| *** openstack has joined #openstack-kolla | 17:04 | |
| sdake | wow no idea where it went | 17:05 |
|---|---|---|
| openstackgerrit | Steven Dake proposed openstack/kolla: Revert "Install master's version of an RDO keystone" https://review.openstack.org/299540 | 17:05 |
| sdake | I wonder if I ereverted soething else by accident | 17:06 |
| sdake | chekcing git logs | 17:06 |
| sdake | thanks inc0 for the catch - that would have been emberrassing :) | 17:07 |
| inc0 | yeah a bit | 17:08 |
| inc0 | trying to get keystone to work :S | 17:08 |
| *** jmccarthy has joined #openstack-kolla | 17:08 | |
| sdake | nihilifer around? | 17:09 |
| sdake | jpeeler around? | 17:10 |
| sdake | please vote on the backport and if you intend to vote on the core nomination please do ;) | 17:10 |
| * sdake wtb a voting service which sends the results automatically to the mailing list | 17:10 | |
| jpeeler | where's this vote? | 17:11 |
| sdake | polla] opentack-dev | 17:11 |
| *** daneyon has joined #openstack-kolla | 17:15 | |
| vhosakot | elemoine_: ping | 17:15 |
| *** stvnoyes has left #openstack-kolla | 17:17 | |
| inc0 | so sdake way I want to deal with backport is same as with upgrade | 17:18 |
| inc0 | once I'm done with ks let everyone take a service and make sure that it deploys both ways | 17:19 |
| inc0 | and upgrades from L to M | 17:19 |
| sdake | inco you need to do glance too | 17:19 |
| inc0 | we need to do all | 17:19 |
| vhosakot | lot of doc gate failures today | 17:19 |
| inc0 | but yeah | 17:19 |
| sdake | what i mean is if its distributed | 17:19 |
| sdake | we need the basics done | 17:19 |
| sdake | like rabbit/maria/keystone | 17:20 |
| sdake | i guess the rest can be distributed | 17:20 |
| inc0 | so here's deal | 17:20 |
| sdake | that wfm ;) | 17:20 |
| sdake | i like distributed backport idea | 17:20 |
| inc0 | I don't do maria and rabbit | 17:20 |
| *** daneyon has quit IRC | 17:20 | |
| inc0 | that we deploy from master | 17:20 |
| sdake | inc0 a distributed backport is off the table until mitaka ships though | 17:20 |
| inc0 | you are afraid for bandwidth? | 17:21 |
| sdake | inc0 we have 36 bugs | 17:21 |
| inc0 | kk | 17:21 |
| sdake | i am not sure which of them are actually crucial | 17:21 |
| sdake | but I'd like as many bugs fixed as possible | 17:21 |
| inc0 | fix these, there is good change that these will affect backport as well | 17:21 |
| sdake | right | 17:21 |
| sdake | lets get mitaka fixed | 17:21 |
| sdake | if you want to lay groundwork for backports by doing the base infra | 17:22 |
| sdake | that sounds good | 17:22 |
| sdake | and then we can distribute the workload | 17:22 |
| sdake | that is actually areally fantastic idea ;) | 17:22 |
| *** gfidente has quit IRC | 17:25 | |
| sdake | gotta jet, doctors appointment | 17:25 |
| sdake | then 2 hours of emetings | 17:25 |
| sdake | i may be out for 2-3 hours but will try to be responsive on irc as possible | 17:25 |
| *** unicell has joined #openstack-kolla | 17:29 | |
| *** ayoung has joined #openstack-kolla | 17:30 | |
| *** salv-orlando has joined #openstack-kolla | 17:33 | |
| *** salv-orlando has quit IRC | 17:40 | |
| *** rajathagasthya has joined #openstack-kolla | 17:41 | |
| vhosakot | sdake: the Member role should be added in Keystone bootstrap ? I see horizon does not have bootstraop | 17:41 |
| *** sdake has quit IRC | 17:42 | |
| *** unicell has quit IRC | 18:02 | |
| *** unicell has joined #openstack-kolla | 18:02 | |
| openstackgerrit | Michal Jastrzebski (inc0) proposed openstack/kolla: Making Mitaka deploying liberty https://review.openstack.org/299563 | 18:08 |
| *** Marga__ has quit IRC | 18:17 | |
| *** Slower_ has joined #openstack-kolla | 18:19 | |
| * Slower_ cries about the channel change *sniff* | 18:19 | |
| *** Slower_ is now known as Slower | 18:19 | |
| inc0 | Slower, nostalgy after #kolla? | 18:23 |
| Slower | yeah :) | 18:24 |
| inc0 | yeah, she will be misse | 18:26 |
| inc0 | d | 18:26 |
| inc0 | but on the other hand, we now have younger channel | 18:26 |
| *** mgoddard_ has joined #openstack-kolla | 18:28 | |
| *** mgoddard_ has quit IRC | 18:30 | |
