*** jthorne has joined #openstack-ansible | 00:02 | |
*** jamielennox|away is now known as jamielennox | 00:02 | |
*** Bjoern has quit IRC | 00:05 | |
*** jthorne has quit IRC | 00:06 | |
cloudnull | cfarquhar: not to my knowledge (for the most part) i mean the cli commands are not 100% idempotent however there's nothing the playbooks in liberty and beyond that cant be run more than once | 00:13 |
---|---|---|
cloudnull | admin0 still around ? | 00:13 |
cfarquhar | cloudnull perfect. thanks. | 00:13 |
cloudnull | admin0 ping me when your back around , if you see this messaage | 00:14 |
*** oneswig has joined #openstack-ansible | 00:23 | |
*** abitha has joined #openstack-ansible | 00:25 | |
*** jthorne has joined #openstack-ansible | 00:26 | |
*** sdake has joined #openstack-ansible | 00:27 | |
*** oneswig has quit IRC | 00:28 | |
Nepoc | quick question, is there a flag I can set so it skips all the "update apt sources" | 00:38 |
*** jamielennox is now known as jamielennox|away | 00:38 | |
*** pegmanm has joined #openstack-ansible | 00:39 | |
*** jamielennox|away is now known as jamielennox | 00:39 | |
*** pegmanm_ has quit IRC | 00:41 | |
*** zhangjn has joined #openstack-ansible | 00:42 | |
openstackgerrit | Merged openstack/openstack-ansible: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/272723 | 00:48 |
*** eil397 has quit IRC | 00:58 | |
shanec | anyone familiar with rpc-maas i can ping with some questions? | 00:59 |
*** spotz is now known as spotz_zzz | 01:00 | |
*** Bjoern has joined #openstack-ansible | 01:11 | |
*** jthorne has quit IRC | 01:12 | |
*** jthorne has joined #openstack-ansible | 01:13 | |
*** skamithi has joined #openstack-ansible | 01:15 | |
*** skamithi has left #openstack-ansible | 01:16 | |
*** baker has joined #openstack-ansible | 01:27 | |
*** baker_ has joined #openstack-ansible | 01:27 | |
*** abitha has quit IRC | 01:30 | |
*** jamielennox is now known as jamielennox|away | 01:30 | |
*** baker has quit IRC | 01:32 | |
*** flwang has quit IRC | 01:39 | |
*** pegmanm_ has joined #openstack-ansible | 01:40 | |
*** pegmanm has quit IRC | 01:43 | |
openstackgerrit | Merged openstack/openstack-ansible: Use Swift venv Python interpreter https://review.openstack.org/275256 | 01:44 |
*** baker_ has quit IRC | 01:45 | |
*** baker has joined #openstack-ansible | 01:46 | |
*** skamithi1 has joined #openstack-ansible | 01:52 | |
*** skamithi1 has left #openstack-ansible | 01:53 | |
*** flwang has joined #openstack-ansible | 01:53 | |
*** skamithi has joined #openstack-ansible | 02:00 | |
*** skamithi has left #openstack-ansible | 02:00 | |
*** sdake has quit IRC | 02:05 | |
*** baker has quit IRC | 02:08 | |
*** baker has joined #openstack-ansible | 02:10 | |
*** oneswig has joined #openstack-ansible | 02:12 | |
*** baker has quit IRC | 02:13 | |
*** oneswig has quit IRC | 02:16 | |
*** baker has joined #openstack-ansible | 02:18 | |
*** jthorne_ has joined #openstack-ansible | 02:19 | |
*** weezS has quit IRC | 02:19 | |
*** baker has quit IRC | 02:22 | |
*** jthorne has quit IRC | 02:23 | |
*** jamielennox|away is now known as jamielennox | 02:31 | |
*** pegmanm has joined #openstack-ansible | 02:41 | |
*** izaakk has quit IRC | 02:43 | |
*** pegmanm_ has quit IRC | 02:43 | |
*** daneyon_ has joined #openstack-ansible | 02:51 | |
*** daneyon has quit IRC | 02:54 | |
*** sdake has joined #openstack-ansible | 03:06 | |
*** pegmanm_ has joined #openstack-ansible | 03:42 | |
*** pegmanm has quit IRC | 03:45 | |
*** shanec has quit IRC | 03:49 | |
*** baker has joined #openstack-ansible | 03:52 | |
openstackgerrit | Logan V proposed openstack/openstack-ansible: Add support for Horizon customization module https://review.openstack.org/275494 | 03:56 |
*** oneswig has joined #openstack-ansible | 04:00 | |
*** oneswig has quit IRC | 04:05 | |
*** jamielennox is now known as jamielennox|away | 04:09 | |
*** sdake has quit IRC | 04:10 | |
*** baker_ has joined #openstack-ansible | 04:15 | |
*** baker has quit IRC | 04:18 | |
*** metral is now known as metral_zzz | 04:20 | |
openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Set hard requirements for pip https://review.openstack.org/275498 | 04:25 |
openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Set hard requirements for pip https://review.openstack.org/275499 | 04:28 |
*** metral_zzz is now known as metral | 04:30 | |
openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Set hard requirements for pip https://review.openstack.org/275500 | 04:30 |
prometheanfire | cloudnull: bed | 04:32 |
*** shanec has joined #openstack-ansible | 04:40 | |
*** shanec_ has joined #openstack-ansible | 04:42 | |
*** pegmanm has joined #openstack-ansible | 04:43 | |
*** shanec has quit IRC | 04:44 | |
*** pegmanm_ has quit IRC | 04:46 | |
*** pegmanm has quit IRC | 04:47 | |
*** pegmanm has joined #openstack-ansible | 04:48 | |
cloudnull | loll | 04:50 |
*** abitha has joined #openstack-ansible | 05:16 | |
*** jthorne_ has quit IRC | 05:32 | |
*** jthorne has joined #openstack-ansible | 05:33 | |
*** Bjoern has quit IRC | 05:33 | |
*** baker_ has quit IRC | 05:42 | |
*** abitha has quit IRC | 05:47 | |
*** oneswig has joined #openstack-ansible | 05:48 | |
*** pegmanm_ has joined #openstack-ansible | 05:49 | |
*** baker has joined #openstack-ansible | 05:51 | |
*** javeriak has joined #openstack-ansible | 05:51 | |
*** pegmanm has quit IRC | 05:52 | |
*** oneswig has quit IRC | 05:53 | |
*** baker has quit IRC | 05:56 | |
*** javeriak has quit IRC | 06:02 | |
*** javeriak has joined #openstack-ansible | 06:02 | |
*** weezS has joined #openstack-ansible | 06:23 | |
*** weezS has quit IRC | 06:27 | |
*** weezS has joined #openstack-ansible | 06:27 | |
*** phiche has joined #openstack-ansible | 06:30 | |
*** phiche1 has joined #openstack-ansible | 06:30 | |
*** jthorne has quit IRC | 06:33 | |
*** weezS has quit IRC | 06:33 | |
*** pegmanm has joined #openstack-ansible | 06:50 | |
*** pegmanm_ has quit IRC | 06:53 | |
*** phiche1 has quit IRC | 06:54 | |
*** phiche has joined #openstack-ansible | 07:02 | |
*** javeriak has quit IRC | 07:09 | |
*** llu has left #openstack-ansible | 07:10 | |
*** cemmason has joined #openstack-ansible | 07:11 | |
*** javeriak has joined #openstack-ansible | 07:11 | |
*** lbragstad has quit IRC | 07:16 | |
*** lbragstad has joined #openstack-ansible | 07:21 | |
*** spotz_zzz is now known as spotz | 07:34 | |
*** spotz is now known as