| *** adrian_otto has quit IRC | 00:00 | |
| *** mnaser has joined #openstack-kolla | 00:04 | |
| mnaser | has anyone done work around taking over an existing cloud with kolla (openstack components seem easy, but im curious around ceph) | 00:04 |
|---|---|---|
| *** kbaegis has joined #openstack-kolla | 00:04 | |
| SamYaple | mnaser: once because someone told me i couldn't do it | 00:06 |
| SamYaple | mnaser: don't recommend it | 00:06 |
| SamYaple | in fact, i dont recommend kolla ceph in production and I never have | 00:06 |
| SamYaple | its stable enough, but not flexible enough | 00:06 |
| mnaser | SamYaple: how is it not flexible enough? it seems to do the base that's needed | 00:07 |
| mnaser | SamYaple: we're doing a step by step "take-over" with kolla and we could do external ceph and move ceph eventually | 00:07 |
| SamYaple | would never recommend moving external ceph to a kolla-ceph deploy | 00:08 |
| SamYaple | i specifically said when i wrote it that it was for small labs and such | 00:08 |
| mnaser | i guess we could stick to our ansible stuff to maintain ceph | 00:09 |
| SamYaple | its not flexible in the sense that its hard to do proper maintenance on things where you might need to fix a journal since the disk died and such | 00:09 |
| mnaser | i guess, in a larger environment you can just scrap the osd and rebuild (at least in our environment, we're okay with this) | 00:09 |
| mnaser | but i see your concerns | 00:09 |
| *** adrian_otto has joined #openstack-kolla | 00:09 | |
| *** gfhellma has joined #openstack-kolla | 00:10 | |
| SamYaple | i mean you *can*. you shouldn't neccesarily do that | 00:10 |
| SamYaple | i just last week recovered a journal in a 1.5PB environment rather than balance out the 4TB drive | 00:10 |
| SamYaple | balancing out the 4TB OSD would have moved like 300TB of data | 00:11 |
| mnaser | jeez | 00:11 |
| SamYaple | granted that cluster is on early subset of the tuneables, but still | 00:11 |
| SamYaple | just because you can doesnt me you should | 00:11 |
| mnaser | we'll see, right now i'm pretty confident in us getting the openstack control plane moved | 00:11 |
| *** sayantan_ has joined #openstack-kolla | 00:12 | |
| mnaser | starting with keystone in the first few days | 00:12 |
| SamYaple | moving the control plane is a great start | 00:13 |
| SamYaple | compute is harder to do, but easily doable with manual intervention | 00:13 |
| mnaser | the only annoyance is that our internal network is the same as the default docker bridge network | 00:13 |
| mnaser | (172.17/16) | 00:14 |
| mnaser | so that's something that might be annoying in the bootstrap process | 00:14 |
| SamYaple | since kolla-ansible doesn't use docker bridge network that network should never come up | 00:14 |
| SamYaple | so no interference | 00:14 |
| SamYaple | then you just use the systemd daemon to set a different network for safety reasons | 00:15 |
| mnaser | SamYaple im fairly sure the docker0 bridge goes up when teh service is started | 00:15 |
| mnaser | and grabs and ip too | 00:15 |
| SamYaple | mnaser: nope. docker0 is down unless there a a container using the docker bridge network | 00:16 |
| mnaser | yup, that's what we'll have to do, kolla has been spoiling us a bit | 00:16 |
| mnaser | oh | 00:16 |
| mnaser | you are right | 00:16 |
| mnaser | it is listed as down | 00:16 |
| mnaser | my bad | 00:16 |
| SamYaple | it will still respond to pings on 172.17.0.1 though | 00:16 |
| mnaser | 172.17.0.0/16 dev docker0 proto kernel scope link src 172.17.0.1 | 00:17 |
| mnaser | the route does get inserted | 00:17 |
| mnaser | which will break network connectivity | 00:17 |
| mnaser | its just a systemd drop in but kolla spoiled us a bit in not having to do a lot :-P | 00:17 |
| SamYaple | depends on ordering | 00:17 |
| *** manheim has quit IRC | 00:18 | |
| SamYaple | but anyway. back to the original point, kolla ceph stuff is meant for fewer than 100 osds (it can't actually do more than that based on naming conventions) and is good for small labs and clusters | 00:18 |
| *** adrian_otto has quit IRC | 00:18 | |
| SamYaple | its hard to call it production | 00:19 |
| mnaser | i guess in that case we'll continue to maintain "external ceph" and see how things go from there | 00:19 |
| SamYaple | i had run it longer than anyone and even i stopped using it (converted to bare-metal ceph) | 00:19 |
| mnaser | docker on compute nodes is something that i am personally a bit ~bleh~ about to be honest | 00:19 |
| SamYaple | gotta have imaging | 00:20 |
| SamYaple | without it no rollback | 00:20 |
| SamYaple | luckily it is not performance impacting at all | 00:20 |
| SamYaple | VMs are run in the host namepsace and the same cgroups as if they were launched from the host | 00:20 |
| SamYaple | 100% same performance | 00:20 |
| mnaser | oh yeah i understand that but its just | 00:21 |
| mnaser | its not what we're used to :) | 00:21 |
| mnaser | we might work on getting monitoring based on sensu added at some point | 00:21 |
| mnaser | or help around an effort surrounding taht | 00:21 |
| *** sbezverk has joined #openstack-kolla | 00:23 | |
| *** yingjun has joined #openstack-kolla | 00:31 | |
| *** sbezverk has quit IRC | 00:32 | |
| *** sbezverk has joined #openstack-kolla | 00:34 | |
| *** manheim has joined #openstack-kolla | 00:37 | |
| *** sbezverk has quit IRC | 00:45 | |
| *** duonghq has joined #openstack-kolla | 00:54 | |
| duonghq | morning guys | 00:54 |
| *** dave-mccowan has joined #openstack-kolla | 00:59 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla master: [wip] Install panko in ceilometer container https://review.openstack.org/444680 | 01:05 |
| *** daidv has joined #openstack-kolla | 01:05 | |
| *** jtriley has joined #openstack-kolla | 01:06 | |
| *** cuongnv has joined #openstack-kolla | 01:08 | |
| duonghq | http://logs.openstack.org/61/444561/4/check/gate-kolla-ansible-dsvm-deploy-ubuntu-source-ubuntu-xenial-nv/e2999ea/console.html#_2017-03-12_06_55_40_351982 seem that we got ara error again? | 01:09 |
| *** zhurong has joined #openstack-kolla | 01:11 | |
| *** qiliang27 has joined #openstack-kolla | 01:16 | |
| *** zhubingbing has joined #openstack-kolla | 01:16 | |
| duonghq | morning zhubingbing | 01:16 |
| zhubingbing | morning duonghq | 01:17 |
| zhubingbing | ;) | 01:17 |
| *** Margin has joined #openstack-kolla | 01:17 | |
| *** schwicht has quit IRC | 01:24 | |
| *** schwicht has joined #openstack-kolla | 01:25 | |
| spsurya | duonghq: morning | 01:32 |
| spsurya | zhubingbing: | 01:32 |
| spsurya | mrng | 01:32 |
| duonghq | sup spsurya | 01:33 |
| zhubingbing | morning | 01:34 |
| spsurya | hw was your weekend guys ? | 01:38 |
| *** manheim has quit IRC | 01:38 | |
| *** salv-orlando has joined #openstack-kolla | 01:41 | |
| *** salv-orlando has quit IRC | 01:45 | |
| mnaser | is there an upgrade path in place that updates deployments from heka to fluentd | 01:48 |
| mnaser | as we go newton -> ocata | 01:48 |
| *** schwicht has quit IRC | 01:52 | |
| SamYaple | mnaser: there should always be an upgrade path for things marked stable | 01:55 |
| SamYaple | if its not implemented, its a bug | 01:55 |
| duonghq | mnaser, zhubingbing is the guys who worked on fluentd | 02:01 |
| zhubingbing | yes | 02:01 |
| zhubingbing | there should always be an upgrade path for things marked stable | 02:01 |
| zhubingbing | agree sample | 02:02 |
| duonghq | zhubingbing, mnaser wants to know how we migrate from heka to fluentd when upgrade takes place | 02:02 |
| zhubingbing | to solve a lot of log collection problems | 02:03 |
| zhubingbing | but the basic function is the same as heka | 02:04 |
| Margin | Hi , I opened a bug https://bugs.launchpad.net/kolla/+bug/1672240 | 02:06 |
| openstack | Launchpad bug 1672240 in kolla "destroy tool is incomplete " [Undecided,New] | 02:06 |
| *** daidv has quit IRC | 02:06 | |
| *** MarginHu has joined #openstack-kolla | 02:07 | |
| *** Margin has quit IRC | 02:08 | |
| zhubingbing | hi Margin | 02:08 |
| zhubingbing | destroy tool is incomplete ? what's mean ? | 02:08 |
| duonghq | zhubingbing, imo, high and critical bug should always be targeted to some milestone | 02:18 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla master: Install panko in ceilometer base container https://review.openstack.org/444680 | 02:19 |
| zhubingbing | em | 02:19 |
| zhubingbing | i will fix this bug | 02:19 |
| zhubingbing | now | 02:19 |
| duonghq | zhubingbing, nice | 02:24 |
| MarginHu | zhubingbing , maybe I'm wrong | 02:27 |
| zhubingbing | don't remove /var/lib/mariadb ? | 02:28 |
| MarginHu | I removed the /var/lib/docker/volumes | 02:31 |
| MarginHu | but also report error | 02:31 |
| MarginHu | I update the bug , please check it . | 02:31 |
| zhubingbing | OK | 02:33 |
| kbaegis | Anyone know where the diskimage builder output folder is? | 02:33 |
| MarginHu | I know | 02:34 |
| MarginHu | it is "pwd" | 02:34 |
| *** Raj12 has joined #openstack-kolla | 02:34 | |