| *** mgoddard_ has joined #openstack-kolla | 18:31 | |
| *** mgoddard has quit IRC | 18:31 | |
| *** mgoddard_ has quit IRC | 18:35 | |
| vhosakot | longer channel as well.... #openstack-kolla is longer than #kolla | 18:36 |
| *** Marga_ has joined #openstack-kolla | 18:42 | |
| *** shardy has quit IRC | 18:45 | |
| *** dwalsh has quit IRC | 18:49 | |
| *** sdake has joined #openstack-kolla | 18:57 | |
| sdake | vhosakot add a bootstrap to horizon then | 18:57 |
| sdake | manjeets any data for me ? :) | 18:58 |
| vhosakot | sdake: I was planning to add "OpenStack role create Member" in keystone's bootstrap in https://github.com/openstack/kolla/blob/master/docker/keystone/keystone_bootstrap.sh | 18:58 |
| sdake | nah | 18:58 |
| sdake | the member requirement comes from horiozn | 18:59 |
| sdake | not from keystone | 18:59 |
| sdake | lets double check in keystone channel now | 18:59 |
| manjeets | sdake working with heat stack | 18:59 |
| manjeets | not yet | 18:59 |
| inc0 | sdake, https://review.openstack.org/299563 | 19:00 |
| sdake | inc0 its that simple? | 19:01 |
| inc0 | question remains what breaks | 19:01 |
| inc0 | but it did deploy aio | 19:01 |
| manjeets | sdake I have this reported to track https://bugs.launchpad.net/kolla/+bug/1564008 | 19:01 |
| sdake | i think deploy_liberty should be determined y openstack_release | 19:01 |
| openstack | Launchpad bug 1564008 in kolla "Kolla should be able to handle connection limit for database according to infra " [Undecided,New] | 19:01 |
| vhosakot | sdake: I agree... since role is a keystone thing, I asked about it.... I see kolla-toolbox uses keystone_client to create new role https://github.com/openstack/kolla/blob/master/docker/kolla-toolbox/kolla_keystone_user.py#L69 | 19:02 |
| ccesario | Hey guys, I have a little doubt... today, after kolla deployment, if I want add a new controller node or compute node ... or replace a node ... is it possible ? | 19:02 |
| *** mgoddard has joined #openstack-kolla | 19:02 | |
| vhosakot | sdake: I agree your point that Member is horizon's requirement... | 19:02 |
| inc0 | ccesario, add new entry to inventory and re-run deploy | 19:02 |
| inc0 | you might need to run kolla-ansible reconfigure as well | 19:02 |
| inc0 | if you're adding controllers | 19:02 |
| rhallisey | ccesario, just need the node ip in the multinode inv | 19:02 |
| inc0 | so rabbitmq, keepalived and such will change conf | 19:03 |
| vhosakot | sdake: let me know what you think | 19:03 |
| *** salv-orlando has joined #openstack-kolla | 19:03 | |
| ccesario | all configs ... data... will be preserved!?! | 19:03 |
| ccesario | right!? | 19:03 |
| sdake | have meeting have to run | 19:03 |
| sdake | ccesario ye but reconfigure is broken atm | 19:04 |
| *** rajathagasthya has quit IRC | 19:04 | |
| sdake | jeffrey4l is working on it | 19:04 |
| vhosakot | does haproxy get updated when compute is added... I agree about haproxy and keepalived needs to update their confs when new controller is added | 19:04 |
| ccesario | thanks by infos guys!! | 19:04 |
| ccesario | I will try test it | 19:04 |
| vhosakot | sdake: after you come back, let me know... add horizon bootstrap with Member role or update keystone bootstrap ? | 19:05 |
| ccesario | I suppose that if I lost one node (compute/controller) due hardware problem by example.... It is only change the inventory pointing to other node and re-configure/deploy .... right!? | 19:07 |
| openstackgerrit | Mauricio Lima proposed openstack/kolla: Add precheck about nested environment https://review.openstack.org/299585 | 19:08 |
| manjeets | sdake i doubt steak-rg.yaml used by launch | 19:09 |
| sdake | manjeets it works for me ;) | 19:11 |
| sdake | manjeets try it unmodified | 19:11 |
| manjeets | okay | 19:11 |
| sdake | it just wont work with the way you have defiend networks | 19:11 |
| sdake | it works the way networks are defined by init-runonce | 19:11 |
| manjeets | defined networks works by other ways though | 19:12 |
| sdake | what i suggest is copy the network defn by init-runonce but make them 65k ips | 19:12 |
| sdake | and call the networks something else | 19:12 |
| *** mgoddard has quit IRC | 19:12 | |