spotz_zzz | 07:35 | |
*** spotz_zzz is now known as spotz | 07:35 | |
*** oneswig has joined #openstack-ansible | 07:36 | |
*** javeriak has quit IRC | 07:39 | |
*** oneswig has quit IRC | 07:41 | |
*** shanec_ has quit IRC | 07:44 | |
*** pegmanm_ has joined #openstack-ansible | 07:51 | |
*** targon has joined #openstack-ansible | 07:52 | |
*** pegmanm has quit IRC | 07:54 | |
*** tomoe_ has joined #openstack-ansible | 07:54 | |
*** spotz is now known as spotz_zzz | 07:57 | |
*** zhangjn has quit IRC | 08:07 | |
*** zhangjn has joined #openstack-ansible | 08:09 | |
*** zhangjn has quit IRC | 08:10 | |
*** zhangjn has joined #openstack-ansible | 08:10 | |
*** zhangjn has quit IRC | 08:11 | |
*** zhangjn has joined #openstack-ansible | 08:11 | |
*** zhangjn has quit IRC | 08:11 | |
*** zhangjn has joined #openstack-ansible | 08:12 | |
*** zhangjn has quit IRC | 08:12 | |
*** javeriak has joined #openstack-ansible | 08:13 | |
*** zhangjn has joined #openstack-ansible | 08:15 | |
*** zhangjn has quit IRC | 08:16 | |
*** zhangjn has joined #openstack-ansible | 08:17 | |
*** mikelk has joined #openstack-ansible | 08:25 | |
*** zhangjn has quit IRC | 08:25 | |
*** jthorne has joined #openstack-ansible | 08:33 | |
*** jthorne has quit IRC | 08:38 | |
*** targon has quit IRC | 08:50 | |
*** pegmanm_ has quit IRC | 08:54 | |
*** pegmanm has joined #openstack-ansible | 08:56 | |
*** MCoLo has joined #openstack-ansible | 09:11 | |
*** sdake has joined #openstack-ansible | 09:15 | |
*** zhangjn has joined #openstack-ansible | 09:22 | |
*** zhangjn has quit IRC | 09:24 | |
*** oneswig has joined #openstack-ansible | 09:25 | |
*** oneswig has quit IRC | 09:29 | |
*** keedya has quit IRC | 09:29 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275578 | 09:30 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275581 | 09:33 |
*** javeriak_ has joined #openstack-ansible | 09:34 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275581 | 09:35 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275581 | 09:37 |
*** javeriak has quit IRC | 09:38 | |
*** tomoe_ has quit IRC | 09:40 | |
odyssey4me | git-harry any chance that you can do the same sort of patch as https://review.openstack.org/273943 for, it seems, horizon-manage, horizon-django, keystone-wsgi, neutron-ha-tool? | 09:44 |
git-harry | odyssey4me: mattt said he might pick up this task but I can sort it out if he's not going to get round to it any time soon. | 09:48 |
* git-harry looks at mattt | 09:48 | |
odyssey4me | thanks :) | 09:48 |
mattt | https://review.openstack.org/#/c/275279/ | 09:49 |
mattt | i'm still working through some edge cases | 09:49 |
mattt | (read: it doesn't work yet :P) | 09:49 |
odyssey4me | ah ok, thanks mattt :) | 09:49 |
mattt | np | 09:50 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Resolve bashate violation E005 https://review.openstack.org/275290 | 09:51 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Added notification options for keystone https://review.openstack.org/257547 | 09:52 |
*** zhangjn has joined #openstack-ansible | 09:57 | |
*** zhangjn_ has joined #openstack-ansible | 09:57 | |
*** zhangjn_ has quit IRC | 09:58 | |
*** zhangjn_ has joined #openstack-ansible | 10:00 | |
*** zhangjn_ has quit IRC | 10:00 | |
*** zhangjn_ has joined #openstack-ansible | 10:02 | |
*** zhangjn_ has quit IRC | 10:02 | |
*** zhangjn_ has joined #openstack-ansible | 10:02 | |
*** gparaskevas has joined #openstack-ansible | 10:06 | |
gparaskevas | Hello guys, any low hanging fruit for me to catch up? | 10:06 |
odyssey4me | morning gparaskevas :) | 10:06 |
odyssey4me | how're you doing? | 10:06 |
openstackgerrit | Merged openstack/openstack-ansible-specs: Spec: LBaaSv2/Octavia https://review.openstack.org/273749 | 10:07 |
gparaskevas | Hello Jesse i am good thanks for asking | 10:07 |
kysse | oh you silly smalltalkers | 10:07 |
odyssey4me | lol | 10:07 |
gparaskevas | xaaxax | 10:08 |
odyssey4me | mattt hughsaunders thoughts on getting this through? https://review.openstack.org/268784 | 10:08 |
gparaskevas | lol | 10:08 |
gparaskevas | many southpark fans around | 10:08 |
*** zhangjn_ has quit IRC | 10:15 | |
*** zhangjn_ has joined #openstack-ansible | 10:15 | |
kysse | odyssey4me: my co-worker asked what is the reason why use lxc containers instead of packages from distro? | 10:25 |
kysse | it can't be updating | 10:25 |
odyssey4me | kysse lxc containers are quite different from packages from the distro | 10:27 |
odyssey4me | kysse lxc containers are used (optionally), and are used for service separation | 10:27 |
kysse | yee that's what I was thinking too. | 10:27 |
kysse | but why? :-) | 10:28 |
odyssey4me | kysse building from source is done instead of using packages because it gives us more freedom to keep up with the source changes, or to use internal sources if that's desired (if you choose to maintain a fork) | 10:28 |
kysse | ye understood. | 10:29 |
odyssey4me | kysse the containers can easily be blown away and rebuilt, whereas that's not so easy on hosts - we also encountered issues where co-hosting services on hosts resulted in issues with tcp connections no timing out properly | 10:29 |
odyssey4me | so the service separation is both to make scaling more flexible (you can increase the number of containers serving a specific service as needed), and reduces issues that result from co-locating services | 10:30 |
kysse | yep yep | 10:31 |
kysse | thanks for answers. | 10:33 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Separate host group var in hosts plays https://review.openstack.org/272869 | 10:37 |
odyssey4me | mattt I added the 'cherry picked from' line to the commit message in https://review.openstack.org/272869 | 10:38 |
mattt | odyssey4me: cool thanks | 10:38 |
mattt | odyssey4me: will look at the other review in a minute, btw hughsaunders is out today | 10:38 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Allow neutron services to move around environment https://review.openstack.org/272871 | 10:39 |
*** javeriak_ has quit IRC | 10:41 | |
*** gparaskevas has quit IRC | 10:42 | |
odyssey4me | ah thanks mattt | 10:42 |
odyssey4me | mattt there's a backport which probably needs to go in asap: https://review.openstack.org/273810 | 10:42 |
*** gparaskevas has joined #openstack-ansible | 10:43 | |
odyssey4me | kysse sure thing, we could probably do better in the docs to add info about that in the preamble | 10:45 |