| kbaegis | MarginHu: As in where I ran the command from? Doesn't appear to be true | 02:34 |
| Raj12 | I have deployed multinode newton with kolla, now I need to do some changes in ml2.conf.ini how can I deploy the changes | 02:35 |
| Raj12 | I have tried adding config in /kolla/config but it is not changing the configure if I do reconfigure | 02:35 |
| MarginHu | kbaegis, yes | 02:36 |
| kbaegis | Raj12: Best strategy is to set COPY ALWAYS in globals and just restart the container | 02:36 |
| Raj12 | Not sure how to set COPY ALWAYS is there any doc? | 02:38 |
| kbaegis | Raj12: It's in /etc/kolla/globals | 02:38 |
| kbaegis | 'config_strategy: "COPY_ALWAYS"' | 02:39 |
| Raj12 | where I need to keep the changes? | 02:39 |
| kbaegis | and then docker restart nova_api or whatever | 02:39 |
| kbaegis | Raj12: /etc/kolla/globals.yml | 02:39 |
| MarginHu | I build ironic-agent image with disk-image-builder . | 02:40 |
| kbaegis | It's not outputting into my current directory | 02:40 |
| Raj12 | added 'config_strategy: "COPY_ALWAYS" and config in /etc/koll/connfig | 02:45 |
| Raj12 | restarted container still no luck | 02:45 |
| Raj12 | it has same old config nothing changed | 02:45 |
| Raj12 | @kbaegis Thanks it worked now. | 02:49 |
| *** Raj12 has quit IRC | 02:59 | |
| kbaegis | h2h | 03:00 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla master: Load iscsi_tcp module when starting ironic-conductor https://review.openstack.org/444699 | 03:32 |
| *** dave-mccowan has quit IRC | 03:32 | |
| *** adrian_otto has joined #openstack-kolla | 03:32 | |
| openstackgerrit | yunfeng zhou proposed openstack/kolla-ansible master: change log-* to flog-*. https://review.openstack.org/444700 | 03:34 |
| openstackgerrit | yunfeng zhou proposed openstack/kolla-ansible master: change log-* to flog-*. https://review.openstack.org/444700 | 03:41 |
| *** adrian_otto has quit IRC | 03:41 | |
| *** salv-orlando has joined #openstack-kolla | 03:42 | |
| *** salv-orlando has quit IRC | 03:46 | |
| *** adrian_otto has joined #openstack-kolla | 03:49 | |
| *** sayantan_ has quit IRC | 03:51 | |
| *** adrian_otto has quit IRC | 03:55 | |
| *** sayantan_ has joined #openstack-kolla | 03:59 | |
| *** sayantan_ has quit IRC | 04:04 | |
| openstackgerrit | Surya Prakash (spsurya) proposed openstack/kolla-ansible master: GerritReview check https://review.openstack.org/444705 | 04:14 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla master: Load iscsi_tcp module when starting ironic-conductor https://review.openstack.org/444699 | 04:20 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible master: Mount /lib/modules folder in ironic-conductor container https://review.openstack.org/444708 | 04:23 |
| *** mdnadeem has joined #openstack-kolla | 04:25 | |
| *** jtriley has quit IRC | 04:34 | |
| *** sayantan_ has joined #openstack-kolla | 04:38 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla master: Install panko in ceilometer base container https://review.openstack.org/444680 | 04:38 |
| *** sayantan_ has quit IRC | 04:42 | |
| *** sudipto has joined #openstack-kolla | 04:45 | |
| *** sayantan_ has joined #openstack-kolla | 04:50 | |
| *** sayantan_ has quit IRC | 04:55 | |
| *** ssurana has joined #openstack-kolla | 05:11 | |
| *** japestinho has quit IRC | 05:20 | |
| *** skramaja has joined #openstack-kolla | 05:23 | |
| *** trinaths has joined #openstack-kolla | 05:23 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible master: Add panko.conf file in ceilometer container https://review.openstack.org/444719 | 05:28 |
| *** MarginHu has quit IRC | 05:38 | |
| *** MarginHu has joined #openstack-kolla | 05:39 | |
| *** MarginHu has quit IRC | 05:42 | |
| *** salv-orlando has joined #openstack-kolla | 05:43 | |
| *** salv-orlando has quit IRC | 05:47 | |
| *** unicell has joined #openstack-kolla | 05:57 | |
| *** unicell has quit IRC | 05:59 | |
| *** tovin07 has joined #openstack-kolla | 06:00 | |
| *** salv-orlando has joined #openstack-kolla | 06:02 | |
| *** unicell has joined #openstack-kolla | 06:04 | |
| *** tovin07_ has joined #openstack-kolla | 06:13 | |
| *** unicell has quit IRC | 06:20 | |
| *** Mannheim has joined #openstack-kolla | 06:31 | |
| *** jrobinson has quit IRC | 06:36 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible master: Add panko.conf file in ceilometer container https://review.openstack.org/444719 | 06:40 |
| *** jrobinson has joined #openstack-kolla | 06:49 | |
| *** bmace has quit IRC | 06:54 | |
| *** bmace has joined #openstack-kolla | 06:54 | |
| *** MarginHu has joined #openstack-kolla | 07:06 | |
| *** satyar has joined #openstack-kolla | 07:09 | |
| *** saneax-_-|AFK is now known as saneax | 07:17 | |
| spsurya | duonghq: | 07:20 |
| spsurya | ping... | 07:20 |
| duonghq | spsurya hey there | 07:22 |
| *** rabel has joined #openstack-kolla | 07:22 | |
| spsurya | what you think about this https://bugs.launchpad.net/kolla/+bug/1576794 | 07:23 |
| openstack | Launchpad bug 1576794 in kolla "drop root for keystone" [Critical,In progress] - Assigned to Surya Prakash Singh (confisurya) | 07:23 |
| spsurya | what can be better way to fix | 07:23 |
| spsurya | this | 07:23 |
| spsurya | ? | 07:23 |
| *** salv-orl_ has joined #openstack-kolla | 07:24 | |
| duonghq | not sure, but iirc, we have some issue relate to apache if we drop root for keystone | 07:24 |
| duonghq | due to Keystone use WSGI for this interface | 07:24 |
| duonghq | not just add USER keystone resolve issue | 07:25 |
| spsurya | duonghq: any way we should not remove root previlege from keystone | 07:25 |
| duonghq | if somebody can prove we cannot drop root from keystone, why can mark this bug is invalid IMO, | 07:26 |
| spsurya | I was thinking same. | 07:26 |
| *** salv-orlando has quit IRC | 07:26 | |
| duonghq | "should not" is not strong evidence | 07:26 |
| duonghq | one possible outcome is" comment in docker file why we do not drop root from keystone | 07:27 |
| duonghq | it'll prevent somebody raise this issue again | 07:27 |
| spsurya | duonghq: suppose a root user deploy OpenStack, and he doesn't want that any other person would create any new credential i.e any normal USER | 07:29 |
| duonghq | spsurya, it's not a rational reason | 07:30 |
| spsurya | duonghq: and if we give root previlege to any USER then he can also distribute other credential and can list all the credential, private cloud functionality would be damaged | 07:32 |
| *** manheim has joined #openstack-kolla | 07:32 | |
| duonghq | spsurya, sorry, misunderstand your last sentence, | 07:32 |
| *** salv-orl_ has quit IRC | 07:33 | |
| *** jrobinson has quit IRC | 07:33 | |
| duonghq | can you explain it again? | 07:33 |
| spsurya | duonghq: i mean any user can create credential as root does. | 07:34 |
| spsurya | duonghq: I think inc0 has filed this bug | 07:34 |
| spsurya | duonghq: may be he can add somthing into this | 07:34 |
| *** manheim has quit IRC | 07:36 | |
| duonghq | hmm, | 07:38 |
| duonghq | I think we should raise it again in this week meeting? | 07:38 |
| *** manheim has joined #openstack-kolla | 07:39 | |
| spsurya | duonghq: sure | 07:39 |
| spsurya | duonghq: we should | 07:39 |
| duonghq | spsurya, can you add this to the agenda | 07:40 |
| duonghq | seem that this week we will have many topic | 07:40 |
| spsurya | yeh i am adding | 07:40 |
| spsurya | duonghq: | 07:40 |
| *** MarginHu has quit IRC | 07:45 | |
| *** MarginHu has joined #openstack-kolla | 07:46 | |
| spsurya | duonghq: done | 07:47 |
| *** MarginHu has quit IRC | 07:48 | |
| *** matrohon has joined #openstack-kolla | 07:50 | |
| *** japestinho has joined #openstack-kolla | 07:55 | |
| *** salv-orlando has joined #openstack-kolla | 08:04 | |
| *** tovin07 has quit IRC | 08:07 | |
| *** kwazar has joined #openstack-kolla | 08:09 | |
| *** zioproto has joined #openstack-kolla | 08:10 | |
| *** ssurana has quit IRC | 08:11 | |
| *** tovin07 has joined #openstack-kolla | 08:12 | |
| *** dciabrin has joined #openstack-kolla | 08:15 | |
| *** Serlex has joined #openstack-kolla | 08:16 | |
| *** openstackgerrit has quit IRC | 08:18 | |
| *** pcaruana has joined #openstack-kolla | 08:25 | |
| *** MarginHu has joined #openstack-kolla | 08:38 | |
| *** egonzalez has joined #openstack-kolla | 08:46 | |
| MarginHu | hi, how to cleanup all related info about mariadb which is generated in last deployment ? | 08:56 |
| *** openstackgerrit has joined #openstack-kolla | 08:57 | |
| openstackgerrit | Eduardo Gonzalez proposed openstack/kolla stable/newton: Sahara needs to run ssh-keygen https://review.openstack.org/444768 | 08:57 |
| *** spsurya__ has joined #openstack-kolla | 08:59 | |
| egonzalez | MarginHu, cleanup in which way? remove containers, config files, images, volumes? | 09:00 |
| MarginHu | yes , I met a mariadb galera issue when redeploy. | 09:01 |
| egonzalez | when executing cleanup-containers mariadb will remove mariadb container and volumes | 09:02 |
