56 | |
| *** absubram has joined #openstack-kolla | 19:58 | |
| *** huikang has quit IRC | 19:58 | |
| openstackgerrit | Stephane Boisvert proposed openstack/kolla: Add python-memcache package cinder,glance,neutron https://review.openstack.org/311765 | 20:02 |
| *** mlima has left #openstack-kolla | 20:05 | |
| *** sdake_ has joined #openstack-kolla | 20:08 | |
| *** sdake has quit IRC | 20:08 | |
| ccesario | Lyncos, make the package list sortable :) as other Dockerfiles | 20:09 |
| openstackgerrit | Stephane Boisvert proposed openstack/kolla: Add python-memcache package cinder,glance,neutron https://review.openstack.org/311765 | 20:12 |
| Lyncos | done | 20:12 |
| *** huikang has joined #openstack-kolla | 20:13 | |
| inc0 | thank you Lyncos | 20:14 |
| Lyncos | np | 20:15 |
| ccesario | Lyncos, thank you ;) | 20:18 |
| Lyncos | I still don't have it installed in my glance image :-( | 20:18 |
| *** serverascode_ has joined #openstack-kolla | 20:23 | |
| *** ravig has quit IRC | 20:27 | |
| *** daneyon has joined #openstack-kolla | 20:28 | |
| *** ravig has joined #openstack-kolla | 20:28 | |
| *** elemoine- has joined #openstack-kolla | 20:28 | |
| *** ntpttr has quit IRC | 20:29 | |
| *** elemoine has quit IRC | 20:29 | |
| *** serverascode has quit IRC | 20:29 | |
| inc0 | Lyncos, did you rebuild it? | 20:29 |
| inc0 | and re-pushed | 20:29 |
| Lyncos | Yes | 20:29 |
| *** daneyon_ has quit IRC | 20:29 | |
| Lyncos | it's like my server are not pulling out the new image | 20:29 |
| Lyncos | even if I set docker_image_pull to always | 20:29 |
| *** daneyon_ has joined #openstack-kolla | 20:31 | |
| *** thumpba has quit IRC | 20:31 | |
| inc0 | strange | 20:31 |
| inc0 | well, you can run cleanup-images on stuff | 20:31 |
| Lyncos | I did | 20:31 |
| inc0 | then something is not right | 20:32 |
| Lyncos | in fact I can't see memcache at build step | 20:32 |
| inc0 | did you push them?;) | 20:32 |
| inc0 | ahh | 20:32 |
| inc0 | hmm | 20:32 |
| Lyncos | even if I do --no-cache | 20:32 |
| inc0 | Lyncos, you're running ubuntu? | 20:32 |
| Lyncos | Yes | 20:32 |
| *** banix has quit IRC | 20:33 | |
| inc0 | ahhh | 20:33 |
| Lyncos | kolla-build --push --type binary --base ubuntu --registry our.repo.org:5000 --no-cache -T 500 --namespace kollaglue glance-api | 20:33 |
| inc0 | hold on | 20:33 |
| Lyncos | It's like it's not reading the right dockerfile | 20:33 |
| ccesario | Lyncos, with me it is working | 20:34 |
| inc0 | https://github.com/openstack/kolla/blob/master/docker/openstack-base/Dockerfile.j2#L108 | 20:34 |
| inc0 | something is wrong | 20:34 |
| inc0 | in your case | 20:34 |
| inc0 | you should have python-memcached installed | 20:34 |
| *** serverascode_ is now known as serverascode | 20:34 | |
| Lyncos | in every images right ? | 20:35 |
| *** daneyon_ has quit IRC | 20:35 | |
| inc0 | every openstack image | 20:35 |
| dims | inc0 : that one is "python-memcached", Lyncos is installing "python-memcache" without the 'd' | 20:35 |
| Lyncos | it's not working | 20:35 |
| inc0 | well, python lib for talking to memchace is python-memcached afair | 20:36 |
| dims | dunno what the diff is | 20:36 |
| inc0 | ubuntu might have renamed it | 20:36 |
| Lyncos | but none is working | 20:36 |
| ccesario | inc0, https://github.com/openstack/kolla/blob/master/docker/openstack-base/Dockerfile.j2#L108 but this is reagarding Redhat based .... right!? | 20:36 |
| dims | global requirements has the python package "python-memcached>=1.56 # PSF" | 20:37 |
| *** jtriley has quit IRC | 20:37 | |
| Lyncos | python-memcached or python-memcache is not installed in both case | 20:37 |