kysse | is there somekinda security angle of running conts? | 10:46 |
*** oneswig has joined #openstack-ansible | 10:47 | |
mattt | odyssey4me: i wrote that, should someone else backport ? | 10:50 |
odyssey4me | kysse perhaps - it doesn't necessarily reduce the attack surface as services are still exposed over the network, but it does in the sense that you can blow them away and rebuild them in order to ensure a pristine environment | 10:51 |
odyssey4me | mattt cloudnull owns the change, so your vote certifies that what was backported is the same as what you wrote | 10:51 |
odyssey4me | afk for a bit | 10:51 |
mattt | odyssey4me: k cool | 10:52 |
*** spotz_zzz is now known as spotz | 10:52 | |
openstackgerrit | Merged openstack/openstack-ansible: Fix rsync service restart in os_swift https://review.openstack.org/275324 | 10:58 |
openstackgerrit | Merged openstack/openstack-ansible: Docs: Using security-hardening.yml https://review.openstack.org/273056 | 10:58 |
*** electrofelix has joined #openstack-ansible | 10:58 | |
*** zhangjn has quit IRC | 11:02 | |
*** zhangjn has joined #openstack-ansible | 11:03 | |
*** zhangjn has quit IRC | 11:05 | |
*** targon has joined #openstack-ansible | 11:05 | |
*** zhangjn_ has quit IRC | 11:13 | |
openstackgerrit | Matt Thompson proposed openstack/openstack-ansible: [WIP] Remove venv activation code https://review.openstack.org/275279 | 11:13 |
*** zhangjn has joined #openstack-ansible | 11:13 | |
*** pegmanm_ has joined #openstack-ansible | 11:13 | |
*** spotz is now known as spotz_zzz | 11:15 | |
*** pegmanm has quit IRC | 11:16 | |
*** javeriak has joined #openstack-ansible | 11:27 | |
*** oneswig has quit IRC | 11:37 | |
*** oneswig has joined #openstack-ansible | 11:37 | |
*** oneswig has quit IRC | 11:42 | |
*** spotz_zzz is now known as spotz | 11:57 | |
*** spotz is now known as spotz_zzz | 12:08 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Restart horizon when package/venvs change https://review.openstack.org/273810 | 12:14 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Separate host group var in hosts plays https://review.openstack.org/272869 | 12:15 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Allow neutron services to move around environment https://review.openstack.org/272871 | 12:15 |
-openstackstatus- NOTICE: Infra running with lower capacity now, due to a temporary problem affecting one of our nodepool providers. Please expect some delays in your jobs. Apologies for any inconvenience caused. | 12:39 | |
*** zhangjn has quit IRC | 12:39 | |
*** javeriak has quit IRC | 12:41 | |
*** zhangjn has joined #openstack-ansible | 12:42 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-apt_package_pinning: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275673 | 12:47 |
*** zhangjn has quit IRC | 12:47 | |
*** zhangjn has joined #openstack-ansible | 12:48 | |
*** zhangjn has quit IRC | 12:48 | |
*** zhangjn has joined #openstack-ansible | 12:49 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rsyslog_client: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275680 | 12:58 |
*** tomoe_ has joined #openstack-ansible | 13:01 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-plugins: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275683 | 13:01 |
*** openstackgerrit has quit IRC | 13:02 | |
*** openstackgerrit has joined #openstack-ansible | 13:03 | |
pabelanger | odyssey4me: Is there a patch in project-config for linters already? | 13:05 |
odyssey4me | pabelanger the templates are there, but none of them are activated for any of the ansible projects | 13:05 |
odyssey4me | pabelanger I'm wanting to get all the tox configs in place before we shift to using it wholesale, which I think is a good idea | 13:06 |
*** tomoe_ has quit IRC | 13:06 | |
openstackgerrit | Matt Thompson proposed openstack/openstack-ansible: [WIP] Remove venv activation code https://review.openstack.org/275279 | 13:07 |
pabelanger | odyssey4me: I'm getting an error: http://pastebin.com/U9CkLMTk | 13:08 |
odyssey4me | pabelanger I hope you don't mind me submitting patches against the ansible-role repositories? :) | 13:08 |
odyssey4me | pabelanger ah, it might be that your pep8 env has no deps specified? | 13:08 |
pabelanger | odyssey4me: no, that is fine. Just to keep in mind, I try to support tox 1.4.2, since that ships on eple7 | 13:08 |
pabelanger | odyssey4me: Ya, let me play with it and get it working | 13:09 |
*** oneswig has joined #openstack-ansible | 13:09 | |
odyssey4me | pabelanger cool, thanks - feel free to hack at the review for your repo, then replicate it to the other ansible-role repositories you're curating... I'll leave those to you | 13:12 |
*** tomoe_ has joined #openstack-ansible | 13:12 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-repo_server: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275688 | 13:12 |
*** tomoe_ has quit IRC | 13:12 | |
*** tomoe_ has joined #openstack-ansible | 13:13 | |
*** oneswig has quit IRC | 13:13 | |
pabelanger | odyssey4me: sounds good | 13:18 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rsyslog_server: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275689 | 13:19 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-pip_install: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275691 | 13:21 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-rabbitmq_server: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275692 | 13:23 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-py_from_git: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275693 | 13:26 |
mattt | odyssey4me: the problem w/ https://review.openstack.org/268784 is that i have no way to test if any of this stuff is fine to remove or not :) | 13:26 |
odyssey4me | mattt it's for an AIO, so does it matter? it's easy enough to add things back in if they're needed... but it definitely still passes the gate without them, and all those actions do is slow the gate down | 13:27 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275698 | 13:30 |
odyssey4me | mattt especially for an AIO, I don't see the purpose of carrying things 'just in case' | 13:31 |
mattt | odyssey4me: oh, i thought there were plans on some downstream projects using run-playbooks.sh | 13:31 |
odyssey4me | going afk for a few hours while I run some errands | 13:31 |
odyssey4me | mattt if any downstream projects use it, it had best be for an AIO and for gating | 13:32 |