| egonzalez | config files aren't an issue when redeploying, since them are replaced by new files, but can be removed with cleanup-host | 09:03 |
| egonzalez | both in tools/ folder | 09:03 |
| *** trinaths1 has joined #openstack-kolla | 09:04 | |
| *** trinaths has quit IRC | 09:05 | |
| MarginHu | you're right, thanks | 09:05 |
| egonzalez | MarginHu, in the bug opened by the same issue, how you make the destroy? through kolla-ansible destroy? | 09:08 |
| MarginHu | but iptable rules | 09:08 |
| MarginHu | yes , kolla-ansible destroy | 09:09 |
| egonzalez | newton, ocata or master? | 09:09 |
| MarginHu | ocata | 09:09 |
| egonzalez | hrm, the mariadb issue shouldn't happen since that role removes containers and volumes | 09:10 |
| MarginHu | how to troube-shoot the issue ? | 09:11 |
| openstackgerrit | Dave Walker proposed openstack/kolla-ansible master: Remove heka_socket vol and unwedge some containers https://review.openstack.org/444771 | 09:13 |
| *** spsurya__ has quit IRC | 09:17 | |
| *** mgoddard has joined #openstack-kolla | 09:18 | |
| *** gfidente has joined #openstack-kolla | 09:24 | |
| *** athomas has joined #openstack-kolla | 09:24 | |
| *** zhubingbing has quit IRC | 09:24 | |
| *** zhubingbing has joined #openstack-kolla | 09:26 | |
| openstackgerrit | Dave Walker proposed openstack/kolla-ansible master: Remove heka_socket vol and unwedge some containers https://review.openstack.org/444771 | 09:29 |
| egonzalez | MarginHu, i'm seeing the issue, when executing cleanup-container <container-name> instead of cleanup-container, volumes are not properly evaluated hence not removed | 09:30 |
| egonzalez | zhubingbing, ^^ | 09:33 |
| zhubingbing | hi | 09:33 |
| zhubingbing | yes | 09:33 |
| zhubingbing | egonzalez | 09:33 |
| egonzalez | what i don't know yet is why destroy role does not remove those volumes, cleanup-containers without regex works fine | 09:34 |
| openstackgerrit | Thomas O'Neill proposed openstack/kolla master: Add neutron-bgp-dragent service. https://review.openstack.org/443732 | 09:35 |
| *** matrohon has quit IRC | 09:35 | |
| egonzalez | zhubingbing, got what happen with cleanup-containers https://github.com/openstack/kolla-ansible/blob/master/tools/cleanup-containers#L15 | 09:36 |
| *** matrohon has joined #openstack-kolla | 09:37 | |
| egonzalez | may notice there are ($( and the double )) at the end of the var, removing the first ( fixes the regex issue) | 09:37 |
| zhubingbing | 7/5000 | 09:38 |
| zhubingbing | Wǒ néng chóng xiàn tā mā | 09:38 |
| zhubingbing | can i reproduce it ? | 09:38 |
| egonzalez | what? hehe | 09:39 |
| MarginHu | I think you can reproduce it . | 09:39 |
| *** tovin07 has quit IRC | 09:39 | |
| zhubingbing | egonzalez | 09:40 |
| egonzalez | MarginHu, btw, before using destroy role, have you removed mariadb container? | 09:40 |
| MarginHu | no | 09:41 |
| zhubingbing | egonzalez i go to fix this bug | 09:41 |
| egonzalez | zhubingbing, yep | 09:41 |
| zhubingbing | ；） | 09:42 |
| zhubingbing | i thinks there volume don‘t remove ；） | 09:43 |
| *** hrw has quit IRC | 09:44 | |
| *** athomas has quit IRC | 09:45 | |
| *** athomas has joined #openstack-kolla | 09:45 | |
| *** hrw has joined #openstack-kolla | 09:45 | |
| *** yingjun has quit IRC | 09:47 | |
| egonzalez | MarginHu, zhubingbing I can't reproduce the issue with mariadb volume using destroy role | 09:48 |
| egonzalez | all kolla volumes are removed | 09:48 |
| zhubingbing | me too | 09:48 |
| zhubingbing | i can't reproduce this bug | 09:48 |
| egonzalez | the only issue i found is using regex in cleanup-containers, but destroy role does not use that | 09:48 |
| zhubingbing | i will talk with MarginHu | 09:49 |
| kbaegis | Hey, anyone using barbican/senlin/magnum? | 09:49 |
| egonzalez | kbaegis, in prod or testing? | 09:49 |
| kbaegis | testing | 09:49 |
| kbaegis | CertificatesToClusterFailed: Failed to create certificates for Cluster | 09:49 |
| egonzalez | I use senlin and magnum but not barbican | 09:50 |
| kbaegis | barbican logs talking about declaring loyalty to the emperor or some bs :) | 09:50 |
| kbaegis | gotcha | 09:50 |
| zhubingbing | i use magnum and barbicna | 09:51 |
| kbaegis | zhubingbing: Did you have to set anything up for certificates? | 09:51 |
| zhubingbing | https://review.openstack.org/#/c/444561/ | 09:51 |
| zhubingbing | this patch | 09:51 |
| zhubingbing | i fix it | 09:51 |
| zhubingbing | yes have some problem barbican | 09:51 |
| kbaegis | tyvm | 09:52 |
| kbaegis | You're awesome | 09:52 |
| kbaegis | So after patch, I have to pip install -U ../kolla-ansible | 09:53 |
| kbaegis | do I have to rebuild the barbican images? | 09:53 |
| kbaegis | And do I have to run reconfigure? | 09:53 |
| zhubingbing | don；t rebuild images | 09:53 |
| kbaegis | okay | 09:53 |
| duonghq | egonzalez, can you review zhubingbing ps: https://review.openstack.org/#/c/444561/ | 09:54 |
| zhubingbing | your kolla version coat？ | 09:54 |
| *** MarginHu has quit IRC | 09:54 | |
| egonzalez | duonghq, done | 09:54 |
| kbaegis | stable/ocata | 09:54 |
| zhubingbing | ocata | 09:54 |
| *** MarginHu has joined #openstack-kolla | 09:54 | |
| *** MarginHu has quit IRC | 09:55 | |
| openstackgerrit | zhubingbing proposed openstack/kolla-ansible master: Fix ERROR barbican.plugin can't find /usr/lib/libCryptoki2_64.so https://review.openstack.org/444561 | 09:56 |
| zhubingbing | egonzalez done | 09:56 |
| duonghq | zhubingbing, you have not resolved 2nd commend of egonzalez | 09:58 |
| duonghq | *comment | 09:58 |
| *** deadnull has quit IRC | 09:58 | |
| zhubingbing | i don't know how to solve it | 10:00 |
| duonghq | just move those lines down to Barbican section ;) | 10:00 |
| *** cuongnv has quit IRC | 10:00 | |
| zhubingbing | ok | 10:01 |
| duonghq | just above crypto_plugin line | 10:01 |
| zhubingbing | ;) | 10:01 |
| openstackgerrit | Merged openstack/kolla master: Sahara needs to run ssh-keygen https://review.openstack.org/436869 | 10:01 |
| egonzalez | btw zhubingbing fix the regex issue in cleanup-containers if you want ;) | 10:02 |
| *** caowei has joined #openstack-kolla | 10:02 | |
| zhubingbing | i have talk with MarginHu | 10:03 |
| openstackgerrit | zhubingbing proposed openstack/kolla-ansible master: Fix ERROR barbican.plugin can't find /usr/lib/libCryptoki2_64.so https://review.openstack.org/444561 | 10:03 |
| zhubingbing | maybe he Is his fault | 10:03 |
| *** pbourke has joined #openstack-kolla | 10:03 | |
| *** satyar has quit IRC | 10:04 | |
| duonghq | zhubingbing, so if he can confirm here, we can close this bug | 10:06 |
| duonghq | I think regex issue is separate bug | 10:06 |
| egonzalez | duonghq, yep, is different issue | 10:07 |
| *** mnasiadka has joined #openstack-kolla | 10:07 | |
| *** rmart04 has joined #openstack-kolla | 10:16 | |
| *** zhurong has quit IRC | 10:21 | |
| *** zhurong has joined #openstack-kolla | 10:22 | |
| *** nathharp has joined #openstack-kolla | 10:25 | |
| *** nathharp has quit IRC | 10:26 | |
| *** rmart04 has quit IRC | 10:28 | |
| *** duonghq has quit IRC | 10:29 | |
| *** tovin07_ has quit IRC | 10:29 | |
| zhubingbing | yes <duonghq | 10:33 |
| zhubingbing | yes duonghq | 10:33 |
| zhubingbing | i have confirm is separate bug | 10:34 |
| zhubingbing | ；） | 10:34 |
| openstackgerrit | Nathan Harper proposed openstack/kolla-ansible master: Nova external-ceph.yml writes keyname into libvirt.base64 Closes-Bug: #1671862 Change-Id: I6b77773238cd65a3cc009d8d9a38bdbea0035212 https://review.openstack.org/444389 | 10:34 |
| openstack | bug 1671862 in kolla "Nova external-ceph.yml writes keyname into libvirt .base64 instead of value" [High,Triaged] https://launchpad.net/bugs/1671862 - Assigned to Nathan Harper (nathharp) | 10:34 |
| kbaegis | Why does neutron suck so badly at deleting ports? | 10:36 |
| *** caowei has quit IRC | 10:36 | |
| *** caowei has joined #openstack-kolla | 10:37 | |
| kbaegis | Neutron is talking dirty database to me, and I'm not into that: There was a conflict when trying to complete your request. | 10:39 |
| zhubingbing | https://bugs.launchpad.net/kolla-ansible/+bug/1672334 | 10:41 |
| openstack | Launchpad bug 1672334 in kolla-ansible "Fix separate bug in cleanup-containers" [High,Confirmed] - Assigned to zhubingbing (zhubingbing) | 10:41 |
| zhubingbing | can u look it ?duonghq | 10:42 |
| *** zhugaoxiao has quit IRC | 10:43 | |
| *** zhugaoxiao has joined #openstack-kolla | 10:44 | |
| openstackgerrit | zhubingbing proposed openstack/kolla-ansible master: Fix separate bug in cleanup-containers https://review.openstack.org/444807 | 10:47 |
| openstackgerrit | Dave Walker proposed openstack/kolla-ansible master: init-runonce - set image distro to "linux" https://review.openstack.org/444808 | 10:49 |
| openstackgerrit | Dave Walker proposed openstack/kolla-ansible master: init-runonce - set image distro to "linux" https://review.openstack.org/444808 | 10:50 |
| openstackgerrit | zhubingbing proposed openstack/kolla-ansible master: Fix separate bug in cleanup-containers https://review.openstack.org/444807 | 10:51 |