| sdake | vhosakot add a bootstrap featue to horizon | 19:12 |
| sdake | vhosakot because apparently it needs one | 19:12 |
| vhosakot | cool.. done.. wanted to confirm before I start fixing cool | 19:13 |
| sdake | this is concllusion of openstack-keystone devs | 19:13 |
| vhosakot | you mean openstack-horizon devs right ? we spoke to in horiozn channel this morning | 19:13 |
| sdake | no keystone devs | 19:14 |
| vhosakot | ah ok | 19:14 |
| sdake | keystone devs said it doesn't make sens e to bootstrap horizon in keystone | 19:14 |
| sdake | but in horizon | 19:14 |
| vhosakot | got it | 19:14 |
| sdake | and only member is used by horizon | 19:14 |
| sdake | no other project | 19:14 |
| sdake | (at the moment) | 19:14 |
| vhosakot | yep.. ok | 19:14 |
| *** dwalsh has joined #openstack-kolla | 19:15 | |
| vhosakot | ccesario: yes, so, you would need to add the new node in the inventory...if you replace a bad node with same IP address, no need to update inventory also... | 19:18 |
| vhosakot | I meant, if you replace a bad node with a good server and retain the old node's IP, then, inventory need not be changed | 19:19 |
| ccesario | vhosakot, in case of replace bad node to new node .... which option need I execute ? reconfigure or deploy? | 19:20 |
| vhosakot | I want to say deploy, reconfigure as well if you want to augemtn/update configs on the new code | 19:21 |
| vhosakot | well, reconfigure for a _new_ node does not make sense as there will be no OpenStack serverice in the firsr place with old configs, right ? | 19:22 |
| vhosakot | so, replace bad server with good server, retain bad server's IP and hostname on good server, run deploy.... | 19:23 |
| ccesario | yes ... make sense. | 19:23 |
| ccesario | vhosakot, thanks again!!! | 19:23 |
| *** achanda has quit IRC | 19:24 | |
| vhosakot | ccesario: haha, let me know the outcome.. it a great test... what are you replacing ? control, compute, storage, deploy node (on which kolla-ansible is run) ? | 19:24 |
| ccesario | vhosakot, for while I 'm only trying understand the procedures..... Because my test env is based on virtual machines, I can try replace any node :) | 19:31 |
| vhosakot | inc0: I was reviewing your PS (Making Mitaka deploying liberty).... I see keystone_admin_token does not exist in Miataka... so, for liberty, keystone_admin_token must be set somewhere or asked in group_vars/all.yml when deploy_liberty=yes | 19:32 |
| vhosakot | ccesario: cool | 19:32 |
| vhosakot | inc0: I'll give this comment in the review | 19:32 |
| ccesario | other thing..... today is it possible make kolla deploy several environments/projects? .... because it read only /etc/kolla/(configs) .... and maybe /etc/kolla/projectname/(configs) would be needed ... | 19:35 |
| inc0 | vhosakot, yeah forgot about this one | 19:36 |
| inc0 | actually place for keytsone_admin_token is password.yml | 19:36 |
| vhosakot | inc0: Liberty has keystone_admin_token: "password" in etc/kolla/passwords.yml | 19:36 |
| inc0 | in mitaka we have generated password | 19:36 |
| inc0 | but I'll add it to passwords.yml | 19:37 |
| rhallisey | ccesario, yes use node_templates_directory | 19:37 |
| inc0 | with a comment | 19:37 |
| vhosakot | but keystone_admin_token is not passwords.yml in Mitaka... shou;dn't it be added dfirst | 19:37 |
| vhosakot | only to deploy Liberty case | 19:37 |
| rhallisey | ccesario, wrong var -> node_config_directory | 19:37 |
| rhallisey | use that one | 19:37 |
| rhallisey | in all.yml | 19:37 |
| vhosakot | cool | 19:38 |
| ccesario | rhallisey, but in case of multiple projects!? | 19:40 |
| vhosakot | ccesario: you mean multiple deploymetns/clouds ? | 19:40 |
| ccesario | yes | 19:40 |
| ccesario | :) | 19:40 |
| vhosakot | hmm | 19:41 |
| rhallisey | ccesario, just do that same thing you did with the first | 19:41 |
| rhallisey | and point that the new config dir | 19:41 |
| rhallisey | that variable* | 19:41 |
| rhallisey | ccesario, I don't see why it wouldn't work | 19:42 |
| ccesario | yes, changing it ansible/group_vars/all.yml in each deploy it works . | 19:43 |
| vhosakot | ccesario: are you using one deploy node for two clouds ? | 19:44 |