odyssey4me | that's the purpose of the script | 13:32 |
mattt | odyssey4me: ok then i feel less concerned about it then | 13:32 |
odyssey4me | that patch is part of my plan to convert the script into some meta plays anyway - I want to be rid of it and use ansible as far as possible | 13:32 |
mattt | odyssey4me: i see a +2 in your future | 13:33 |
odyssey4me | lol | 13:33 |
odyssey4me | bbl | 13:33 |
*** jthorne has joined #openstack-ansible | 13:42 | |
*** jthorne has quit IRC | 13:43 | |
*** jthorne has joined #openstack-ansible | 13:44 | |
*** retreved has joined #openstack-ansible | 13:45 | |
*** retreved has quit IRC | 13:46 | |
*** retreved has joined #openstack-ansible | 13:47 | |
*** jthorne_ has joined #openstack-ansible | 13:49 | |
*** jthorne_ has quit IRC | 13:52 | |
*** jthorne has quit IRC | 13:52 | |
*** pegmanm_ has quit IRC | 14:01 | |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_container_create: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275710 | 14:07 |
mhayden | odyssey4me: i just put a little note here about the lbaasv2 agent -> https://review.openstack.org/#/c/275440/1 | 14:09 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-security: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275713 | 14:10 |
openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_client: Revise all lint tests and add consolidated 'linters' tox environment https://review.openstack.org/275716 | 14:11 |
* mhayden has some concerns around the architecture of lbaasv2/octavia, too | 14:12 | |
*** woodard has joined #openstack-ansible | 14:14 | |
odyssey4me | mhayden maybe it's worth considering having the v2agent using the old method in its own container so that it's easy to remove/replace when the time comes | 14:14 |
mhayden | well, the jump from lbaasv1 -> lbaasv2 + octavia is uh, rather large | 14:15 |
*** skamithi has joined #openstack-ansible | 14:15 | |
mhayden | :) | 14:15 |
odyssey4me | anyway, I'll check your comment out later - I just had a moment to get on with the linter patches... heading out to move some things from central london to home, which will take several hours | 14:15 |
mhayden | enjoy! | 14:15 |
*** zhangjn_ has joined #openstack-ansible | 14:15 | |
odyssey4me | yeah, my concern is that people use lbaasv1 in liberty and then find themselves stuck in a corner... maybe we should just add a release note indicating that it's a known issue that the implementation is v1 and that it is not recommended to use it in production | 14:16 |
mhayden | that might work | 14:16 |
odyssey4me | cite that it's deprecated in Liberty and the v2 implementation will come in Mitaka | 14:16 |
odyssey4me | then we do Octavia in Mitaka | 14:17 |
odyssey4me | if the Horizon plugins don't come through in time, then we just do the back-end initially and carry a known issue for that | 14:17 |
odyssey4me | ok, bbl | 14:17 |
mhayden | cool | 14:18 |
*** scarlisle has joined #openstack-ansible | 14:19 | |
*** tomoe_ has quit IRC | 14:19 | |
*** zhangjn_ has quit IRC | 14:20 | |
*** Bjoern has joined #openstack-ansible | 14:22 | |
*** zhangjn_ has joined #openstack-ansible | 14:22 | |
*** Bjoern is now known as Bjoern_zZzZzZzZ | 14:22 | |
*** tomoe_ has joined #openstack-ansible | 14:23 | |
*** b3rnard0_away is now known as b3rnard0 | 14:25 | |
*** Bjoern_zZzZzZzZ is now known as Bjoern | 14:30 | |
*** galstrom_zzz is now known as galstrom | 14:49 | |
*** oneswig has joined #openstack-ansible | 14:57 | |
*** portante has joined #openstack-ansible | 14:59 | |
*** zhangjn has quit IRC | 15:00 | |
*** pegmanm has joined #openstack-ansible | 15:02 | |
*** oneswig has quit IRC | 15:02 | |
*** spotz_zzz is now known as spotz | 15:03 | |
*** zhangjn has joined #openstack-ansible | 15:04 | |
*** KLevenstein has joined #openstack-ansible | 15:07 | |
*** Mudpuppy has joined #openstack-ansible | 15:10 | |
*** Mudpuppy has quit IRC | 15:10 | |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Docs: Update integration with OSA https://review.openstack.org/273050 | 15:11 |
*** Mudpuppy has joined #openstack-ansible | 15:11 | |
*** CheKoLyN has joined #openstack-ansible | 15:18 | |
*** javeriak has joined #openstack-ansible | 15:22 | |
*** tomoe_ has quit IRC | 15:23 | |
*** sigmavirus24_awa is now known as sigmavirus24 | 15:24 | |
*** tomoe_ has joined #openstack-ansible | 15:26 | |
*** baker has joined #openstack-ansible | 15:27 | |
*** mgoddard_ has joined #openstack-ansible | 15:30 | |
*** mgoddard has quit IRC | 15:33 | |
cloudnull | odyssey4me: and the rest of cores: we need these through https://review.openstack.org/#/q/Iddcbba94c62dc87dac6fc25efb17b99ee8ec37d6,n,z | 15:37 |
cloudnull | as a frame of reference, 12.0.5 is building pip8 | 15:37 |
cloudnull | http://rpc-repo.rackspace.com/os-releases/12.0.5/ | 15:37 |
cloudnull | and 12.0.4 | 15:37 |
mattt | cloudnull: can you respond to odyssey4me's comment ? | 15:37 |
cloudnull | http://rpc-repo.rackspace.com/os-releases/12.0.5/pip-8.0.2-py2.py3-none-any.whl | 15:38 |
cloudnull | http://rpc-repo.rackspace.com/os-releases/12.0.4/pip-8.0.2-py2.py3-none-any.whl | 15:38 |
mattt | also how come we're backporting stuff now before the master commit merges ? | 15:39 |
mattt | i thought that was a hard no-no :) | 15:39 |
cloudnull | i backported it and -2 it because kilo and liberty are effected by this | 15:39 |
cloudnull | it needs to go in with a quick ness. IMO | 15:39 |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Fix Postfix mynetworks if IPv6 is disabled https://review.openstack.org/275768 | 15:39 |
mattt | cloudnull: sure, can you respond to his question? | 15:40 |
mattt | because i too don't know enough about this to know why the multitude of pins aren't working | 15:40 |
cloudnull | replied | 15:41 |
cloudnull | The issue is that we are specifying the requirement which in our requirements.txt file and will always have the most priority because we've defined it. | 15:41 |
cloudnull | mattt: the only place its pinned is in the role | 15:42 |
cloudnull | pip_install | 15:42 |
cloudnull | and anything in one of these files will take priority over whats found in a given role defaults/main.yml file | 15:43 |
cloudnull | https://github.com/openstack/openstack-ansible-plugins/blob/master/lookups/py_pkgs.py#L33-L39 | 15:43 |
cloudnull | i tried to get this through the other week | 15:46 |
cloudnull | https://review.openstack.org/#/c/270153/10 | 15:46 |
