| *** yifei has joined #openstack-ansible | 00:01 | |
| *** yifei has quit IRC | 00:05 | |
| *** mwynne has joined #openstack-ansible | 00:25 | |
| *** yifei has joined #openstack-ansible | 00:27 | |
| *** chyka has quit IRC | 00:30 | |
| *** savvas_ has joined #openstack-ansible | 00:44 | |
| *** savvas_ has quit IRC | 00:45 | |
| *** savvas_ has joined #openstack-ansible | 00:46 | |
| *** masber has quit IRC | 00:47 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible master: [WIP] Implement iptables management https://review.openstack.org/512099 | 00:49 |
|---|---|---|
| *** savvas__ has joined #openstack-ansible | 00:50 | |
| *** savvas__ has quit IRC | 00:50 | |
| *** savvas__ has joined #openstack-ansible | 00:51 | |
| *** savvas_ has quit IRC | 00:53 | |
| *** hw_wutianwei has joined #openstack-ansible | 01:04 | |
| *** xingchao has joined #openstack-ansible | 01:08 | |
| *** xingchao has quit IRC | 01:18 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] Add nspawn container driver https://review.openstack.org/477017 | 01:31 |
| openstackgerrit | Cuong Nguyen proposed openstack/openstack-ansible-os_glance master: Use glance_services to get group name https://review.openstack.org/520989 | 01:34 |
| cloudnull | so I dissected our recent tempest failures. | 01:37 |
| cloudnull | http://paste.openstack.org/show/627157/ | 01:37 |
| cloudnull | looks like we need to put the glance v1 api back | 01:37 |
| cloudnull | ... | 01:37 |
| cloudnull | odyssey4me: evrardjp: hwoarang: logan-: spotz: andymccr: d34dh0r53: jmccrory: mhayden: ^ | 01:38 |
| cloudnull | I have to run | 01:38 |
| cloudnull | if i get time this weekend I'll try and take a look | 01:38 |
| cloudnull | however if someone can try and figure out what is required to either make that test pass or re-instate the glance API it'd be appreciated. | 01:39 |
| spotz | cloudnull I'll poke and see if I can't narrow it if I have time. Gotta finish my course still by next week:( | 01:39 |
| spotz | cloudnull: I do wonder if it just needs to be configured to use the v2 api somewhere | 01:40 |
| *** savvas_ has joined #openstack-ansible | 01:41 | |
| *** savvas__ has quit IRC | 01:42 | |
| cloudnull | this might be. | 01:43 |
| cloudnull | sadly I've got to go and cant look into it ATM. | 01:44 |
| cloudnull | :( | 01:44 |
| *** pester has joined #openstack-ansible | 01:54 | |
| *** fxpester has quit IRC | 01:55 | |
| *** chyka has joined #openstack-ansible | 01:57 | |
| *** japestinho has joined #openstack-ansible | 02:00 | |
| *** xingchao has joined #openstack-ansible | 02:01 | |
| *** chyka has quit IRC | 02:02 | |
| *** xingchao has quit IRC | 02:02 | |
| *** savvas_ has quit IRC | 02:18 | |
| *** savvas_ has joined #openstack-ansible | 02:18 | |
| *** pmannidi has quit IRC | 02:19 | |
| *** pmannidi has joined #openstack-ansible | 02:20 | |
| openstackgerrit | Jimmy McCrory proposed openstack/openstack-ansible-os_tempest master: Disable glance v1 API by default https://review.openstack.org/522427 | 02:26 |
| *** udesale has joined #openstack-ansible | 02:29 | |
| *** gloksinia has joined #openstack-ansible | 02:30 | |
| *** masber has joined #openstack-ansible | 02:38 | |
| *** masuberu has joined #openstack-ansible | 02:42 | |
| gloksinia | Hello, can i ask a question here ?? | 02:47 |
| *** masber has quit IRC | 02:47 | |
| mattt | gloksinia: go for it! someone will answer as soon as they're able to | 02:54 |
| gloksinia | matt: thank you. | 03:00 |
| *** savvas_ has quit IRC | 03:04 | |