| ccesario | vhosakot, I 'm thinking about.... :) | 19:46 |
| ccesario | tests man | 19:46 |
| ccesario | :) | 19:46 |
| rhallisey | ccesario, I think it would still work | 19:47 |
| rhallisey | try it :) | 19:47 |
| vhosakot | hmmm... so, does that mean, you use the same kolla inputs (in globals.yml) for two clouds ? same IPs and VIPs for both clouds.. ? | 19:47 |
| vhosakot | yep.. try it :) great test... | 19:47 |
| *** mbound has joined #openstack-kolla | 19:48 | |
| ccesario | vhosakot, no no... what the rhallisey it would works.... but after the first deploy I will need chaange the "config" directory to read the new configs for new "cloud" | 19:48 |
| vhosakot | ah cool! | 19:49 |
| ccesario | the idea would be if kolla read automatically these configs :) | 19:50 |
| *** achanda has joined #openstack-kolla | 19:51 | |
| vhosakot | ccesario: so, would "kolla-ansible deploy" use the same globals.yml to deploy services on both clouds ? | 19:51 |
| ccesario | vhosakot, no ..... like "kolla-ansible deploy project1" "kolla-ansible deploy project2" and each deploy read the var in each directory config ... eg. /etc/kolla/project1/globals.yml and /etc/kolla/project2/globals.yml ..... | 19:54 |
| ccesario | only ideas... only :) | 19:54 |
| ccesario | with this, the operator/admin only touch in globals config ..... no changes in "core" code | 19:55 |
| rhallisey | ccesario, also need to change the inventory file | 19:55 |
| rhallisey | but right, no core code | 19:55 |
| ccesario | rhallisey, yes yes (I only put a short comment) | 19:55 |
| rhallisey | :) | 19:56 |
| ccesario | rhallisey, I think that you understood the idea :) | 19:57 |
| rhallisey | ccesario, ya I gotcha. Try it! Curious to see it work | 19:57 |
| vhosakot | rhallisey: ccesario: yes, I think it should work.. same _core_ code, different globals.yml files, different inventory files, _same_ deploy node, two clouds... nice | 20:00 |
| vhosakot | what about docker registry... will it be same for btoh clouds ? | 20:00 |
| *** sdake_ has joined #openstack-kolla | 20:00 | |
| rhallisey | vhosakot, ya | 20:01 |
| rhallisey | use the same images | 20:01 |
| rhallisey | only the configs change per service | 20:01 |
| vhosakot | yep.. right | 20:02 |
| *** sdake has quit IRC | 20:02 | |
| vhosakot | cool | 20:02 |
| ccesario | rhallisey, vhosakot maybe creating a kolla-ansible script wrapper to receive the params like "project" name and inventory can make this easy .... I think | 20:03 |
| vhosakot | but, two gloabls.yml files means _two_ docker_registry pram for kolla... | 20:05 |
| vhosakot | ccesario: I think keep docker_registry IP address same in both globals.yml so that the same registry is sued for both clouds | 20:07 |
| ccesario | vhosakot, necessarily the tw deploys don't need to be simultaneous ... | 20:08 |
| ccesario | well ... only ideas to make it more easy ... eheheh | 20:11 |
| *** rajathagasthya has joined #openstack-kolla | 20:13 | |
| vhosakot | yep | 20:13 |
| vhosakot | I think it should work... | 20:13 |
| ccesario | vhosakot, I will try test it | 20:17 |
| *** banix_ has joined #openstack-kolla | 20:17 | |
| *** banix has quit IRC | 20:19 | |
| *** banix_ is now known as banix | 20:19 | |
| ccesario | vhosakot, thinking in this.... the kolla-genpw could have the option to specify the output directory :) | 20:20 |
| sdake_ | inc0 can you get ahold of gammaha and let him know my schedule has changed | 20:23 |
| manjeets | sdake i think this solution to type of instances we want http://docs.openstack.org/developer/heat/template_guide/basic_resources.html#create-and-associate-security-groups-to-an-instance | 20:23 |
| sdake_ | inc0 and come to irc to chat | 20:23 |
| inc0 | manjeets, is Ganesh around? | 20:24 |
| sdake_ | manjeets the templates work a sis | 20:24 |
| sdake_ | if yuo can explain to me what precisely the problem is with them as they stnd i am open to listening ;) | 20:24 |
| manjeets | inc0: i see he is away | 20:25 |
| *** dwalsh has quit IRC | 20:27 | |
| *** sdake has joined #openstack-kolla | 20:32 | |
| gmmaha | inc0: hey | 20:33 |
| inc0 | gmmaha, sdake....talk;) | 20:33 |