*** jthorne has joined #openstack-ansible | 15:46 | |
cloudnull | but the actual constraint was removed in the final revision | 15:46 |
*** klamath has joined #openstack-ansible | 15:47 | |
*** jthorne has quit IRC | 15:47 | |
*** rgogunskiy has joined #openstack-ansible | 15:49 | |
*** zhangjn has quit IRC | 15:50 | |
*** zhangjn has joined #openstack-ansible | 15:51 | |
*** zhangjn has quit IRC | 15:51 | |
mattt | cloudnull: let me look at that now, but quick question for you | 15:52 |
mattt | cloudnull: http://paste.openstack.org/show/485863/ | 15:52 |
mattt | cloudnull: why is argparse not going into the venv? | 15:52 |
cloudnull | whats that ? | 15:52 |
mattt | cloudnull: it's skipping putting argparse into container because "requirement already met" | 15:53 |
mattt | but it's pointing to the host's python path | 15:53 |
cloudnull | thats an issue with pip | 15:53 |
cloudnull | pip8 in particular | 15:53 |
cloudnull | idk if you have pip8 there but that was an issue with pip8 | 15:54 |
mattt | pip 7 | 15:54 |
*** raddaoui has joined #openstack-ansible | 15:54 | |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Docs: Add HAProxy SSL configuration options https://review.openstack.org/275778 | 15:54 |
cloudnull | the other cause is that argparse is in py27 | 15:54 |
*** raddaoui_ has joined #openstack-ansible | 15:54 | |
cloudnull | so maybe its already there because its not a package that needs to be installed ? | 15:54 |
*** jthorne has joined #openstack-ansible | 15:55 | |
cloudnull | are you seeing an issue with argparse not being found ? | 15:55 |
mattt | cloudnull: yeah on keystone, specifically this change https://review.openstack.org/#/c/275279/3/playbooks/roles/os_keystone/templates/keystone-httpd.conf.j2 | 15:55 |
mattt | cloudnull: i'm trying ot lock apache to the venv so we can remove the activate cruft from the wsgi file | 15:55 |
*** galstrom is now known as galstrom_zzz | 15:57 | |
cloudnull | http://logs.openstack.org/79/275279/3/check/gate-openstack-ansible-dsvm-commit/b1c70a4/console.html#_2016-02-03_14_40_12_735 so its throwing a 503 | 15:59 |
mattt | cloudnull: yeah because apache is puking | 15:59 |
*** jthorne has quit IRC | 15:59 | |
*** raddaoui_ has quit IRC | 15:59 | |
*** raddaoui has quit IRC | 15:59 | |
mattt | cloudnull: 2016-02-03 15:04:26.000735 DistributionNotFound: The 'argparse' distribution was not found and is required by oslo.config, python-keystoneclient, stevedore, keystoneauth1 | 15:59 |
*** raddaoui has joined #openstack-ansible | 16:00 | |
cloudnull | looks like the apache error is -- ArgsAlreadyParsedError: arguments already parsed: cannot register CLI option | 16:00 |
mattt | cloudnull: http://logs.openstack.org/79/275279/3/check/gate-openstack-ansible-dsvm-commit/b1c70a4/logs/aio1_keystone_container-972d463b/keystone-apache-error.log search for DistributionNotFound | 16:00 |
mattt | if i install argparse into venv it starts fine | 16:00 |
*** raddaoui_ has joined #openstack-ansible | 16:01 | |
*** raddaoui_ has quit IRC | 16:01 | |
*** raddaoui has quit IRC | 16:01 | |
*** javeriak has quit IRC | 16:01 | |
*** raddaoui has joined #openstack-ansible | 16:01 | |
*** TheIntern has joined #openstack-ansible | 16:02 | |
cloudnull | mattt: http://cdn.pasteraw.com/dbboozqtaem1d5d0bou8bws091xosy9 | 16:03 |
cloudnull | i just checked a current master build | 16:03 |
cloudnull | the venv doesnt have argparse in it because its baked into the core python | 16:03 |
*** tomoe_ has quit IRC | 16:04 | |
cloudnull | and in my quick test I can import argparse and show the pathing | 16:04 |
*** tomoe_ has joined #openstack-ansible | 16:04 | |
*** galstrom_zzz is now known as galstrom | 16:05 | |
cloudnull | i'd be curious if you can do the same | 16:05 |
mattt | cloudnull: i can actually | 16:05 |
mattt | just tested | 16:06 |
cloudnull | so maybe something else is going on with the wsgi setup | 16:06 |
mattt | cloudnull: sounds like it! let me look into it further -- thanks for the info :) | 16:09 |
*** javeriak has joined #openstack-ansible | 16:10 | |
*** jthorne has joined #openstack-ansible | 16:10 | |
*** skamithi has quit IRC | 16:11 | |
*** galstrom is now known as galstrom_zzz | 16:11 | |
*** jthorne has quit IRC | 16:12 | |
*** mgoddard_ has quit IRC | 16:12 | |
*** mgoddard has joined #openstack-ansible | 16:12 | |
*** jthorne has joined #openstack-ansible | 16:12 | |
cloudnull | Bjoern cfarquhar: I'm rebuilding 12.0.5 (http://rpc-repo.rackspace.com/os-releases/12.0.5/) ensuring that pip7x is used instead of pip8 which will stabalize that release ill ping when its complete | 16:13 |
*** baker has quit IRC | 16:14 | |
cfarquhar | thanks cloudnull | 16:18 |
*** jthorne has quit IRC | 16:20 | |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Docs: nova config options for Ceph RBD https://review.openstack.org/275796 | 16:23 |
*** weezS has joined #openstack-ansible | 16:24 | |
*** daneyon has joined #openstack-ansible | 16:24 | |
spotz | mhayden you ready for me to peek at that or still working? | 16:24 |
*** baker has joined #openstack-ansible | 16:25 | |
*** daneyon_ has quit IRC | 16:27 | |
cloudnull | also mattt odyssey4me hughsaunders andymccr d34dh0r53 stevelle we need this https://review.openstack.org/#/c/274932/ | 16:27 |
cloudnull | and we might want to think about backporting it to liberty | 16:27 |
openstackgerrit | Miguel Alex Cantu (alextricity25) proposed openstack/openstack-ansible: Allow apache file overrides https://review.openstack.org/267749 | 16:28 |
*** raddaoui has quit IRC | 16:28 | |
openstackgerrit | Merged openstack/openstack-ansible: Separate host group var in hosts plays https://review.openstack.org/272869 | 16:28 |
openstackgerrit | Miguel Alex Cantu (alextricity25) proposed openstack/openstack-ansible: Allow apache file overrides https://review.openstack.org/267749 | 16:29 |
*** raddaoui has joined #openstack-ansible | 16:29 | |
*** shausy has joined #openstack-ansible | 16:34 | |
mattt | cloudnull: if this was available since kilo does it really matter if we don't backport to liberty? | 16:34 |
cloudnull | no. it doesnt. | 16:35 |
openstackgerrit | Miguel Alex Cantu (alextricity25) proposed openstack/openstack-ansible: Changed ceilometer configs to use keystonev2 https://review.openstack.org/274905 | 16:35 |
cloudnull | Im thinking that we backport it to liberty so that when we upgrade to mitaka the DB already exists. | 16:35 |
cloudnull | but operationally it shouldnt make a difference | 16:35 |