| Lyncos | I think I was wrong with python-memcache and it is really python-memcached which is not installed | 20:38 |
| inc0 | ccesario, this one is not https://github.com/openstack/kolla/blob/master/docker/openstack-base/Dockerfile.j2#L307 | 20:38 |
| Lyncos | in cinder_api I can see python-pymemcache module | 20:38 |
| *** ravig has quit IRC | 20:39 | |
| Lyncos | nothing in glance or neutron | 20:39 |
| inc0 | hmf | 20:39 |
| inc0 | thats bad | 20:39 |
| ccesario | inc0, yes. But the problem only happen in binary mode ... in source mode it is ok :) | 20:40 |
| Lyncos | you maybe better to double check on your side ... maybe my environment is wrong... | 20:40 |
| inc0 | Lyncos, you're running ubuntu-source? | 20:40 |
| inc0 | if not, do this | 20:40 |
| Lyncos | Binary | 20:40 |
| inc0 | I stopped running binary | 20:40 |
| inc0 | and ubuntu binary was never really well tested | 20:41 |
| sbezverk | ccesario have you repro the same issue as Lyncos reports? | 20:41 |
| Lyncos | in openstack-base you don't have any memcache | 20:41 |
| ccesario | inc0, the problem is related to ubuntu-binary | 20:41 |
| ccesario | sbezverk, yes | 20:41 |
| inc0 | yes, well, I would recomend against running ubuntu binary | 20:41 |
| sbezverk | ccesario cool, thanks for confirming,, as I could not find any logs.. | 20:41 |
| inc0 | Lyncos, is there anything specific about source that would prevent you from using it? | 20:42 |
| ccesario | sbezverk, I post the logs in launchpad | 20:42 |
| Lyncos | inc0 .. I was thinking binary would be faster to build ? no real reason | 20:42 |
| Lyncos | That make my bug request useless then | 20:43 |
| Lyncos | if you guys fix it in openstack-base | 20:43 |
| *** macsz has quit IRC | 20:43 | |
| inc0 | Lyncos, use source, much more tested with ubuntu | 20:43 |
| Lyncos | ok | 20:43 |
| inc0 | keep the bug, it's valid | 20:43 |
| Lyncos | will do | 20:43 |
| *** macsz has joined #openstack-kolla | 20:43 | |
| Lyncos | I'm trying source right now | 20:44 |
| inc0 | ubuntu-source should work, but frankly everyone runs source builds that I know of | 20:44 |
| inc0 | centos doesn't show any apparent benefit of running binary | 20:44 |
| inc0 | besides having to deal with repo issues | 20:44 |
| *** ravig has joined #openstack-kolla | 20:45 | |
| *** rajathagasthya has quit IRC | 20:45 | |
| Lyncos | I'll fix the bug tomorrow ... removing modifications from individual base files (cinder, glance neutron) and modify openstack-base instead | 20:46 |
| inc0 | would be better | 20:46 |
| inc0 | thank you, I'll make comment on patchset | 20:46 |
| Lyncos | this will make it more 'clean' | 20:46 |
| *** rajathagasthya has joined #openstack-kolla | 20:47 | |
| inc0 | definetly | 20:51 |
| inc0 | that's what it's for | 20:51 |
| *** ravig has quit IRC | 20:53 | |
| *** ravig has joined #openstack-kolla | 20:54 | |
| *** harlowja has joined #openstack-kolla | 20:56 | |
| *** inc0 has quit IRC | 20:58 | |
| *** dmk0202 has joined #openstack-kolla | 21:00 | |
| openstackgerrit | Stephane Boisvert proposed openstack/kolla: Add python-memcache package openstack base https://review.openstack.org/311765 | 21:01 |
| *** sdake_ has quit IRC | 21:03 | |
| *** sdake has joined #openstack-kolla | 21:03 | |
| *** ccesario___ has joined #openstack-kolla | 21:03 | |
| sdake | hey cats | 21:04 |
| sdake | inc0 not true at all, centos binary is faster then sourc ebuilds | 21:07 |
| sdake | because centos binary compiles the python bytecode | 21:07 |
| *** intr1nsic has quit IRC | 21:09 | |
| *** intr1nsic has joined #openstack-kolla | 21:10 | |
| *** ByPasS has quit IRC | 21:10 | |