| sdake | gmmaha hey my schedule changed | 20:33 |
| sdake | i am open now | 20:33 |
| gmmaha | sdake: ohhh | 20:33 |
| sdake | but may not be in about 1 hr or so | 20:33 |
| gmmaha | cool.. i am game | 20:33 |
| gmmaha | lets do it | 20:33 |
| *** sdake_ has quit IRC | 20:33 | |
| sdake | ok so lets get to it | 20:33 |
| sdake | can you setup some screen sharing so we can talk rather then type - will be faster | 20:34 |
| sdake | google groups or something | 20:34 |
| gmmaha | sdake: definitely can do | 20:34 |
| * gmmaha goes to get his google account | 20:34 | |
| *** achanda has quit IRC | 20:37 | |
| *** jtriley has quit IRC | 20:37 | |
| *** openstackgerrit has quit IRC | 20:48 | |
| *** openstackgerrit has joined #openstack-kolla | 20:48 | |
| manjeets | sdake https://github.com/openstack/kolla/blob/master/demos/heat/steak.yaml#L24 | 20:51 |
| *** daneyon has joined #openstack-kolla | 20:51 | |
| manjeets | i used defaults with updated network capacity still had issue with heat | 20:51 |
| manjeets | what is this key for | 20:52 |
| *** daneyon has quit IRC | 20:56 | |
| *** achanda has joined #openstack-kolla | 20:56 | |
| *** achanda has quit IRC | 20:57 | |
| *** achanda has joined #openstack-kolla | 20:58 | |
| *** daneyon has joined #openstack-kolla | 21:04 | |
| *** daneyon_ has quit IRC | 21:07 | |
| *** jtriley has joined #openstack-kolla | 21:11 | |
| *** prithiv has joined #openstack-kolla | 21:11 | |
| gmmaha | root@3b2a0f3f8615:/# dpkg -l | grep libvirt | 21:12 |
| gmmaha | ii libvirt-bin 1.3.1-1ubuntu4~cloud0 amd64 programs for the libvirt library | 21:12 |
| gmmaha | ii libvirt0:amd64 1.3.1-1ubuntu4~cloud0 amd64 library for interfacing with different virtualization systems | 21:12 |
| gmmaha | vhosakot: can you check if your libvirt containers are running the same versions of libvirt? | 21:12 |
| sdake | vhosakot can you erify the above is the same on our box? | 21:12 |
| *** ccesario_ has joined #openstack-kolla | 21:12 | |
| vhosakot | what command should I run ? | 21:12 |
| vhosakot | ah dpkg.. 1 sec | 21:12 |
| gmmaha | vhosakot: dpk -l | grep libvirt | 21:12 |
| gmmaha | dpkg -l | grep libvirt | 21:12 |
| sdake | inside container | 21:12 |
| gmmaha | root@3b2a0f3f8615:/# dpkg -l | grep libvirt | 21:13 |
| * manjeets saw a new minimum for sql connection today after re deploy | 21:13 | |
| manjeets | sdake may want that 492 | 21:14 |
| vhosakot | gmmaha: sdake: I dont have libvirt installed on the host.. on the host, "dpkg -l | grep libvirt" is empty....I have libvirt installed inside the nova_libvirt container.... | 21:15 |
| vhosakot | $ docker exec nova_libvirt dpkg -l | grep libvirt | 21:16 |
| vhosakot | ii libvirt-bin 1.3.1-1ubuntu4~cloud0 amd64 programs for the libvirt library | 21:16 |
| vhosakot | ii libvirt0:amd64 1.3.1-1ubuntu4~cloud0 amd64 library for interfacing with different virtualization systems | 21:16 |
| gmmaha | vhosakot: thanks.. | 21:16 |
| gmmaha | mathes | 21:16 |
| *** jtriley has quit IRC | 21:19 | |
| *** haplo37 has quit IRC | 21:19 | |
| vhosakot | do you have libvirt on the host ... ? it should be on the compute host... yes | 21:20 |
| *** ccesario_ has quit IRC | 21:23 | |
| *** ccesario__ has joined #openstack-kolla | 21:27 | |
| *** jtriley has joined #openstack-kolla | 21:30 | |
| *** rhallisey has quit IRC | 21:31 | |
| gmmaha | vhosakot: yeah i d o libvirt on the host.. will be disabling it soon-ish | 21:31 |
| *** ccesario__ has quit IRC | 21:31 | |
| vhosakot | cool | 21:31 |
| *** ccesario__ has joined #openstack-kolla | 21:31 | |
| *** achanda has quit IRC | 21:32 | |
| rajathagasthya | Has anyone seen this error before during kolla-build? "INFO:kolla.cmd.build:base:File contains no section headers. | 21:34 |
| rajathagasthya | INFO:kolla.cmd.build:base:file: file:///etc/yum.repos.d/delorean-deps.repo, line: 1 | 21:34 |
| rajathagasthya | INFO:kolla.cmd.build:base:'<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"\n'" | 21:34 |
| rajathagasthya | I'm in a Vagrant setup with Virtualbox provider. | 21:34 |
| *** ccesario__ has quit IRC | 21:38 | |
| gmmaha | vhosakot: how do you disable libvirt on ubuntu 14.04? | 21:38 |