palendae | Making a new table mid-lifecycle seems eh | 16:36 |
mattt | palendae: yeah that's what i was thinking | 16:36 |
mattt | feels a bit late in the game to slip that in :P | 16:37 |
cloudnull | fair enough | 16:37 |
mattt | but like cloudnull said, it could be one less thing to do when you go to do a database upgrade for mitaka | 16:37 |
palendae | Initially I misundertood the report in nova that *they* brought it back to liberty, but that wasn't the case | 16:38 |
*** javeriak has quit IRC | 16:39 | |
palendae | mattt: Yeah, but one more thing to do between a minor version of liberty | 16:39 |
mattt | but that is a smaller scope of changes than liberty->mitaka | 16:39 |
*** javeriak has joined #openstack-ansible | 16:39 | |
palendae | Operationally, it's mostly pushing the time spent around the table | 16:39 |
palendae | Sure, though I expect DBs to appear on a major version, a little more surprising if it appears in a minor | 16:40 |
mattt | yeah true | 16:40 |
palendae | Would be worthwhile asking people outside ourselves, too :) | 16:41 |
*** javeriak_ has joined #openstack-ansible | 16:42 | |
cloudnull | I can add that to the meeting | 16:44 |
cloudnull | and see what folks thing | 16:44 |
*** javeriak has quit IRC | 16:44 | |
cloudnull | *think | 16:44 |
palendae | Yeah | 16:44 |
*** oneswig has joined #openstack-ansible | 16:45 | |
openstackgerrit | Miguel Alex Cantu (alextricity25) proposed openstack/openstack-ansible: Allow apache file overrides https://review.openstack.org/267749 | 16:47 |
*** gparaskevas has quit IRC | 16:48 | |
palendae | " LBaaS V2 reference driver is now based on Octavia, an operator grade scalable, reliable Load Balancer platform." | 16:50 |
*** oneswig has quit IRC | 16:50 | |
palendae | Whoops | 16:51 |
*** cmkirkla has joined #openstack-ansible | 16:52 | |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Docs: Explain container_mtu https://review.openstack.org/275819 | 16:52 |
*** tomoe_ has quit IRC | 16:53 | |
*** mgoddard_ has joined #openstack-ansible | 16:58 | |
*** mgoddard has quit IRC | 16:58 | |
*** phalmos has joined #openstack-ansible | 17:00 | |
*** phalmos has quit IRC | 17:03 | |
*** rgogunskiy has quit IRC | 17:04 | |
*** izaakk has joined #openstack-ansible | 17:04 | |
*** phiche has quit IRC | 17:06 | |
Nepoc | Hello everyone, is there any way to add in more provider networks that are not bridges? I have the required bridges but I need to add in 3 more physical connection to be used by VM's. When I've added them in as bridges the compute nodes end up creating new bridges when an instance is connected to those networks. | 17:10 |
*** tomoe_ has joined #openstack-ansible | 17:13 | |
*** tomoe_ has quit IRC | 17:17 | |
*** baker has quit IRC | 17:23 | |
*** baker has joined #openstack-ansible | 17:24 | |
openstackgerrit | Miguel Alex Cantu (alextricity25) proposed openstack/openstack-ansible: Changed ceilometer configs to use keystonev3 https://review.openstack.org/274905 | 17:24 |
openstackgerrit | Miguel Alex Cantu (alextricity25) proposed openstack/openstack-ansible: Allow apache file overrides https://review.openstack.org/267749 | 17:25 |
*** baker has quit IRC | 17:26 | |
*** cmkirkla has quit IRC | 17:28 | |
*** cmkirkla has joined #openstack-ansible | 17:30 | |
*** cmkirkla has left #openstack-ansible | 17:32 | |
*** jthorne has joined #openstack-ansible | 17:33 | |
*** targon has quit IRC | 17:36 | |
*** jthorne has quit IRC | 17:37 | |
*** jthorne has joined #openstack-ansible | 17:40 | |
*** raddaoui has quit IRC | 17:51 | |
*** TheIntern has quit IRC | 17:51 | |
*** weezS has quit IRC | 17:53 | |
*** jthorne has quit IRC | 17:57 | |
*** retreved has quit IRC | 17:59 | |
*** mikelk has quit IRC | 18:00 | |
bgmccollum | sigmavirus24 -- FYI, happened again...galera looks OK, but all cinder services are down, and not attempting to re-connect -- "WSREP has not yet prepared node for application use" | 18:02 |
*** baker has joined #openstack-ansible | 18:02 | |
sigmavirus24 | O_o | 18:05 |
sigmavirus24 | That's a much different error | 18:05 |
sigmavirus24 | Did you try tinkering with cinder's database connection retry settings and transaction retry settings? | 18:06 |
sigmavirus24 | bgmccollum: any chance that the cinder services are running with different schema? I can't imagine why that would be generating WSREP errors, but this seems genuinely bizarre | 18:07 |
*** automagically has joined #openstack-ansible | 18:08 | |
bgmccollum | sigmavirus24 i didn't touch anything -- they all died around 11PM last night, scheduler, volume (lvm backend), and volume (ceph backend), all on different physical nodes. | 18:11 |
sigmavirus24 | bgmccollum: wasn't saying you had touched anything | 18:11 |
* sigmavirus24 is just thinking of what could possibly go wrong between the service and the database | 18:11 | |
bgmccollum | sigmavirus24 hmm...different schemas like different database schemas? different version of the code? | 18:12 |
sigmavirus24 | right | 18:12 |
sigmavirus24 | Again, this doesn't seem liek an error that that situation would produce | 18:12 |
bgmccollum | something must be happening on the galera side, where briefly it loses quorem...then regains it...and in the window cinder just throws its hands and quits | 18:13 |
sigmavirus24 | This seems very much like a galera <-> maria problem but it's odd that it only happens with cinder | 18:13 |
sigmavirus24 | bgmccollum: right, that's why I'm suggesting looking at setting the retry values for cinder | 18:13 |
sigmavirus24 | it sounds like retries are off for those cinder services | 18:13 |
bgmccollum | where / what do i set that? | 18:13 |
sigmavirus24 | So in the [database] section of the config file you'll want to tinker with the following vars (which I don't think are templatized in a kilo deployment) max_retries (https://github.com/openstack/oslo.db/blob/stable/kilo/oslo_db/options.py#L77), retry_interval (https://github.com/openstack/oslo.db/blob/stable/kilo/oslo_db/options.py#L86), db_retry_interval (https://github.com/openstack/oslo.db/blob/stable/kilo/oslo_db/options. | 18:16 |
sigmavirus24 | py#L120), and db_max_retries (https://github.com/openstack/oslo.db/blob/stable/kilo/oslo_db/options.py#L132) | 18:16 |
sigmavirus24 | The stacktrace you showed yesterday makes me lean towards the latter two | 18:16 |
sigmavirus24 | That said, those all have good defaults so I'm not sure what's happening with them | 18:16 |