| gloksinia | I want to know about deploying openstack-ansible with Openvswtich. I can't make it work at all. | 03:05 |
| gloksinia | Forgot it i try it on pike. | 03:07 |
| gloksinia | So my questions is after following https://docs.openstack.org/openstack-ansible-os_neutron/latest/app-openvswitch.html this documentation after successful deploy. | 03:07 |
| gloksinia | I need to create all port and bridge by myself to make it work. i saw only br-int with ovs-vsctl show command and i cant bring it up. | 03:10 |
| *** gie has joined #openstack-ansible | 03:40 | |
| *** gie has quit IRC | 03:41 | |
| *** gie has joined #openstack-ansible | 03:43 | |
| *** lbragstad has joined #openstack-ansible | 03:46 | |
| *** lbragstad has quit IRC | 03:52 | |
| *** poopcat has quit IRC | 04:01 | |
| *** Dinesh_Bhor has joined #openstack-ansible | 04:03 | |
| *** lbragstad has joined #openstack-ansible | 04:04 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-plugins master: Add basic plugin support for nspawn type containers https://review.openstack.org/490689 | 04:07 |
| logan- | which branch are you seeing that on cloudnull | 04:08 |
| cloudnull | the tempest issue? | 04:09 |
| cloudnull | that's in master. | 04:10 |
| logan- | k | 04:10 |
| cloudnull | looks like jmccrory put up a patch for it already | 04:10 |
| cloudnull | https://review.openstack.org/#/c/522427 | 04:10 |
| openstackgerrit | Logan V proposed openstack/openstack-ansible-os_tempest master: Disable Glance v1 API usage by default https://review.openstack.org/522435 | 04:11 |
| logan- | oh | 04:11 |
| logan- | lol | 04:12 |
| logan- | yeah sure enough | 04:12 |
| cloudnull | If that's all we need, it'd be wonderful :D | 04:12 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Modify LXC container create to JIT create https://review.openstack.org/490976 | 04:15 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] Add nspawn container driver https://review.openstack.org/477017 | 04:19 |
| *** lbragstad has quit IRC | 04:34 | |
| *** gkadam has joined #openstack-ansible | 04:36 | |
| *** germs1 has joined #openstack-ansible | 04:41 | |
| *** germs has quit IRC | 04:41 | |
| *** cjloader has joined #openstack-ansible | 04:41 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] Add nspawn container driver https://review.openstack.org/477017 | 04:46 |
| *** hybridpollo has quit IRC | 05:09 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] Add nspawn container driver https://review.openstack.org/477017 | 05:11 |
| *** germs1 has quit IRC | 05:22 | |
| *** germs has joined #openstack-ansible | 05:22 | |
| *** chyka has joined #openstack-ansible | 05:27 | |
| *** nshetty has joined #openstack-ansible | 05:30 | |
| *** nshetty_ has joined #openstack-ansible | 05:42 | |
| *** nshetty has quit IRC | 05:42 | |
| *** udesale has quit IRC | 05:44 | |
| *** chyka has quit IRC | 05:46 | |
| *** udesale has joined #openstack-ansible | 05:53 | |
| *** udesale has quit IRC | 05:53 | |
| *** udesale has joined #openstack-ansible | 05:54 | |
| *** gokhan has quit IRC | 05:59 | |
| *** gokhan has joined #openstack-ansible | 06:01 | |
| *** germs has quit IRC | 06:06 | |
| openstackgerrit | Logan V proposed openstack/openstack-ansible master: [WIP] Implement iptables management https://review.openstack.org/512099 | 06:07 |
| *** wspthr has quit IRC | 06:07 | |
| *** threestrands has quit IRC | 06:10 | |
| *** wspthr has joined #openstack-ansible | 06:10 | |
| *** pradiprwt has joined #openstack-ansible | 06:10 | |
| pradiprwt | Hi Everyone, I am facing problem in installing galera, here is console output.. http://paste.openstack.org/show/627170/ | 06:15 |