| vhosakot | service libvirt-bin stop | 21:39 |
| vhosakot | update-rc.d libvirt-bin disable | 21:39 |
| gmmaha | vhosakot: sudo update-rc.d -f libvirtd remove | 21:42 |
| gmmaha | vhosakot: aah.. thansk | 21:42 |
| gmmaha | thanks | 21:42 |
| *** banix has quit IRC | 21:44 | |
| *** jtriley has quit IRC | 21:47 | |
| vhosakot | rajathagasthya: I'd ping manndre.. he is not around now.. ping SiRiuS_ if he is around | 21:47 |
| rajathagasthya | vhosakot: Thanks, will do. | 21:48 |
| vhosakot | rajathagasthya: you could refer http://docs.openstack.org/developer/kolla/quickstart.html to install kolla without vagrant... but, you'd need to create the VM youself | 21:49 |
| *** jtriley has joined #openstack-kolla | 21:51 | |
| rajathagasthya | vhosakot: I think it's not a Vagrant problem. It's something I encounter during kolla-build. | 21:52 |
| rajathagasthya | "ERROR:kolla.cmd.build:base:The command '/bin/sh -c rpm --import /etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7 && yum install -y epel-release yum-plugin-priorities && rpm --import /etc/pki/rpm-gpg/RPM-GPG-KEY-EPEL-7 && yum clean all' returned a non-zero code: 1" | 21:52 |
| vhosakot | ah, so vagrant is good ? | 21:52 |
| rajathagasthya | Yeah, vagrant is good. | 21:53 |
| vhosakot | just rebuild multiple times... I have seen some images faul to build if repos are unavailable... rebuild fixes it for me | 21:53 |
| *** jtriley has quit IRC | 21:56 | |
| *** dougs1 has joined #openstack-kolla | 21:57 | |
| manjeets | http://paste.openstack.org/show/492504/ | 21:59 |
| *** dougs has quit IRC | 22:00 | |
| vhosakot | manjeets: awesome! | 22:00 |
| gmmaha | https://bugs.launchpad.net/kolla/+bug/1564108 | 22:00 |
| openstack | Launchpad bug 1564108 in kolla "Add libvirt disable at prechecks" [Undecided,New] | 22:00 |
| * manjeets was warming up deployment horizon gave up | 22:00 | |
| vhosakot | manjeets: can you sent "neutron port-list | grep ip | wc -l" | 22:00 |
| manjeets | neutron port-list | grep ip | wc -l | 22:01 |
| manjeets | 1023 | 22:01 |
| *** mlima_ has joined #openstack-kolla | 22:01 | |
| vhosakot | one network ? | 22:01 |
| sdake | manjeets was that with heat? | 22:01 |
| inc0 | do we have meeting now? | 22:02 |
| manjeets | no horizon | 22:02 |
| sdake | inc0 in 1 hour bro | 22:02 |
| vhosakot | 58 more mins... DLS | 22:02 |
| inc0 | DLS | 22:02 |
| inc0 | -.- | 22:02 |
| manjeets | it was 1000 last night | 22:02 |
| sdake | we run on utc time in openstac k;-) | 22:02 |
| manjeets | sdake i see multiple ip issue need to debug that | 22:02 |
| sdake | manjeets did yu figureou thow to dleete 1000 nova vms quickly? | 22:03 |
| * manjeets will use re deploy | 22:03 | |
| manjeets | he gave up with heat temprarily | 22:03 |
| *** GonZo2000 has quit IRC | 22:03 | |
| manjeets | re deploy took 11 minutes | 22:04 |
| *** mlima has quit IRC | 22:05 | |
| *** GonZo2000 has joined #openstack-kolla | 22:05 | |
| *** GonZo2000 has joined #openstack-kolla | 22:05 | |
| inc0 | keystone is strange | 22:05 |
| inc0 | -.- | 22:05 |
| inc0 | I almost forgot how I used to hate it | 22:06 |
| vhosakot | gmmaha: I see you filed https://bugs.launchpad.net/kolla/+bug/1564108... | 22:06 |
| openstack | Launchpad bug 1564108 in kolla "Add libvirt disable at prechecks" [Undecided,New] | 22:06 |
| vhosakot | gmmaha: does precehcks pass for you when libvirt is enabled ? | 22:06 |
| openstackgerrit | Hui Kang proposed openstack/kolla: Add etcd Docker container https://review.openstack.org/298451 | 22:07 |
| manjeets | http://paste.openstack.org/show/492508/ | 22:08 |
| manjeets | some got 2 ips | 22:08 |
| gmmaha | vhosakot: yeah.. if i have stopped the service, it passes.. but on reboot it gets restarted again.. | 22:08 |
| gmmaha | would be nice to have it say 'go ahead and disable it with such commands' | 22:08 |
| *** jtriley has joined #openstack-kolla | 22:09 | |
| gmmaha | and ohh if you have cloud-init running the disable option that you shared or the remove option that i shared with update-rc.d doesnt work | 22:09 |
| gmmaha | i had to force un-isntalled libvirt-bin from the machines | 22:09 |
| vhosakot | gmmaha: wait... I want to understand more | 22:11 |