bgmccollum | ok, ill fire up cinder in debug and see it dump what those values are too... | 18:17 |
*** shausy has quit IRC | 18:20 | |
sigmavirus24 | bgmccollum: also if you're going to be using debug mode you might want to try out setting https://github.com/openstack/oslo.db/blob/stable/kilo/oslo_db/options.py#L100 | 18:23 |
sigmavirus24 | (connection_debug in [database]) | 18:23 |
bgmccollum | ok, ill add that, and wait for it to tip over again...everything else looks to be the defaults. | 18:25 |
sigmavirus24 | Interesting | 18:27 |
*** skamithi has joined #openstack-ansible | 18:27 | |
sigmavirus24 | I wonder if this exception is just totally unhandled by cinder | 18:28 |
bgmccollum | thats what i was thinking...the retry just doesn't kick in | 18:28 |
cloudnull | Nepoc: you can use a veth from one of the bridges to provide for more networks while using the same local fabric | 18:28 |
cloudnull | also if you have the ability to create more vlans you can create vlan tagged interfaces | 18:29 |
cloudnull | Nepoc: example on creating a veth pair from a bridge https://github.com/openstack/openstack-ansible/blob/master/etc/network/interfaces.d/aio_interfaces.cfg#L52-L58 | 18:30 |
cloudnull | Nepoc: example on creating vlan tagged interfaces https://github.com/openstack/openstack-ansible/blob/master/etc/network/interfaces.d/openstack_interface.cfg.example#L62-L72 | 18:30 |
*** electrofelix has quit IRC | 18:33 | |
*** raddaoui has joined #openstack-ansible | 18:33 | |
*** oneswig has joined #openstack-ansible | 18:34 | |
*** oneswig has quit IRC | 18:38 | |
*** eil397 has joined #openstack-ansible | 18:39 | |
Nepoc | cloudnull: I have tried both without success. Lets say I have eth0-eth4. eth0.997-eth0.999 in bridges br-vlan, br-mgmt, br-vxlan. eth0 is external, eth1 is network1, eth2 is network2 eth3 is network3. In neutron I need vlan 130 on eth0, vlan 110 on eth1 and vlan 120 on eth2 and finally external routes out eth0 flat. | 18:44 |
*** sdake has quit IRC | 18:47 | |
Nepoc | cloudnull: http://pastebin.com/NNhB82Xq | 18:48 |
Nepoc | cloudnull: interfaces and provider_networks | 18:48 |
Nepoc | cloudnull: that was my latest attempt | 18:50 |
*** michaelgugino has joined #openstack-ansible | 18:55 | |
*** raddaoui_ has joined #openstack-ansible | 18:55 | |
*** raddaoui has quit IRC | 18:59 | |
*** retreved has joined #openstack-ansible | 19:00 | |
*** retreved has quit IRC | 19:00 | |
*** retreved has joined #openstack-ansible | 19:01 | |
*** phiche has joined #openstack-ansible | 19:02 | |
*** Bjoern is now known as Bjoern_zZzZzZzZ | 19:03 | |
cloudnull | Nepoc: was that not working ? | 19:04 |
cloudnull | the provider networks entries look correct | 19:04 |
cloudnull | you might want to define "ip_from_q:" within the entries | 19:05 |
cloudnull | to ip them within the containers | 19:05 |
cloudnull | however they look ok | 19:05 |
*** jthorne has joined #openstack-ansible | 19:05 | |
Nepoc | The external network kills the network connection if I include it in the config. So it's currently commented out. The control network once the network is created in openstack and a vm connected it doesn't create eth0.130 and put it in a bridge on the compute node, it just puts eth0 in the bridge with the tap interface. | 19:07 |
Nepoc | Not sure how or why it differs from eth1.110 and eth2.120 as those are created correctly | 19:07 |
*** weezS has joined #openstack-ansible | 19:10 | |
*** jthorne has quit IRC | 19:10 | |
Nepoc | ah nevermind I see my problem there! | 19:10 |
*** jthorne has joined #openstack-ansible | 19:11 | |
Nepoc | let me just just try a little something something | 19:11 |
*** Bjoern_zZzZzZzZ is now known as Bjoern | 19:14 | |
openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Added FWAAS support https://review.openstack.org/275894 | 19:17 |
cloudnull | Nepoc: ^ | 19:17 |
cloudnull | added the option to the horizon dash with the latest change set | 19:18 |
cloudnull | and tested it, which looks good to go | 19:18 |
Nepoc | Excellent | 19:19 |
cloudnull | prometheanfire: 18f1605bddd428a014d0e43ef52d1af6305e1e03 | 19:19 |
*** phiche has quit IRC | 19:23 | |
michaelgugino | mhayden: did you mean you're getting a 404 from review.openstack.org, or are you building the docs somewhere? | 19:28 |
Nepoc | okay so that didn't work as I anticipated... container_type: "vlan" configures the vlan I need just phys instead it seems. | 19:29 |
*** raddaoui_ has quit IRC | 19:29 | |
*** phiche has joined #openstack-ansible | 19:29 | |
Nepoc | tcpdump on the containers eth2 interface gave me the vlan traffic from the compute node | 19:30 |
Nepoc | however if I use phys for eth0 I think that's when I lose connection to the infra host | 19:31 |
*** rgogunskiy has joined #openstack-ansible | 19:40 | |
*** sdake has joined #openstack-ansible | 19:47 | |
*** galstrom_zzz is now known as galstrom | 19:48 | |
*** galstrom is now known as galstrom_zzz | 19:48 | |
*** phiche has quit IRC | 19:49 | |
*** metral has quit IRC | 19:56 | |
*** javeriak has joined #openstack-ansible | 19:57 | |
*** hybridpollo has joined #openstack-ansible | 19:59 | |
*** javeriak_ has quit IRC | 20:00 | |
*** sdake_ has joined #openstack-ansible | 20:00 | |
Nepoc | cloudnull: okay... using phys results in lxc_conf - conf.c:instantiate_phys:2969 - failed to retrieve the index for br-egress | 20:00 |
Nepoc | for the neutron_agents container | 20:01 |
*** sdake has quit IRC | 20:03 | |
*** jthorne_ has joined #openstack-ansible | 20:05 | |
*** metral_zzz has joined #openstack-ansible | 20:05 | |
*** metral_zzz is now known as metral | 20:05 | |
*** gtt116 has joined #openstack-ansible | 20:06 | |
openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: [WIP] Add support for upgrading from Kilo to Liberty https://review.openstack.org/272182 | 20:07 |
*** jthorne has quit IRC | 20:08 | |
*** gtt116_ has quit IRC | 20:09 | |
*** javeriak has quit IRC | 20:11 | |
Nepoc | Okay I guess I should just ask it this way. How do I use eth0 for my router external gateway and for vlan 130? | 20:13 |
cloudnull | Nepoc: sorry on calls will be back in a bit | 20:16 |
Nepoc | cloudnull: okay no problem, will continue to attempt random things | 20:17 |
*** hybridpollo has quit IRC | 20:20 | |
*** oneswig has joined #openstack-ansible | 20:22 | |
*** oneswig has quit IRC | 20:26 | |
*** Mudpuppy has quit IRC | 20:27 | |
*** Mudpuppy has joined #openstack-ansible | 20:27 | |
cloudnull | Nepoc: would you mind sharing your interface configs ? | 20:28 |