| *** matrohon has quit IRC | 10:57 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Nova oslo_messaging_notifications config error https://review.openstack.org/444504 | 10:57 |
| *** trinaths1 has quit IRC | 10:57 | |
| *** matrohon has joined #openstack-kolla | 10:58 | |
| *** caowei has quit IRC | 11:03 | |
| egonzalez | anyone have ever configured two VIP in the same keepalived interface? | 11:12 |
| egonzalez | lets say, internal_vip and external_vip in the same interface with different VIP addresses | 11:12 |
| *** matrohon has quit IRC | 11:21 | |
| *** ccesario__ has quit IRC | 11:22 | |
| *** matrohon has joined #openstack-kolla | 11:23 | |
| *** zhubingbing has quit IRC | 11:24 | |
| *** rwallner has joined #openstack-kolla | 11:27 | |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: ironic-pxe: handle non-x86 architectures https://review.openstack.org/434817 | 11:30 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: debian: enable all images enabled for Ubuntu https://review.openstack.org/432787 | 11:30 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: Add support for non-x86 architectures (aarch64, ppc64le) https://review.openstack.org/430940 | 11:30 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: nova-libvirt: handle ppc64le https://review.openstack.org/434810 | 11:30 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: ceph: move ceph-fuse package to cephfs-fuse image https://review.openstack.org/440634 | 11:30 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: nova-compute: handle rtslib(-fb) for debian-binary builds too https://review.openstack.org/435941 | 11:30 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: debian: move to stretch https://review.openstack.org/434453 | 11:30 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: debian: install gnupg and dirmngr needed for apt-key https://review.openstack.org/434431 | 11:30 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: openstack-base/kolla-toolbox: CentOS has mariadb-devel https://review.openstack.org/444832 | 11:30 |
| *** furlongm has quit IRC | 11:37 | |
| *** furlongm has joined #openstack-kolla | 11:38 | |
| *** papacz has joined #openstack-kolla | 11:40 | |
| *** rhallisey has joined #openstack-kolla | 11:41 | |
| *** trinaths has joined #openstack-kolla | 11:41 | |
| *** yingjun has joined #openstack-kolla | 11:51 | |
| *** erlon has joined #openstack-kolla | 11:51 | |
| *** sudipto has quit IRC | 11:56 | |
| *** rwellum has joined #openstack-kolla | 11:59 | |
| *** matrohon has quit IRC | 12:03 | |
| *** matrohon has joined #openstack-kolla | 12:05 | |
| *** trinaths has quit IRC | 12:06 | |
| *** yingjun has quit IRC | 12:11 | |
| *** athomas has quit IRC | 12:12 | |
| *** matrohon has quit IRC | 12:13 | |
| *** saneax is now known as saneax-_-|AFK | 12:14 | |
| *** matrohon has joined #openstack-kolla | 12:16 | |
| *** manheim has quit IRC | 12:19 | |
| *** DuncanT has joined #openstack-kolla | 12:26 | |
| *** athomas has joined #openstack-kolla | 12:26 | |
| *** lamt has quit IRC | 12:28 | |
| *** jheroux has joined #openstack-kolla | 12:35 | |
| *** zhubingbing has joined #openstack-kolla | 12:38 | |
| *** spsurya__ has joined #openstack-kolla | 12:43 | |
| *** zhubingbing__ has joined #openstack-kolla | 12:43 | |
| *** zhubingbing has quit IRC | 12:44 | |
| *** salv-orlando has quit IRC | 12:50 | |
| *** manheim has joined #openstack-kolla | 12:57 | |
| *** yingjun has joined #openstack-kolla | 13:01 | |
| *** zhurong has quit IRC | 13:01 | |
| *** matrohon has quit IRC | 13:04 | |
| *** matrohon has joined #openstack-kolla | 13:06 | |
| *** matrohon has quit IRC | 13:07 | |
| *** matrohon has joined #openstack-kolla | 13:07 | |
| *** lamt has joined #openstack-kolla | 13:09 | |
| *** brice has joined #openstack-kolla | 13:12 | |
| hrw | yo | 13:12 |
| hrw | who can tell me more about DELOREAN? | 13:12 |
| *** sudipto has joined #openstack-kolla | 13:13 | |
| hrw | and why it is buried in kolla/common/config.py instead of being in docker/base/Dockerfile.j2? | 13:13 |
| *** jtriley has joined #openstack-kolla | 13:14 | |
| *** brice has quit IRC | 13:14 | |
| egonzalez | hrw, to make it configurable at kolla-build.conf | 13:14 |
| hrw | egonzalez: I am trying to find a way to make it arch dependend | 13:15 |
| *** salv-orlando has joined #openstack-kolla | 13:15 | |
| hrw | now it is forces for all rpm builds and brakes !x86-64 ones | 13:15 |
| egonzalez | hrw, https://github.com/openstack/kolla/blob/master/kolla/common/config.py#L228 is used as default setup_rpm_config | 13:18 |
| egonzalez | in stable branch that default is removed | 13:18 |
| *** jtriley has quit IRC | 13:19 | |
| hrw | egonzalez: I will test changing DELOREAN(_OPTS) from string to set and using porper set depending on arch | 13:19 |
| egonzalez | hrw, makes sense | 13:19 |
| *** g3ek has quit IRC | 13:20 | |
| *** jtriley has joined #openstack-kolla | 13:20 | |
| *** schwicht has joined #openstack-kolla | 13:22 | |
| *** skramaja has quit IRC | 13:23 | |
| *** salv-orl_ has joined #openstack-kolla | 13:24 | |
| *** gfidente^2nd has joined #openstack-kolla | 13:25 | |
| *** salv-orlando has quit IRC | 13:26 | |
| *** gfidente has quit IRC | 13:27 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Neutron oslo_messaging_notifications config error https://review.openstack.org/444506 | 13:29 |
| *** g3ek has joined #openstack-kolla | 13:29 | |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: ironic-pxe: handle non-x86 architectures https://review.openstack.org/434817 | 13:32 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: debian: enable all images enabled for Ubuntu https://review.openstack.org/432787 | 13:32 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: Add support for non-x86 architectures (aarch64, ppc64le) https://review.openstack.org/430940 | 13:32 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: openstack-base/kolla-toolbox: CentOS has mariadb-devel https://review.openstack.org/444832 | 13:32 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: nova-libvirt: handle ppc64le https://review.openstack.org/434810 | 13:32 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: ceph: move ceph-fuse package to cephfs-fuse image https://review.openstack.org/440634 | 13:32 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: nova-compute: handle rtslib(-fb) for debian-binary builds too https://review.openstack.org/435941 | 13:32 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: debian: move to stretch https://review.openstack.org/434453 | 13:32 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: debian: install gnupg and dirmngr needed for apt-key https://review.openstack.org/434431 | 13:32 |
| hrw | sorry for flood but that set got quite integrated to each other | 13:33 |
| hrw | docker/base/Dockerfile.j2 is a formatting mess ;( | 13:35 |
| hrw | but reformatting it in my patch would add more work for reviewers. refomatting it before my patch, submitting and then rebasing my set on it would not be better | 13:36 |
| *** eaguilar has joined #openstack-kolla | 13:40 | |
| *** gfidente^2nd is now known as gfidente | 13:41 | |
| *** eaguilar has quit IRC | 13:45 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes master: WIP Replace unconditional wait for compute kit https://review.openstack.org/444613 | 13:46 |
| *** srwilkers has joined #openstack-kolla | 13:46 | |
| *** mdnadeem has quit IRC | 13:46 | |
| *** sudipto has quit IRC | 13:50 | |
| *** salv-orl_ has quit IRC | 13:51 | |
| srwilkers | good morning | 13:51 |
| hrw | uf. finally centos/aarch64 builds are going | 13:52 |
| *** Jezekus has joined #openstack-kolla | 14:00 | |
| *** lucasxu has joined #openstack-kolla | 14:00 | |
| *** yingjun has quit IRC | 14:02 | |
| *** manheim has quit IRC | 14:09 | |
| *** sudipto has joined #openstack-kolla | 14:10 | |
| *** Jeffrey4l has joined #openstack-kolla | 14:12 | |
| *** sayantan_ has joined #openstack-kolla | 14:15 | |
| *** spsurya__ has quit IRC | 14:15 | |
| *** Jezekus has quit IRC | 14:16 | |
| *** Jeffrey4l has quit IRC | 14:17 | |
| *** manheim has joined #openstack-kolla | 14:18 | |
| *** Jeffrey4l has joined #openstack-kolla | 14:18 | |
| *** kbaegis has quit IRC | 14:26 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Designate: manage Api and Central service workers https://review.openstack.org/444929 | 14:28 |
| *** eaguilar has joined #openstack-kolla | 14:29 | |
| *** satyar has joined #openstack-kolla | 14:33 | |
| *** kbaegis has joined #openstack-kolla | 14:35 | |
| *** dave-mccowan has joined #openstack-kolla | 14:35 | |
| *** kbaegis has quit IRC | 14:37 | |
| egonzalez | is aodh required for ceilometer alarms? without aodh are not working for me | 14:38 |
| Guest35019 | egonzalez: hey, qq do you guys see gate failure on centos? | 14:42 |
| *** Guest35019 is now known as sbezverk | 14:42 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Designate: manage Api and Central service workers https://review.openstack.org/444929 | 14:43 |