| vhosakot | gmmaha: does precehecks pass if libvirt is enabled ? | 22:12 |
| gmmaha | vhosakot: prechecks pass if libvirt is enabled and not running | 22:12 |
| vhosakot | ok | 22:12 |
| vhosakot | gmmaha: does precehecks pass if libvirt is running ? | 22:12 |
| sdake | meaning when you reboot, your deployment no longer works | 22:12 |
| gmmaha | vhosakot: no.. they wont pass if its running on the host | 22:12 |
| vhosakot | gmmaha: ah that is good... | 22:13 |
| vhosakot | gmmaha: the bug-fix would be flag an error if libvirt is enabled and not running ? | 22:13 |
| gmmaha | vhosakot: right.. | 22:13 |
| vhosakot | ok | 22:13 |
| gmmaha | if its even enabled, lets fail precheck and ask them to remove it | 22:14 |
| vhosakot | gmmaha: do you run prechecs after every reboot | 22:14 |
| sdake | vhosakot you shouldn't need to run prechecks after a reboot ;-) | 22:14 |
| gmmaha | vhosakot: no.. | 22:14 |
| *** jtriley has quit IRC | 22:15 | |
| vhosakot | gmmaha: what happens if you reboot _before_ deploying ? | 22:16 |
| vhosakot | gmmaha: is libvirt enabled again after reboot ? | 22:16 |
| gmmaha | vhosakot: libvirt would restart and deploy will happen, but VMs wont work | 22:16 |
| vhosakot | gmmaha: wait | 22:17 |
| gmmaha | vhosakot: right.. it gets re-enabled *if* you have cloud-init on the machine.. which i think they will have | 22:17 |
| *** ccesario_ has joined #openstack-kolla | 22:17 | |
| vhosakot | you reboot the compute node or deploy node ? | 22:17 |
| gmmaha | compute nodes.. | 22:18 |
| vhosakot | gmmaha: you deploy kolla, reboot compute node, VM's dont work.. correct ? | 22:18 |
| gmmaha | vhosakot: the cloud-init is a speculation.. i mgiht be wrong on why libvirt service gets restarted after reboot.. even after i disable using update-rc.d and even remove it | 22:19 |
| gmmaha | remove it == 'update-rc.d -f libvirt-bin remove' | 22:19 |
| vhosakot | and VMs work after removing ? | 22:20 |
| gmmaha | vhosakot: not sure yet.. | 22:21 |
| gmmaha | testing itnow | 22:21 |
| *** diogogmt_ has quit IRC | 22:21 | |
| *** vishwanathj has quit IRC | 22:27 | |
| *** mbound has quit IRC | 22:28 | |
| *** unicell has quit IRC | 22:32 | |
| *** unicell1 has joined #openstack-kolla | 22:32 | |
| *** rhallisey has joined #openstack-kolla | 22:35 | |
| sdake | even after removing libvirt, the shared object is still busted | 22:37 |
| sdake | i think the ld.so.cache may be busted | 22:37 |
| sdake | but hard to tell | 22:37 |
| sdake | containers = magic ;) | 22:37 |
| vhosakot | ha | 22:37 |
| *** rajathagasthya has quit IRC | 22:42 | |
| *** sdake has quit IRC | 22:46 | |
| *** sdake has joined #openstack-kolla | 22:46 | |
| *** rajathagasthya has joined #openstack-kolla | 22:48 | |
| gmmaha | vhosakot: should the kolla_toolbox have 'bin/sleep infinity' on it? the container seems to be up with that command | 22:50 |
| gmmaha | sdake: same error.. libvirt container is restarting | 22:51 |
| gmmaha | and the same libvirt-admin.so.0: cannot open shared object file: Permission denied | 22:51 |
| sdake | gamyes | 22:51 |
| sdake | gmmaha yes | 22:51 |
| sdake | gmmaha type dmessg and see if anytyhing pops up | 22:51 |
| gmmaha | sdake: aah ok.. (guessing yes is about kolla_toolbox) | 22:51 |
| sdake | yes toolbox | 22:52 |
| gmmaha | holy!! | 22:52 |
| gmmaha | http://paste.openstack.org/show/492513/ | 22:52 |
| vhosakot | kolla_toolbox has always had sleep infinity... it should not affect any containers... | 22:52 |
| sdake | ya apparmor | 22:53 |
| sdake | i suggested turning that off early on | 22:53 |
| gmmaha | sdake: but its not running and i thought i had turned it off. | 22:53 |
| sdake | clearly its not turned off ;-) | 22:53 |
| gmmaha | :) | 22:54 |
| *** SiRiuS_ has quit IRC | 22:54 | |
| * gmmaha goes to go through the ritual | 22:55 | |
| ntpttr | gmmaha: when did apparmor start running? was it installed when we had a working deployment? | 22:55 |
| gmmaha | ntpttr: not sure whne it got isntalled.. | 22:56 |
| sdake | i have heard reports kolla works with apparmor | 22:56 |
| gmmaha | i for one, definitely did not install it | 22:56 |