| *** ntpttr- has quit IRC | 21:10 | |
| *** ByPasS has joined #openstack-kolla | 21:12 | |
| *** dmk0202 has quit IRC | 21:12 | |
| jpeeler | i thought the pyc files were generated after first execution anyway | 21:12 |
| jpeeler | is that not true sdake ? | 21:13 |
| *** ntpttr has joined #openstack-kolla | 21:13 | |
| *** daneyon has quit IRC | 21:16 | |
| sdake | inc0 i got my lab setup | 21:17 |
| sdake | jpeeler in rpms they are created prior to first start | 21:18 |
| sdake | on from source we drop root so the binaries never get executed | 21:18 |
| sdake | so there is never any precompiled bytecode | 21:18 |
| sdake | its probably 5-15% difference in performance | 21:18 |
| jpeeler | oh right, drop root | 21:18 |
| sdake | considering openstack is bloated anyway | 21:18 |
| sdake | it probably doesn't matter all that much | 21:18 |
| *** haplo37 has quit IRC | 21:19 | |
| sdake | inc0 i have 3 of my workstations now setup | 21:21 |
| sdake | guess inc0 is gone for the day | 21:21 |
| *** mbound has joined #openstack-kolla | 21:25 | |
| *** ccesario___ has quit IRC | 21:28 | |
| *** huikang has quit IRC | 21:29 | |
| sbezverk | sdake if you have 5 minutes, could you check this PS please https://review.openstack.org/#/c/308447/ | 21:35 |
| patchbot | sbezverk: patch 308447 - kolla - Addining ability to specify capabilities and security | 21:35 |
| *** ravig has quit IRC | 21:42 | |
| *** ravig has joined #openstack-kolla | 21:43 | |
| *** ravig has quit IRC | 21:46 | |
| *** ravig has joined #openstack-kolla | 21:53 | |
| *** banix has joined #openstack-kolla | 21:56 | |
| *** daneyon has joined #openstack-kolla | 21:59 | |
| *** thumpba has joined #openstack-kolla | 22:01 | |
| *** ccesario___ has joined #openstack-kolla | 22:01 | |
| *** ntpttr- has joined #openstack-kolla | 22:05 | |
| *** ccesario___ has quit IRC | 22:06 | |
| *** thumpba has quit IRC | 22:08 | |
| *** mbound has quit IRC | 22:11 | |
| *** betheynyx has quit IRC | 22:13 | |
| *** absubram has quit IRC | 22:16 | |
| *** fragatin_ has joined #openstack-kolla | 22:19 | |
| *** ravig has quit IRC | 22:20 | |
| *** fragatina has quit IRC | 22:22 | |
| *** ravig has joined #openstack-kolla | 22:26 | |
| *** rajathagasthya has quit IRC | 22:27 | |
| *** ravig has quit IRC | 22:32 | |
| *** ravig has joined #openstack-kolla | 22:32 | |
| *** ravig has quit IRC | 22:34 | |
| *** ravig has joined #openstack-kolla | 22:34 | |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla: Doc rework for quickstart, multinode, and image-building https://review.openstack.org/311884 | 22:34 |
| *** daneyon has quit IRC | 22:36 | |
| *** ravig has quit IRC | 22:38 | |
| *** ravig has joined #openstack-kolla | 22:38 | |
| *** fragatina has joined #openstack-kolla | 22:39 | |
| *** fragatina has quit IRC | 22:39 | |
| *** fragatina has joined #openstack-kolla | 22:40 | |
| *** banix has quit IRC | 22:42 | |
| *** ravig has quit IRC | 22:42 | |
| *** ravig has joined #openstack-kolla | 22:42 | |
| *** asalkeld has joined #openstack-kolla | 22:42 | |
| *** dwalsh has quit IRC | 22:43 | |
| sdake | rhallisey 20 comments on spec - please address ;) | 22:47 |
| *** betheynyx has joined #openstack-kolla | 22:57 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla: Addining ability to specify capabilities and security https://review.openstack.org/308447 | 23:03 |
| asalkeld | sdake: fyi, I am keeping out of the discussion on kube. I pretty much disagree with both points (abi and and seperate repo) and don't want to get into an anti kolla-mesos thing. IMHO technically there was nothing wrong with our approach and it was purely business/management descision to drop it that caused us to step away. | 23:04 |