| egonzalez | sbezverk, any error in special? i've seen one today with oslo log not iterable object or something like that, after recheck issue was gone | 14:43 |
| sbezverk | egonzalez: http://paste.openstack.org/show/602502/ | 14:46 |
| sbezverk | looks like bindep is failing.. | 14:46 |
| sbezverk | egonzalez: it looks to me as infra issue but wanted to confirm if you see the same before going and complaining on infra channel ;) | 14:48 |
| *** sbezverk has quit IRC | 14:50 | |
| *** mkoderer has joined #openstack-kolla | 14:53 | |
| *** manheim has quit IRC | 15:01 | |
| *** manheim has joined #openstack-kolla | 15:02 | |
| *** manheim has quit IRC | 15:03 | |
| *** sayantan_ has quit IRC | 15:05 | |
| *** adrian_otto has joined #openstack-kolla | 15:07 | |
| *** sayantan_ has joined #openstack-kolla | 15:09 | |
| *** ipsecguy_ has joined #openstack-kolla | 15:10 | |
| *** ipsecguy has quit IRC | 15:11 | |
| *** eaguilar_ has joined #openstack-kolla | 15:12 | |
| *** eaguilar has quit IRC | 15:14 | |
| *** sayantan_ has quit IRC | 15:16 | |
| *** yingjun has joined #openstack-kolla | 15:17 | |
| *** gfhellma has quit IRC | 15:21 | |
| *** hieulq_ has joined #openstack-kolla | 15:26 | |
| *** zhubingbing__ has quit IRC | 15:27 | |
| *** satyar has quit IRC | 15:28 | |
| *** yingjun has quit IRC | 15:33 | |
| openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible master: Fix heat ec2 keystone auth https://review.openstack.org/444953 | 15:36 |
| egonzalez | k8s guys, can anyone check if this ^^ also affect you too? | 15:37 |
| kfox1111 | morning. | 15:39 |
| kfox1111 | egonzalez: interesting. the v3 is required for it to function properly? | 15:40 |
| *** eaguilar_ has quit IRC | 15:40 | |
| egonzalez | kfox1111, yep, at least in ansible was not working without v3, heat was not able to create a valid ec2 auth in keystone | 15:40 |
| kfox1111 | ok. then the k8s one should have the same fix. | 15:41 |
| kfox1111 | we haven't tested that too much yet. | 15:41 |
| *** sbezverk has joined #openstack-kolla | 15:44 | |
| openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible master: Fix heat ec2 keystone auth https://review.openstack.org/444953 | 15:44 |
| *** mnasiadka has quit IRC | 15:47 | |
| hrw | who is using kolla to build centos based images? | 15:49 |
| mnaser | hrw we dont build but we consume the tagged ones | 15:49 |
| *** fooliouno has joined #openstack-kolla | 15:50 | |
| hrw | ok | 15:50 |
| openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible master: Add cpu_mode in quickstart guide https://review.openstack.org/444958 | 15:51 |
| *** crushil has joined #openstack-kolla | 15:59 | |
| mnaser | has anyone used kolla to scale a control plane up (more specifically my biggest concern is how scaling of mariadb is handled) | 16:01 |
| kfox1111 | mnaser: mariadb doesn't really scale. | 16:02 |
| kfox1111 | all you can do is manually shard. | 16:03 |
| kfox1111 | or throw bigger iron it it. | 16:03 |
| mnaser | kfox1111: well maybe scale was the wrong term but adding new nodes to the galera cluster db | 16:03 |
| kfox1111 | more then 3? | 16:03 |
| mnaser | so if i have 1 controller and later decide to go up to 3 for ha | 16:03 |
| kfox1111 | ah. | 16:03 |
| kfox1111 | haven't tried that. :/ | 16:03 |
| mnaser | it would be nice to have small deployments start with 1 node and still be able to go u pto 3 | 16:06 |
| *** srwilkers has left #openstack-kolla | 16:07 | |
| *** srwilkers has joined #openstack-kolla | 16:07 | |
| *** imcsk8 has quit IRC | 16:07 | |
| *** imcsk8 has joined #openstack-kolla | 16:09 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible master: Add panko.conf file in ceilometer container https://review.openstack.org/444719 | 16:09 |
| SamYaple | mnaser: are you talking about kolla-ansible? | 16:09 |
| mnaser | yup SamYaple | 16:09 |
| SamYaple | mnaser: i wrote it to go from 1-n number of nodes, yes | 16:10 |
| SamYaple | it scales up just fine | 16:10 |
| SamYaple | scaling down is trickier and requires a bit of manual work | 16:10 |
| mnaser | SamYaple: yeah that's not something that we want to get into, but getting the ability of scaling up is perfect | 16:10 |
| mnaser | i assume its as straight forward as adding it to the inventory and redeploying? | 16:10 |
| SamYaple | yep | 16:10 |
| mnaser | that's great to hear | 16:11 |
| *** kbaegis has joined #openstack-kolla | 16:12 | |
| fooliouno | Hello .. I am trying to bring up mariadb in kolla-kubernetes. It works in a single node configuration, but when I try to start mariadb on a worker node, I get this error: http://pastebin.com/b9qSPjzj | 16:13 |
| fooliouno | Any ideas on what might be wrong | 16:14 |
| *** david-lyle has joined #openstack-kolla | 16:15 | |
| inc0 | good morning | 16:23 |
| hrw | is there a way to inject time into each INFO: line from kolla output? | 16:29 |
| *** gfhellma has joined #openstack-kolla | 16:32 | |
| openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Designate: manage Api and Central services workers https://review.openstack.org/444929 | 16:33 |
| *** matrohon has quit IRC | 16:33 | |
| mnaser | hrw you could probably edit this - https://github.com/openstack/kolla/blob/master/docker/base/set_configs.py | 16:35 |
| *** salv-orlando has joined #openstack-kolla | 16:35 | |
| mnaser | however, docker can print timestamps natively | 16:35 |
| *** hieulq_ has quit IRC | 16:35 | |
| mnaser | hrw docker logs -t <container> | 16:35 |
| hrw | mnaser: "INFO:kolla.image.build.gnocchi-base:Cleaning up everything" is still kolla imho ;) will check set_configs | 16:36 |
| mnaser | ooo | 16:36 |
| mnaser | i was thinking the docker start up things | 16:36 |
| hrw | ;) | 16:36 |
| mnaser | if it uses oslo_log it could be easy | 16:36 |
| hrw | mnaser: so far I do not yet care of how docker runs but rather about how kolla builds | 16:37 |
| mnaser | yeah it looks like it uses native python logging so https://github.com/openstack/kolla/blob/master/kolla/image/build.py#L56-L77 | 16:37 |
| hrw | yep | 16:38 |
| *** pcaruana has quit IRC | 16:38 | |
| *** rwsu has joined #openstack-kolla | 16:40 | |
| *** zioproto has quit IRC | 16:41 | |
| rwellum | ping sdake | 16:43 |
| *** kbaegis has quit IRC | 16:49 | |
| *** sayantan_ has joined #openstack-kolla | 16:51 | |
| *** Jezekus has joined #openstack-kolla | 16:54 | |
| *** kbaegis has joined #openstack-kolla | 17:00 | |
| *** Mannheim has quit IRC | 17:01 | |
| *** kbaegis has quit IRC | 17:03 | |
| *** lucasxu has quit IRC | 17:03 | |
| *** Jezekus has quit IRC | 17:05 | |
| *** rabel_b1 has joined #openstack-kolla | 17:06 | |
| *** Jezekus has joined #openstack-kolla | 17:06 | |
| *** egonzalez has quit IRC | 17:08 | |
| *** rabel has quit IRC | 17:09 | |
| *** kbaegis has joined #openstack-kolla | 17:10 | |
| *** gfidente is now known as gfidente|afk | 17:15 | |
| rwellum | I'm trying kolla-kubernetes, and seeing this issue: https://review.openstack.org/#/c/417331/ - which was fixed a long time ago. My full output is: http://paste.openstack.org/show/602537/ | 17:17 |
| *** openstackgerrit has quit IRC | 17:18 | |
| *** sudipto has quit IRC | 17:19 | |
| inc0 | rwellum: are you using stable kolla/ansible or master? | 17:19 |
| *** Jezekus has quit IRC | 17:21 | |
| rwellum | master I think: git clone http://github.com/openstack/kolla-kubernetes | 17:22 |
| inc0 | how about kolla-ansible? | 17:22 |
| inc0 | for config gen | 17:22 |
| rwellum | Same: I am following: https://etherpad.openstack.org/p/kolla-kubernetes-deploy-guide-BP | 17:23 |
| rwellum | git clone http://github.com/openstack/kolla-ansible | 17:23 |
| *** unicell has joined #openstack-kolla | 17:25 | |
| *** unicell has quit IRC | 17:25 | |
| *** unicell has joined #openstack-kolla | 17:25 | |
| *** Rowan_ has joined #openstack-kolla | 17:26 | |
| Rowan_ | Hi, am trying to deploy newton using Kolla and when I deploy with default config it is deploy with flat networks. | 17:27 |
| inc0 | Rowan_: well, yeah | 17:28 |
| Rowan_ | I want to deploy using vlans are there parameters we can set to change it to vlans and configure our own bridges? | 17:28 |
| inc0 | that and tunnels | 17:28 |
| inc0 | as for "configure our own bridges" depends how much you want to configure them a | 17:29 |
| inc0 | for vlans, hold on | 17:29 |
| *** lucasxu has joined #openstack-kolla | 17:30 | |
| inc0 | //docs.openstack.org/developer/kolla-ansible/advanced-configuration.html#openstack-service-configuration-in-kolla | 17:31 |
| inc0 | https://docs.openstack.org/developer/kolla-ansible/advanced-configuration.html#openstack-service-configuration-in-kolla | 17:31 |
| inc0 | use this mechanism | 17:31 |
| inc0 | to override paste.ini and provide vlan ranges | 17:31 |
| inc0 | we should probably document how to do that | 17:32 |
| inc0 | I'll submit bp | 17:32 |
| Rowan_ | I tried it node_custom_config: "/etc/kolla/config" but no deploying with the changes | 17:32 |
| Rowan_ | not* | 17:32 |