| sdake | apparmor is on by default | 22:56 |
| sdake | vhosakot do you run with apparmor? | 22:56 |
| vhosakot | no.. whatever is default | 22:56 |
| vhosakot | sudo apparmor_status | 22:57 |
| vhosakot | $ sudo apparmor_status | grep libvirt | 22:58 |
| *** Jeffrey4l has joined #openstack-kolla | 22:59 | |
| *** prithiv1 has joined #openstack-kolla | 23:00 | |
| inc0 | meeting time? | 23:00 |
| *** prithiv has quit IRC | 23:00 | |
| inc0 | or my clock lies to me again? | 23:00 |
| asalkeld_ | yeah | 23:00 |
| inc0 | (I don't trust this bastard any more) | 23:00 |
| asalkeld_ | haha | 23:00 |
| sdake | irc meeting now folks | 23:02 |
| sdake | feel free to join us on #openstack-meeting-4 | 23:02 |
| sdake | jpeeler rhallisey | 23:03 |
| gmmaha | vhosakot: http://paste.openstack.org/show/492514/ | 23:04 |
| vhosakot | dont know if it affects | 23:04 |
| gmmaha | vhosakot: it says its in enforce mode | 23:04 |
| gmmaha | http://paste.openstack.org/show/492515/ | 23:05 |
| *** blahRus has quit IRC | 23:05 | |
| *** prithiv has joined #openstack-kolla | 23:07 | |
| *** prithiv1 has quit IRC | 23:08 | |
| *** Jeffrey4l_ has joined #openstack-kolla | 23:15 | |
| gmmaha | vhosakot: for now i will go and remove apparmor for our testing | 23:16 |
| vhosakot | yes, good test... | 23:16 |
| *** Jeffrey4l has quit IRC | 23:16 | |
| vhosakot | although I doubt default apparmour affects OpenStack | 23:16 |
| *** banix has joined #openstack-kolla | 23:18 | |
| *** salv-orl_ has joined #openstack-kolla | 23:24 | |
| *** prithiv has quit IRC | 23:26 | |
| *** salv-orlando has quit IRC | 23:26 | |
| *** Jeffrey4l__ has joined #openstack-kolla | 23:32 | |
| *** Jeffrey4l__ is now known as Jeffrey4l | 23:32 | |
| *** Jeffrey4l_ has quit IRC | 23:33 | |
| *** prithiv has joined #openstack-kolla | 23:33 | |
| gmmaha | wooohoooo.. | 23:36 |
| vhosakot | ? | 23:36 |
| gmmaha | remove apparmor and things work now! | 23:36 |
| vhosakot | really ? | 23:36 |
| gmmaha | atleast libvirt container hasnt died | 23:37 |
| gmmaha | ohh yeah! | 23:37 |
| vhosakot | nova_libvirt container is up ? | 23:37 |
| * gmmaha goes to check the dashboard | 23:37 | |
| gmmaha | vhosakot: yeah | 23:37 |
| vhosakot | can you cehckt he VMs please ? | 23:37 |
| gmmaha | vhosakot: yeah starting to create the role and will spin up a VM | 23:37 |
| gmmaha | ohh hold on | 23:38 |
| gmmaha | deployment failed | 23:38 |
| gmmaha | http://paste.openstack.org/show/492517/ | 23:38 |
| vhosakot | re-deloy ? | 23:38 |
| gmmaha | vhosakot: yeah | 23:38 |
| gmmaha | doing that now | 23:38 |
| *** inc0 has quit IRC | 23:44 | |
| *** inc0 has joined #openstack-kolla | 23:44 | |
| *** achanda has joined #openstack-kolla | 23:45 | |
| *** Jeffrey4l_ has joined #openstack-kolla | 23:45 | |
| *** Jeffrey4l has quit IRC | 23:46 | |
| gmmaha | vhosakot: re-deployed nad still the same error.. | 23:47 |
| gmmaha | there is definnitely somethign wrong in the patches since the last time i had | 23:47 |
| gmmaha | where i could rdeploy consistently | 23:47 |
| * gmmaha goes to look at the patch log | 23:48 | |
| *** rajathagasthya has quit IRC | 23:50 | |
| *** rajathagasthya has joined #openstack-kolla | 23:53 | |
| inc0 | ok, I'm going to grab a scotch, have a fun evening guys | 23:55 |
| inc0 | I will | 23:55 |
| vhosakot | enjoy! :) | 23:55 |
| gmmaha | have fun inc0 | 23:55 |
| *** inc0 has quit IRC | 23:55 | |
| Jeffrey4l_ | sdake, could u explain why kolla-toolbox need root privilege? https://review.openstack.org/#/c/298479/7/docker/kolla-toolbox/Dockerfile.j2 | 23:56 |
| Jeffrey4l_ | vhosakot, ^ | 23:56 |
| openstackgerrit | Steven Dake proposed openstack/kolla: Fix gate by adding MariaDB-shared dependnecy https://review.openstack.org/299681 | 23:56 |
| sdake | needs workflow here plz | 23:56 |
| sdake | Jeffrey4l_ th patch thta merged removed that debug cruft | 23:57 |
| sdake | i needed root so i coudl debug the enironment temporarily | 23:58 |
| sdake | and accidentially git reviewed that atch | 23:58 |
| sdake | dont look at changeset 7 look at changeset 8 | 23:58 |
| Jeffrey4l_ | sdake, clearly. my bad. Thanks | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!