| sdake | asalkeld fair enough | 23:05 |
| sdake | just to be clear though, you mean you prefer one repo? | 23:05 |
| sdake | or you want separate repo | 23:05 |
| sdake | (you were unclear above) | 23:06 |
| asalkeld | no, per deployment makes senes to me | 23:06 |
| sdake | per COE repo yo umean | 23:06 |
| asalkeld | i don't know what a "coe" is | 23:06 |
| sdake | asalkeld what i'd do is send a note to hte mailing liststating sincey ou are not really actively involvedi nthe projet you do not wish your vote to be counted | 23:06 |
| sdake | container orchestration engine | 23:06 |
| sdake | because as is we have 12 cores | 23:06 |
| sdake | which means if its 6/6, we will hae a tiebreak problem | 23:07 |
| asalkeld | i'll do that if my vote is needed | 23:07 |
| sdake | with 11 people eligble to vote, only 6 votes are needed rather then 7 | 23:08 |
| sdake | so you see you could end up being a tiebreaker | 23:08 |
| sdake | which i am sure you dont want after what you just said above ;) | 23:08 |
| sdake | I think it may come to a tie break :) | 23:08 |
| asalkeld | ok, well let me know | 23:08 |
| sdake | with the current voting | 23:08 |
| sdake | will do | 23:08 |
| sdake | thanks for your honesty | 23:08 |
| asalkeld | also if you remember correctly i did give you a detailed response about the abi question | 23:09 |
| sdake | i dont sorry | 23:09 |
| sdake | but not saying you didn't | 23:09 |
| sdake | just dont recall - brain mush | 23:09 |
| asalkeld | you said i replied "meh" | 23:09 |
| sdake | my mistatement | 23:10 |
| sdake | would you like me to correct it ? | 23:10 |
| sdake | i'm sorry angus i can't remember every single detailed converstaion i have about every technical issue | 23:10 |
| sdake | buti do know it wasn't resolved | 23:10 |
| sdake | if you want me to take action to correct it i can do that | 23:10 |
| asalkeld | docker abi is like a library, define what can be done the format. not exact content. This is like saying every heat template must be written by the heat team | 23:10 |
| sdake | since i'm obviously inthe wrong here | 23:11 |
| asalkeld | sdake: don't worry | 23:11 |
| sdake | again apologies if that got your bloodpressure up ;( | 23:11 |
| sdake | as you can tell this morning was prety heated :) | 23:12 |
| asalkeld | i wrote up an etherpad, can't find it - I'll hunt out my irc history | 23:12 |
| sdake | need a thermometor | 23:12 |
| sdake | i definately dont remember an etherpad | 23:12 |
| sdake | what all that thread shows is that the kolla team really cares about kolla strongly | 23:13 |
| sdake | but has different opinions | 23:13 |
| *** daneyon has joined #openstack-kolla | 23:13 | |
| sdake | since we all have different logical thought patterns :) | 23:13 |
| sdake | asalkeld anyway i wouldn't worry about it, nobody actually reads what I write :) | 23:14 |
| *** daneyon has quit IRC | 23:18 | |
| *** ravig has quit IRC | 23:19 | |
| *** jrist has quit IRC | 23:19 | |
| asalkeld | sdake: here is the etherpad https://etherpad.openstack.org/p/kolla-set_configs | 23:21 |
| sdake | wow didn't see that at all | 23:22 |
| sdake | ya i like the utopia model | 23:23 |
| sdake | even though its unused | 23:24 |
| sdake | its one abi | 23:24 |
| sdake | vs two | 23:24 |
| sdake | thanks for sharing that document | 23:24 |
| sdake | i hope we have utopia model in k8s and ansible abi | 23:25 |
| sdake | rhallisey can you somehow work that utopia model into the spec | 23:25 |
| asalkeld | sdake: mistake | 23:26 |