| Rowan_ | ex: am trying to change ml2_conf.ini so created /etc/kolla/config/neutron/plugins/ml2/ml2_conf.ini and reconfigured the env | 17:33 |
| *** mgoddard has quit IRC | 17:33 | |
| Rowan_ | but nodes are not getting new config | 17:34 |
| inc0 | https://docs.openstack.org/developer/kolla-ansible/advanced-configuration.html#openstack-service-configuration-in-kolla | 17:35 |
| inc0 | sry | 17:35 |
| inc0 | wrong link | 17:35 |
| inc0 | https://github.com/openstack/kolla/blob/stable/newton/ansible/roles/neutron/tasks/config.yml#L99-L101 | 17:35 |
| inc0 | correct place for ml2 would be /etc/kolla/config/neutron/ml2_conf.ini | 17:36 |
| inc0 | Rowan_: also submitted this https://blueprints.launchpad.net/kolla-ansible/+spec/docs-config-examples | 17:37 |
| Rowan_ | Thanks inc0 really appreciate your help and timely response. | 17:37 |
| inc0 | everyone, feel free to add comments about other use-cases ^ | 17:37 |
| inc0 | Rowan_: np:) | 17:37 |
| inc0 | let me know if this helps you | 17:37 |
| rwellum | inc0, so should I not be using master? | 17:38 |
| inc0 | reason I ask is that I don't think kolla-ansible has this change backported to stable/ocata | 17:38 |
| Rowan_ | Will give a try and update you about vlans config | 17:38 |
| inc0 | so if you'd use stable/ocata, it might not have code | 17:39 |
| inc0 | but I'll confirm that | 17:39 |
| rwellum | inc0, I checked and I see the code. | 17:39 |
| hogepodge | ugh, realized in organizing this meeting that daylight saving happened over the weekend | 17:39 |
| rwellum | in my clone | 17:39 |
| inc0 | ahh nvm then | 17:39 |
| inc0 | bug is still open about that in kolla-k8s, I'll check it out when my brain starts properly | 17:39 |
| inc0 | hogepodge: yeah.... it's always problem | 17:40 |
| inc0 | we can place bets on how many people will miss this weeks meeting | 17:40 |
| inc0 | because of dst | 17:40 |
| hogepodge | going to stick with 1830 UTC and we can work out scheduling | 17:40 |
| hogepodge | dst is the worst | 17:40 |
| hogepodge | I'll probably go back and delete all my standing meetings and resubscribe to make sure my calendar didn't do the wrong thing | 17:41 |
| inc0 | I wonder if someone actually made research to determine how much time we *lose* because of dst | 17:41 |
| inc0 | all the meetings without quorum and such | 17:41 |
| jascott1 | DST is about money | 17:42 |
| jascott1 | more shopping hours | 17:43 |
| *** salv-orlando has quit IRC | 17:43 | |
| *** salv-orlando has joined #openstack-kolla | 17:43 | |
| inc0 | how does that help? besides making everyone sleepy and not willing to leave home | 17:43 |
| jascott1 | haha no doubt | 17:43 |
| inc0 | (yes, I'm working from home and thinking of taking a nap) | 17:44 |
| *** eaguilar has joined #openstack-kolla | 17:44 | |
| jascott1 | well in the winter you get more daylight, which more time to shop | 17:44 |
| inc0 | I mean I understand why that was an important thing *before* we had street lights everywhere | 17:44 |
| -openstackstatus- NOTICE: restarting gerrit to address performance problems | 17:44 | |
| inc0 | btw kolla was oficially released | 17:46 |
| inc0 | all glory to Ocata | 17:46 |
| jascott1 | inc0 "In 1984, Fortune magazine estimated that a seven-week extension of DST would yield an additional $30 million for 7-Eleven stores, and the National Golf Foundation estimated the extension would increase golf industry revenues $200 million to $300 million." | 17:49 |
| jascott1 | https://en.wikipedia.org/wiki/Daylight_saving_time | 17:49 |
| rwellum | inc0, line 136 https://etherpad.openstack.org/p/kolla-kubernetes-deploy-guide-BP - seems to suggest that kolla is up and running - but it's not afaik. Does this seem wrong to you too? | 17:50 |
| *** rabel_b1 has quit IRC | 17:51 | |
| *** manheim has joined #openstack-kolla | 17:54 | |
| *** adrian_otto has quit IRC | 17:57 | |
| *** eaguilar_ has joined #openstack-kolla | 17:58 | |
| *** eaguilar has quit IRC | 18:00 | |
| *** Jezekus has joined #openstack-kolla | 18:01 | |
| *** adrian_otto has joined #openstack-kolla | 18:02 | |
| inc0 | rwellum: adding labels to nodes? | 18:04 |
| rwellum | inc0, yeah because at this point in guide - we don't have nodes I think. | 18:05 |
| inc0 | rwellum: so labels are meant to tell k8s where deploy what | 18:06 |
| inc0 | some services can be just scheduled k8s style | 18:06 |
| *** crushil has left #openstack-kolla | 18:06 | |
| inc0 | but some, like compute nodes should be deployed in specific nodes | 18:06 |
| portdirect | rwellum: I'm starting to work through the guide and will be running a deployment tonight, but that line looks like its in the right place | 18:06 |
| inc0 | it's for k8s more than it's for kolla itself | 18:06 |
| portdirect | inc0's explanation is spot on - its just to make the nodes that kolla pods will run on | 18:07 |
| portdirect | s/make/mark | 18:07 |
| inc0 | yeah I still fight the initial k8s deploy:/ | 18:07 |
| inc0 | halcyon hates me | 18:07 |
| portdirect | :( thats not good inc0, if theres any way I can help just give me a holler | 18:08 |
| inc0 | portdirect: well, issue I'm dealing with is after I deploy k8s, I run tiller and it fails liveness probe:/ | 18:08 |
| inc0 | imho it fails to connect to underlying k8s | 18:08 |
| inc0 | (no logs available) | 18:09 |
| rwellum | 'minime-02' is the node according to that guid portdirect, inc0 - that doesn't make too much sense to me | 18:09 |
| rwellum | Now: ALLINONENODE=$(hostname) | 18:09 |
| rwellum | kubectl label node $ALLINONENODE kolla_compute=true | 18:09 |
| portdirect | inc0: does the dns test work? | 18:09 |
| inc0 | rwellum: minime-02 is sdake's node, it's just hostname | 18:09 |
| rwellum | that makes more sense | 18:09 |
| rwellum | right... | 18:09 |
| inc0 | portdirect: let me rebuild whole env and start fresh | 18:10 |
| inc0 | I'm using flannel btw | 18:10 |
| rwellum | portdirect, great that you're looking at this now. | 18:10 |
| portdirect | yeah - just relocated to the States, glad to be getting ramped back up now :D | 18:10 |
| inc0 | oh did you? | 18:11 |
| inc0 | cool, congrats | 18:11 |
| inc0 | now I won't have any second thoughts to ping you | 18:11 |
| inc0 | I always felt a little guilty before because of late night thing | 18:11 |
| * portdirect moves back to UK :P | 18:11 | |
| sbezverk | inc0: has infra fixed broken centos mirror? | 18:11 |
| portdirect | yeah - I'm really looking forward o working normal hours again :) | 18:12 |
| inc0 | I had no idea about centos broken mirror in the first place;) | 18:12 |
| SamYaple | portdirect: define normal | 18:12 |
| sbezverk | inc0: yeah this morning all centos jobs were failing | 18:12 |
| SamYaple | inc0: sounds like dns! | 18:12 |
| *** Jezekus has quit IRC | 18:12 | |
| portdirect | 6am to 5:59am SamYaple | 18:12 |
| SamYaple | portdirect: whats an am? | 18:13 |
| inc0 | anno morningum? | 18:13 |
| inc0 | (that was bad...) | 18:13 |
| inc0 | I hate dst | 18:13 |
| SamYaple | just ignore it | 18:15 |
| SamYaple | itll go away | 18:15 |
| *** MasterOfBugs has joined #openstack-kolla | 18:16 | |
| *** adrian_otto has quit IRC | 18:17 | |
| *** adrian_otto has joined #openstack-kolla | 18:19 | |
| *** Mannheim has joined #openstack-kolla | 18:21 | |
| *** mgoddard has joined #openstack-kolla | 18:21 | |
| jascott1 | portdirect which state did you end up in? | 18:23 |
| portdirect | missouri, now I can throw things at srwilkers :) | 18:24 |
| jascott1 | excellent! | 18:24 |
| srwilkers | portdirect, that would require you to have sufficient aim | 18:24 |
| jascott1 | http://i.imgur.com/eCkyv.jpg | 18:25 |
| portdirect | I'll just use larger objects to compensate | 18:25 |
| inc0 | jascott1: we should get a nerf gun | 18:26 |
| inc0 | with API ideally | 18:26 |
| jascott1 | i think our neighbors would revolt | 18:26 |
| jascott1 | haha | 18:26 |
| jascott1 | the guy between our cubes would be prob be quite pissed | 18:27 |
| inc0 | well, they'll already know rules of engagement | 18:27 |
| inc0 | idea for project | 18:27 |
| portdirect | jascott1: "collateral damage" they should know the risks | 18:27 |
| jascott1 | maybe those that are like a mortar where we can get the trajectory dialed in | 18:27 |
| inc0 | nerfgun to automatically shoot down nerfs | 18:27 |
| jascott1 | lol | 18:27 |
| jascott1 | nerf drone | 18:27 |
| inc0 | computer vision+tons and tons of algebra to figure out ballistics | 18:28 |
| jascott1 | nerf to shoot down nerf would be very impressive | 18:28 |
| inc0 | probably couple hundreds of hours writing code that will do something totally useless | 18:28 |
| jascott1 | then 2 nerfs would be raining down on Brian each time | 18:29 |
| inc0 | not to mention hardware part | 18:29 |
| jascott1 | it would prob get confused and just shoot him everytime he stook up ;) | 18:29 |
| inc0 | I mean you need some really quick and precise servos out there | 18:29 |
| jascott1 | stook/stood | 18:29 |
| inc0 | or another idea | 18:29 |
| inc0 | drone with nerfgun (as you mentioned) | 18:30 |