cloudnull | Seems like you are going to need to bring your physical interfaces into the bridges and then vlan tag from there | 20:28 |
cloudnull | https://gist.github.com/cloudnull/e81115119dddee5e2a06 | 20:28 |
cloudnull | ^ thats an example of doing that with a single bond | 20:28 |
cloudnull | Bofu2U: put that together | 20:29 |
Nepoc | So this all works fine for the required bridges. But my simplest problem is, how do I get for example bond0 to be used as the router gateway in openstack. | 20:30 |
Nepoc | Next problem, how do I then add a network with vlan 130 on bond0 | 20:31 |
*** galstrom_zzz is now known as galstrom | 20:31 | |
Nepoc | I'll share my interfaces one sec | 20:31 |
Nepoc | http://pastebin.com/na6NibdB | 20:32 |
*** phalmos has joined #openstack-ansible | 20:36 | |
cloudnull | Nepoc: http://paste.openstack.org/show/485889/ | 20:36 |
cloudnull | so that would pull the vlan trunk all the way through br-vlan | 20:36 |
cloudnull | which is where all of the interfaces will be untagged | 20:37 |
cloudnull | br-vlan can be given to neutron and neutron can create networks with segmentation_ids (vlan tags) | 20:37 |
cloudnull | that essentially makes br-vlan w/ a bridge port of eth0 the gateway | 20:38 |
cloudnull | which is assuming your network carries a native vlan | 20:38 |
*** hybridpollo has joined #openstack-ansible | 20:38 | |
Nepoc | Okay I think I'm following | 20:39 |
cloudnull | the limitation is that the kernel will only untag traffic once | 20:39 |
*** olivier427 has joined #openstack-ansible | 20:39 | |
cloudnull | so you have to pull it all the way through to the interface where the kernel begins untagging traffic | 20:39 |
cloudnull | doing this creates weid looking interfaces (br-vlan.997) but it works quite well | 20:40 |
*** galstrom is now known as galstrom_zzz | 20:40 | |
Nepoc | So in openstack I would create a new network, type vlan, provider physical network would be br-vlan segmentation id 130 | 20:40 |
cloudnull | yes | 20:41 |
Nepoc | ok that works and then br-vlan flat for my external | 20:41 |
cloudnull | you'd want to extend br-vlan with a veth pair for the flat network but yes | 20:41 |
Nepoc | okay and that's in the example you provided before | 20:42 |
cloudnull | neutron will not let you have more than one network type on a single interface | 20:42 |
Nepoc | Well I'll try this out and be back soon, thank you again | 20:43 |
*** raddaoui_ has joined #openstack-ansible | 20:43 | |
cloudnull | example with the veth | 20:44 |
cloudnull | http://paste.openstack.org/show/485890/ | 20:44 |
cloudnull | which will create a "flat-external" interface name | 20:44 |
cloudnull | in your provider_networks array your flat network entry would be http://paste.openstack.org/show/485891/ | 20:47 |
cloudnull | which will wire your container to the flat-external network which is plugged into the br-vlan device | 20:47 |
Nepoc | thank you very much | 20:48 |
openstackgerrit | Hector I Gonzalez Mendoza proposed openstack/openstack-ansible-pip_install: Make Certificates validation configurable https://review.openstack.org/275918 | 20:52 |
*** phalmos has quit IRC | 21:03 | |
openstackgerrit | Merged openstack/openstack-ansible: Restart horizon when package/venvs change https://review.openstack.org/273810 | 21:03 |
*** galstrom_zzz is now known as galstrom | 21:05 | |
palendae | dolphm: From the keystone upgrade notes for liberty: "Several configuration options have been deprecated, renamed, or moved to new sections in the keystone.conf file." Where might I find that list, besides digging into keystone's git log? | 21:13 |
palendae | Actually http://docs.openstack.org/liberty/config-reference/content/keystone-conf-changes-liberty.html looks like what I wanted | 21:14 |
dolphm | palendae: if there's not a list published into the release notes... i'd start with stable/kilo's config file and look for things that are deprecated there, because they could be removed in stable/liberty | 21:14 |
dolphm | palendae: https://github.com/openstack/keystone/blob/stable/kilo/etc/keystone.conf.sample#L14 | 21:14 |
palendae | Thanks | 21:15 |
*** galstrom is now known as galstrom_zzz | 21:16 | |
*** olivier427 has quit IRC | 21:30 | |
*** abitha has joined #openstack-ansible | 21:33 | |
*** phiche has joined #openstack-ansible | 21:37 | |
*** CheKoLyN has quit IRC | 21:38 | |
*** jthorne_ has quit IRC | 21:45 | |
*** jthorne has joined #openstack-ansible | 21:45 | |
*** weezS has quit IRC | 21:47 | |
*** olivier427 has joined #openstack-ansible | 21:50 | |
*** CheKoLyN has joined #openstack-ansible | 21:53 | |
*** raddaoui_ has quit IRC | 22:03 | |
*** sdake_ has quit IRC | 22:06 | |
*** woodard_ has joined #openstack-ansible | 22:07 | |
*** woodard has quit IRC | 22:10 | |
*** oneswig has joined #openstack-ansible | 22:10 | |
*** woodard_ has quit IRC | 22:11 | |
*** oneswig has quit IRC | 22:15 | |
*** retreved has quit IRC | 22:16 | |
openstackgerrit | Michael Gugino proposed openstack/openstack-ansible-rsyslog_server: [WIP] Multi-distro support for rsyslog-server role https://review.openstack.org/274290 | 22:17 |
*** Mudpuppy has quit IRC | 22:26 | |
*** sdake has joined #openstack-ansible | 22:32 | |
*** CheKoLyN has quit IRC | 22:36 | |
*** sdake has quit IRC | 22:44 | |
*** phiche has quit IRC | 22:45 | |
*** CheKoLyN has joined #openstack-ansible | 22:49 | |
*** weezS has joined #openstack-ansible | 22:53 | |
*** TheIntern has joined #openstack-ansible | 22:58 | |
*** jthorne has quit IRC | 23:12 | |
*** baker has quit IRC | 23:14 | |
*** raddaoui_ has joined #openstack-ansible | 23:17 | |
*** jthorne has joined #openstack-ansible | 23:20 | |
*** michaelgugino has quit IRC | 23:23 | |
*** elopez has quit IRC | 23:26 | |
*** klamath has quit IRC | 23:27 | |
*** spotz is now known as spotz_zzz | 23:29 | |
*** sigmavirus24 is now known as sigmavirus24_awa | 23:30 | |
*** Bjoern has quit IRC | 23:33 | |
*** raddaoui_ has quit IRC | 23:43 | |
*** jamielennox|away is now known as jamielennox | 23:47 | |
logan- | is xtrabackup used anywhere besides galera containers? trying to figure out why galera_client pulls it in everywhere | 23:52 |
*** scarlisle has quit IRC | 23:53 | |
*** KLevenstein has quit IRC | 23:56 | |
*** TheIntern has quit IRC | 23:58 | |
*** oneswig has joined #openstack-ansible | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!