| asalkeld | just saying, i think you are nuts | 23:26 |
| sdake | i am nuts | 23:26 |
| sdake | your right | 23:26 |
| *** daneyon has joined #openstack-kolla | 23:26 | |
| sdake | why else would anyone be a ptl of a project in openstack? :) | 23:27 |
| asalkeld | content of those files != ABI, | 23:27 |
| sdake | well there are no dependencies in the k8s model | 23:27 |
| asalkeld | ABI is the format and they ENV | 23:27 |
| sdake | so the utopia model is possible | 23:27 |
| sdake | abi si the json file and env right | 23:27 |
| asalkeld | format, not specific files - IMHO | 23:28 |
| *** daneyon has quit IRC | 23:28 | |
| asalkeld | anyway, i'll stop arguing... | 23:28 |
| sdake | thanks for findin the link asalkeld | 23:29 |
| *** daneyon has joined #openstack-kolla | 23:29 | |
| sdake | woud you like me to post it to the thread to correct my "meH" statemet | 23:29 |
| asalkeld | no, don't give it fuel | 23:29 |
| sdake | I thought the lack of action was a meh, but you did indeed to action of writing up an etherpad | 23:29 |
| *** ccesario___ has joined #openstack-kolla | 23:30 | |
| sdake | to/do action | 23:30 |
| sdake | cool thanks for being mature about it :) | 23:30 |
| sdake | but i want you to understand that i thought you pretty much ignored my request | 23:31 |
| sdake | which kind of pissed me off | 23:31 |
| sdake | so it was a communication problem on my end | 23:31 |
| sdake | so please accept my apology on the meh statement atleast on irc :) | 23:31 |
| asalkeld | k | 23:31 |
| sdake | what i should have done is harassed you again about it | 23:32 |
| sdake | at which point you would have pointed out the etherpad | 23:32 |
| sdake | and then i wouldn't have written that meh statement | 23:32 |
| sdake | but i was neck deep in bugs and you were neck deep in othe rthings | 23:33 |
| sdake | so that didn't happen | 23:33 |
| sdake | both human, my error :) | 23:33 |
| sdake | if we could just read each other's minds :) | 23:33 |
| *** jrist has joined #openstack-kolla | 23:33 | |
| asalkeld | sdake: it says "if you are on the list" ... "please vote" - I am not on that list | 23:34 |
| sdake | you are - under kolla-core | 23:34 |
| sdake | maybe that was befoer kolla-core was added to the list | 23:35 |
| sdake | a couple cores have approached me and said ony cores should vote on this since its a policy decision | 23:35 |
| sdake | which i tend to agree with | 23:35 |
| sdake | breaking new ground here, not quite sure what to do | 23:35 |
| sdake | its not like i imagined half our cores want it integrated from the start | 23:36 |
| sdake | that seems obvious from the ml posts :) | 23:36 |
| *** daneyon has quit IRC | 23:36 | |
| sdake | it affects the new contribtors as much as the current kolla core team | 23:37 |
| *** ssurana has quit IRC | 23:37 | |
| sdake | and I want to be inclusive of our new contributors | 23:37 |
| *** sdake_ has joined #openstack-kolla | 23:41 | |
| ccesario___ | Evening folks!! | 23:41 |
| *** jtriley has joined #openstack-kolla | 23:41 | |
| *** sdake has quit IRC | 23:43 | |
| *** banix has joined #openstack-kolla | 23:46 | |
| sdake_ | hey ccesario___ | 23:47 |
| *** diogogmt has quit IRC | 23:51 | |
| ccesario___ | Hey sdake_ ! | 23:52 |
| *** diogogmt has joined #openstack-kolla | 23:54 | |
| *** sdake_ has quit IRC | 23:55 | |
| *** sdake has joined #openstack-kolla | 23:56 | |
| *** sdake has quit IRC | 23:56 | |
| *** sdake has joined #openstack-kolla | 23:57 | |
| *** sdake has quit IRC | 23:58 | |
| ccesario___ | sdake, how is going the kubernetes directions? | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!