| inc0 | but with autonomus flying and infrared target painting | 18:30 |
| inc0 | so you can point invisible laser at someone at the outher side of openspace | 18:30 |
| jascott1 | heheh | 18:30 |
| inc0 | then we gradually extend range and after a while have outdoor drone for random pedestrians;) | 18:31 |
| *** fooliouno has quit IRC | 18:31 | |
| inc0 | nerf headshot from 300m away would be impressive;_) | 18:31 |
| inc0 | hmm...but that's not a bad idea for things like video | 18:32 |
| inc0 | have drone with some pre-coded flights patterns to record video of thing you point invisible laser on | 18:33 |
| inc0 | like idk, music band on stage | 18:33 |
| jascott1 | Intel has already demoed a drone that can follow a moving target | 18:33 |
| inc0 | yeah, I saw few ideas of that | 18:33 |
| inc0 | but having target painter would be good, or invisible led | 18:34 |
| inc0 | like I could install this thing on my paraglider and have drone flying around me taping me | 18:34 |
| jascott1 | yeah that seems like a great idea | 18:34 |
| inc0 | pretty fun gopro appliance idea, who wants to start a startup?:P | 18:35 |
| jascott1 | my wife says im not allowed to play with startups anymore | 18:35 |
| jascott1 | :D | 18:35 |
| inc0 | :( | 18:35 |
| *** sayantan_ has quit IRC | 18:35 | |
| inc0 | yeah you have morgage now | 18:35 |
| jascott1 | only 360 easy monthly payments left! | 18:36 |
| SamYaple | jascott1: just do ten at a time | 18:36 |
| jascott1 | now why didnt I think of that | 18:37 |
| SamYaple | too many startups | 18:37 |
| inc0 | or even better, one startup, sell it to intel, pay off house and still have 14.993 bil $ left | 18:37 |
| jascott1 | i have been looking into paying down extra on principal, something like paying extra $400 mo will save me about 8+ years | 18:38 |
| SamYaple | jascott1: doubling payments on a 30 year pays it off in 7 years | 18:38 |
| portdirect | inc0 you can act on intels behalf here? My patent for cloud lego just came through this morning. | 18:38 |
| portdirect | jascott1: you snooze, you loose. | 18:38 |
| jascott1 | portdirect >:( | 18:39 |
| inc0 | portdirect: cloud lego? | 18:39 |
| portdirect | 100% my idea | 18:39 |
| inc0 | who uses cloud any more? cloud is passe | 18:39 |
| jascott1 | Missouri huh | 18:39 |
| portdirect | hybrid cloud lego? | 18:39 |
| Rowan_ | inc0 sorry to interrupt you. | 18:39 |
| jascott1 | (and your important conversation) | 18:39 |
| inc0 | Rowan_: it's not like you interrupted anything constructive;) | 18:39 |
| Rowan_ | Deployed with vlans and having issues with networking | 18:40 |
| inc0 | aren't we all... whats happening? | 18:40 |
| Rowan_ | while deploying instances with vlans network unable to reach metadata | 18:40 |
| Rowan_ | Lease of 172.24.51.54 obtained, lease time 86400 route: SIOCADDRT: File exists WARN: failed: route add -net "0.0.0.0/0" gw "172.24.51.1" cirros-ds 'net' up at 0.93 checking http://169.254.169.254/2009-04-04/instance-id failed 1/20: up 0.94. request failed | 18:41 |
| SamYaple | Rowan_: its just the vlan networks that don't have metadata? | 18:41 |
| SamYaple | so you can do a vxlan with no problem? | 18:41 |
| Rowan_ | With vxlan no issues | 18:41 |
| SamYaple | are the vlan networks behind neutron routers? | 18:41 |
| Rowan_ | but atatching float IP to vxlan instances not working | 18:41 |
| Rowan_ | nope for vlans I configured default gw | 18:42 |
| SamYaple | so its an isolated subnet | 18:43 |
| SamYaple | you need to enable metadata on isolated subnets | 18:43 |
| Rowan_ | am able to ping the vlan instances but metadata failing | 18:43 |
| SamYaple | or, alternatively, setup a metadata network | 18:43 |
| inc0 | ok portdirect sooo....let's start with k8s debugging if you have a moment:) | 18:44 |
| portdirect | np around for 10-15 | 18:45 |
| *** sayantan_ has joined #openstack-kolla | 18:45 | |
| portdirect | you run the container here? https://docs.openstack.org/developer/kolla-kubernetes/development-environment.html#testing-the-deployed-environment | 18:45 |
| inc0 | http://paste.ubuntu.com/24172091/ | 18:45 |
| Rowan_ | Samyaple it is not isloated subnet subnet gateway_ip enabled | 18:48 |
| *** manheim has quit IRC | 18:48 | |
| inc0 | portdirect: http://paste.openstack.org/show/602551/ <- dns seems fine, but 10.96.0.10 doesn;t | 18:49 |
| SamYaple | Rowan_: if its not connected to a neutron router it counts as isolated | 18:49 |
| Rowan_ | ok | 18:51 |
| Rowan_ | how to enable metadata on isolated subnets? | 18:54 |
| Rowan_ | when I check dhcp namespace it has default to metadata | 18:55 |
| SamYaple | dhcp agent, option enable_isolated_metadata=True | 18:55 |
| Rowan_ | default via 172.24.51.1 dev tap28fe3684-3b 169.254.0.0/16 dev tap28fe3684-3b proto kernel scope link src 169.254.169.254 172.24.51.0/24 dev tap28fe3684-3b proto kernel scope link src 172.24.51.50 | 18:55 |
| *** kbaegis has quit IRC | 18:55 | |
| SamYaple | metadata, by default, is handled in the neutron router namespace | 18:56 |
| SamYaple | you do not have a neutron router | 18:56 |
| SamYaple | you have to tell the instance how to handle metadata now | 18:57 |
| SamYaple | in this case, through the dhcp agent. but you need to tell the dhcp agent that | 18:57 |
| Rowan_ | enable_isolated_metadata is set to true by default in my env | 18:58 |
| SamYaple | right. so there is another piece to this when you also have an external gateway | 19:00 |
| SamYaple | you have to remove the gateway_ip and set it as a host route | 19:00 |
| SamYaple | so you need to set an additional route of 0.0.0.0/0 via 172.24.51.1 | 19:00 |
| SamYaple | and remove the gateway_ip from the subnet for the metadata to work | 19:00 |
| Rowan_ | so I need to remove gw for the vlan? | 19:00 |
| SamYaple | yea | 19:00 |
| inc0 | yeah, pretty much if your gateway is external to cloud, nothing will redirect to metadata - it's iptables rule on routers normally | 19:00 |
| Rowan_ | Will try now | 19:01 |
| SamYaple | inc0: no thats not true | 19:01 |
| inc0 | :( | 19:01 |
| SamYaple | inc0: it still works | 19:01 |
| SamYaple | dhcp injects a route | 19:01 |
| SamYaple | but it only does that when gateway_ip is not set | 19:01 |
| SamYaple | so to get internet connectivity and metadata you ahve to inject teh default route | 19:01 |
| inc0 | ahh, right | 19:01 |
| SamYaple | you end up with 169.254.169.254/32 pointed to dhcp server and 0.0.0.0/0 pointed to the gateway | 19:02 |
| Rowan_ | no luck | 19:03 |
| Rowan_ | disable gw added host route | 19:04 |
| Rowan_ | WARN: failed: route add -net "0.0.0.0/0" gw "172.24.51.1" cirros-ds 'net' up at 0.89 checking http://169.254.169.254/2009-04-04/instance-id failed 1/20: up 0.89. request failed failed 2/20: up 12.93. request failed | 19:04 |
| inc0 | Rowan_: tcpdump time? | 19:04 |
| *** athomas has quit IRC | 19:04 | |
| *** zhugaoxiao has quit IRC | 19:04 | |
| SamYaple | you removed gateway_ip from neutron subnet, restarted the agents, restarted the instance and still no go? | 19:04 |
| inc0 | check qdhcp ns if it gets traffic from cirros | 19:04 |
| *** zhugaoxiao has joined #openstack-kolla | 19:05 | |
| *** Jeffrey4l has quit IRC | 19:05 | |
| *** Jeffrey4l has joined #openstack-kolla | 19:05 | |
| inc0 | Jeffrey4l: in case you're awake, we have oficially released Kolla:) | 19:05 |
| Rowan_ | able to ping instance from qdhcp | 19:06 |
| inc0 | check if qdhcp gets traffic if you try http://169... | 19:07 |
| inc0 | and if it doesn, what happends next | 19:07 |
| *** eaguilar_ has quit IRC | 19:07 | |
| inc0 | check also iptables -t nat (afair) for metadata rule | 19:07 |
| Rowan_ | sorry no unable to ping instance from qhdcp | 19:08 |
| *** eaguilar has joined #openstack-kolla | 19:17 | |
| *** Mannheim has quit IRC | 19:20 | |
| *** salv-orl_ has joined #openstack-kolla | 19:24 | |
| *** salv-orlando has quit IRC | 19:27 | |
| *** openstackgerrit has joined #openstack-kolla | 19:32 | |
| *** mgoddard has quit IRC | 19:41 | |
| hrw | uf. looks like centos/source build of gnocchi-base will succeed | 19:44 |
| *** salv-orl_ has quit IRC | 19:45 | |
| hrw | looks like no one built that or it worked magically | 19:46 |
| inc0 | hrw: I am pretty sure we build it in gates;) | 19:47 |
| hrw | inc0: than it is magic | 19:48 |
| hrw | s/than/then | 19:48 |
| hrw | inc0: without adding two devel packages it fails to build scipy from source | 19:49 |
| hrw | hm. same for debian/source | 19:50 |
| hrw | hm. on x86-64 it built for centos, debian, ubuntu. | 19:50 |
| hrw | suspect some weird stuff in repos or sth | 19:51 |
| hrw | no. pip fetches binary version of scipy on x86-64 | 19:52 |
| *** sgordon` is now known as sgordon | 19:52 | |
| inc0 | there you go:) | 19:52 |
| inc0 | I thought it's wierd because I routinelly build whole stack of ubuntu containers | 19:53 |
| inc0 | and I know others do the same for centos | 19:54 |
| inc0 | not sure about debian | 19:54 |
| hrw | no one does debian | 19:54 |
| hrw | otherwise it would work ;d | 19:54 |
| inc0 | well, we need gates | 19:54 |
| inc0 | for debian | 19:55 |
| inc0 | otherwise we can't ensure | 19:55 |
| inc0 | it works | 19:55 |
| hrw | sure | 19:55 |
| hrw | gnocchi container images will be a bit larger then | 19:55 |
| *** jmccarthy has joined #openstack-kolla | 19:56 | |
| hrw | and take long time... | 19:57 |
| hrw | INFO:kolla.image.build.gnocchi-base: Running setup.py install for scipy: still running... | 19:57 |
| inc0 | yeah I remember that in nova some time ago too | 19:57 |
| inc0 | not for scipy tho | 19:58 |
| hrw | on x86-64 pip fetches precompiled python packages often | 19:58 |
| hrw | btw... my stack of patches gets longer and longer while some of them could be merged individually. should I split them out? | 19:59 |
| inc0 | hrw: it's always easier to break down work to multiple patches and try to limit cross-dependency between them | 20:00 |
| *** Jeffrey4l has quit IRC | 20:00 | |
| inc0 | btw, you're Marcin?;) | 20:00 |
| hrw | yes, I am | 20:01 |
| hrw | /whois hrw helps | 20:01 |
| inc0 | hej, Michal z tej strony;) | 20:01 |
| *** Jeffrey4l has joined #openstack-kolla | 20:01 | |
| hrw | inc0: ;D | 20:02 |
| inc0 | ok, yeah, I never really used that irc wizardry | 20:02 |
| hrw | inc0: will you be at OpenStack Day in Warsaw on 22nd March? | 20:02 |
| *** gfidente|afk has quit IRC | 20:02 | |
| hrw | inc0: I do irc for over 20 years ;D | 20:02 |
| inc0 | a bit long flight for me - I live in west coast now:) | 20:02 |
| hrw | ;D | 20:02 |
| hrw | inc0: I also in West... but West North corner of Poland: Szczecin ;d | 20:03 |
| hrw | INFO:kolla.image.build.gnocchi-base: Running setup.py install for scipy: finished with status 'done' | 20:03 |
| hrw | will try to split out some patches tomorrow for easier merging | 20:04 |
| inc0 | I'm from Wroclaw originally | 20:04 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: openstack-base/kolla-toolbox: CentOS has mariadb-devel https://review.openstack.org/444832 | 20:05 |
| hrw | inc0: lived there for ~year | 20:05 |
| *** jheroux has quit IRC | 20:05 | |
| inc0 | best city in Europe;) Portland, where I live now, has pretty similar vibe | 20:06 |
| inc0 | also lots of bridges and hipsters;) | 20:06 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: debian: install gnupg and dirmngr needed for apt-key https://review.openstack.org/434431 | 20:06 |
| *** rhallisey has quit IRC | 20:08 | |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: ironic-pxe: handle non-x86 architectures https://review.openstack.org/434817 | 20:09 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: nova-compute: handle rtslib(-fb) for debian-binary builds too https://review.openstack.org/435941 | 20:10 |
| openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: ceph: move ceph-fuse package to cephfs-fuse image https://review.openstack.org/440634 | 20:12 |
| hrw | now those patches are not depending on other ones - should be easier | 20:12 |
| hrw | let's see how jenkins will fail now | 20:13 |
| *** rabel has joined #openstack-kolla | 20:20 | |
| *** rabel_b1 has joined #openstack-kolla | 20:21 | |
| *** 07EAAIRVF has joined #openstack-kolla | 20:23 | |
| *** jtriley has quit IRC | 20:24 | |
| *** rabel has quit IRC | 20:24 | |
| *** adrian_otto has quit IRC | 20:27 | |
| *** jmccarthy has quit IRC | 20:28 | |
| *** fooliouno has joined #openstack-kolla | 20:41 | |
| fooliouno | Hello .. kolla-kubernetes documentation has a step to create bootstrap for glance-endpoints, but the jinja2 file it is missing. Is this a bug | 20:43 |
| *** jmccarthy has joined #openstack-kolla | 20:45 | |
| *** rwallner has quit IRC | 20:47 | |
| *** rwallner has joined #openstack-kolla | 20:49 | |
| *** rabel_b1 has quit IRC | 20:51 | |
| *** 07EAAIRVF has quit IRC | 20:52 | |
| *** rwallner has quit IRC | 20:53 | |
| *** eaguilar has quit IRC | 20:54 | |
| *** satyar has joined #openstack-kolla | 20:59 | |
| *** gfhellma has quit IRC | 21:00 | |
| *** jmccarthy has quit IRC | 21:02 | |
| Rowan_ | Is there any wayt to rebuild the faulty container with existing config? | 21:07 |
| *** Rowan_ has quit IRC | 21:14 | |
| *** sayantan_ has quit IRC | 21:15 | |
| *** sayantan_ has joined #openstack-kolla | 21:20 | |
| *** srwilkers has quit IRC | 21:22 | |
| *** salv-orlando has joined #openstack-kolla | 21:30 | |
| *** Jeffrey4l has quit IRC | 21:35 | |
| *** satyar has quit IRC | 21:35 | |
| *** Jeffrey4l has joined #openstack-kolla | 21:35 | |
| *** fooliouno has quit IRC | 21:36 | |
| *** papacz has quit IRC | 21:46 | |
| *** Mannheim has joined #openstack-kolla | 21:49 | |
| *** jrobinson has joined #openstack-kolla | 21:55 | |
| *** schwicht has quit IRC | 21:56 | |
| *** rwallner has joined #openstack-kolla | 21:59 | |
| *** Serlex has quit IRC | 21:59 | |
| *** rwallner has quit IRC | 22:03 | |
| sbezverk | fooliouno: we are not using jinj2 templates any longer, please consider using helm charts. | 22:11 |
| *** manheim has joined #openstack-kolla | 22:30 | |
| *** Mannheim has quit IRC | 22:33 | |
| *** dave-mccowan has quit IRC | 22:36 | |
| inc0 | yay I thin I know whats wrong with my k8s | 22:48 |
| inc0 | since it timeouts on 443 and not secure connection works | 22:49 |
| inc0 | my question is | 22:49 |
| inc0 | how exactly kube passes self signed CA certs to containers? | 22:49 |
| SamYaple | inc0: is that a question in general or a kolla-kubernetes question? | 22:59 |
| inc0 | general | 22:59 |
| inc0 | I deployed halcyon and from within container kube-api-ip:443 is no bueno | 22:59 |
| *** srwilkers has joined #openstack-kolla | 23:00 | |
| SamYaple | im not sure it does pass in self-signed certs in general.... | 23:01 |
| SamYaple | production deploys are not using self-signed | 23:01 |
| inc0 | production deploys have ca authority somewhere? | 23:02 |
| SamYaple | non-production can easily just ignore and not verify | 23:02 |
| SamYaple | you can get real certs for all the controllers | 23:02 |
| SamYaple | thats what I do | 23:02 |
| inc0 | well wget https://10.0.2.1:443 times out | 23:02 |
| SamYaple | times out is unrelated to the cert | 23:03 |
| inc0 | right....iptables or fragmented vxlans? | 23:04 |
| SamYaple | are you, you know, sure you launched the api with ssl on port 443? | 23:04 |
| SamYaple | i think 8081 is the other "normal" https port for the api | 23:05 |
| inc0 | if I wouldn't, that would be connection refused right? | 23:05 |
| SamYaple | no | 23:05 |
| SamYaple | because of the fact that ip address doesnt live anywhere | 23:05 |
| SamYaple | its just iptables weirdness | 23:05 |
| inc0 | well, 8081 did refuse my connection | 23:05 |
| SamYaple | do random ports refuse or timeout? | 23:07 |
| inc0 | refuse | 23:07 |
| SamYaple | it really sounds like you aren't listening properly | 23:07 |
| SamYaple | how did you launch the api? | 23:07 |
| SamYaple | cli options wise | 23:08 |
| inc0 | riiightsume | 23:10 |
| inc0 | right | 23:10 |
| inc0 | so, you | 23:10 |
| inc0 | you're right, it does deploy api with secure port 6443 | 23:11 |
| inc0 | question is, why halcyon doesn't do it properly | 23:11 |
| SamYaple | i was right? man i just guessed for all of that | 23:13 |
| *** ipsecguy_ has quit IRC | 23:13 | |
| SamYaple | made it all up | 23:13 |
| SamYaple | yay me | 23:13 |
| *** ipsecguy has joined #openstack-kolla | 23:13 | |
| inc0 | well | 23:14 |
| inc0 | I | 23:14 |
| inc0 | I'm nto sure yet, but somethigns funky out there | 23:14 |
| *** lamt has quit IRC | 23:15 | |
| SamYaple | i do know you need to launch with ssl in mind | 23:16 |
| SamYaple | otherwise its all going to be http from the container | 23:16 |
| SamYaple | or pod | 23:16 |
| SamYaple | WHATEVER | 23:16 |
| *** jrobinson has quit IRC | 23:18 | |
| *** rwellum has quit IRC | 23:19 | |
| *** lucasxu has quit IRC | 23:19 | |
| *** Pavo has joined #openstack-kolla | 23:20 | |
| *** Pavo has quit IRC | 23:21 | |
| jemcevoy | inc0 - I have run into another issue with ceilometer. This time an SSL error during install. Do you know what the fix might be? http://paste.openstack.org/show/602559/ | 23:22 |
| inc0 | ssl... | 23:23 |
| *** rhallisey has joined #openstack-kolla | 23:23 | |
| *** srwilkers has quit IRC | 23:24 | |
| *** erlon has quit IRC | 23:25 | |
| jemcevoy | I did the certificate generate thing for a self sighed cert... I plan to setup dogtag CA with FreeIPA but have not finished a real CA... I see that you and SamYaple are discussing ssl issues too | 23:27 |
| *** japestinho has quit IRC | 23:30 | |
| *** saneax-_-|AFK is now known as saneax | 23:35 | |
| *** sayantan_ has quit IRC | 23:41 | |
| spsurya | morning all | 23:41 |
| *** sayantan_ has joined #openstack-kolla | 23:49 | |
| sdake | jascott1 grats on the house dude ;-) | 23:53 |
| jascott1 | hey sdake whats up? | 23:54 |
| jascott1 | thanks, yeah we are waiting on the appraisal results but its looking good so far | 23:54 |
| *** sayantan_ has quit IRC | 23:54 | |
| *** spsurya__ has joined #openstack-kolla | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