| pradiprwt | can anyone please help me | 06:15 |
| *** hamza has joined #openstack-ansible | 06:16 | |
| *** m|y|k has joined #openstack-ansible | 06:19 | |
| *** gokhan has quit IRC | 06:22 | |
| cloudnull | pradiprwt normally this is a load balancer issue | 06:23 |
| cloudnull | is the lb up? | 06:23 |
| cloudnull | and can you curl http://172.29.236.9:8181 ? | 06:23 |
| *** gokhan has joined #openstack-ansible | 06:23 | |
| pradiprwt | cloudnull: yeah, you are right load balancer is not reacheble | 06:25 |
| pradiprwt | Is there any way to make it up ? | 06:25 |
| cloudnull | are you running haproxy | 06:26 |
| cloudnull | you can simply rerun the haproxy playbook if something is off there. | 06:26 |
| cloudnull | if you're using an external LB you'll need to resolve that first. | 06:27 |
| pradiprwt | cloudnull: sure I will run now | 06:27 |
| *** nshetty_ is now known as nshetty | 06:31 | |
| *** hamza has quit IRC | 06:33 | |
| pradiprwt | cloudnull: Still facing same problem after running haproxy playbook, I got this issue while running "openstack-ansible setup-infrastructure.yml" | 06:33 |
| cloudnull | what, if anything in the haproxy playbook changed? | 06:34 |
| cloudnull | is haproxy running on the host? | 06:34 |
| yifei | Hi all, I always failed when run setup-infrastructure. The actual failure is that it failed to install nginx in repo lxc. "Public key for gperftools-libs-2.4-7.el7.x86_64.rpm is not installed". The OS of hosts is centos7.4. | 06:35 |
| yifei | If I manually installed nginx with nogpgcheck, it will also failed at some else place e.g. installing memchached | 06:36 |
| pradiprwt | cloudnull: yes, haproxy is running on all infra hosts | 06:36 |
| *** hamza has joined #openstack-ansible | 06:36 | |
| pradiprwt | haproxy playbook executed without any error | 06:37 |
| yifei | So if I missed some configuration related to yum repos | 06:37 |
| yifei | Is there anybody can help me with this? Thanks | 06:37 |
| *** gkadam has quit IRC | 06:59 | |
| pradiprwt | cloudnull: here is log after restarting haproxy http://paste.openstack.org/show/627182/ | 07:02 |
| *** cjloader has quit IRC | 07:03 | |
| *** jvidal has joined #openstack-ansible | 07:11 | |
| *** michelv has joined #openstack-ansible | 07:13 | |
| *** jvidal has quit IRC | 07:26 | |
| *** nshetty is now known as nshetty|brb | 07:30 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/pike: Set the nova-lxd lib to use PIKE https://review.openstack.org/522155 | 07:30 |
| *** nshetty|brb is now known as nshetty|lunch | 07:39 | |
| *** jvidal has joined #openstack-ansible | 07:40 | |
| *** pcaruana has joined #openstack-ansible | 07:45 | |
| *** shardy has joined #openstack-ansible | 08:02 | |
| *** armaan has joined #openstack-ansible | 08:05 | |
| *** mbuil has joined #openstack-ansible | 08:06 | |
| evrardjp | cloudnull: still around? | 08:06 |
| evrardjp | jmccrory: thanks again! | 08:08 |
| evrardjp | odyssey4me: or andymccr are you around? | 08:09 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: [WIP] Add nspawn container driver https://review.openstack.org/477017 | 08:10 |
| cloudnull | evrardjp: about to pass out. | 08:10 |
| evrardjp | :) | 08:11 |
| cloudnull | have a good night ! | 08:11 |
| evrardjp | you've deserved some sleep, go to bed! | 08:11 |
| openstackgerrit | Dimitrios Markou proposed openstack/openstack-ansible-os_neutron master: Add OvS-NSH support https://review.openstack.org/517259 | 08:11 |
| *** gie has quit IRC | 08:11 | |
| *** gloksinia has quit IRC | 08:11 | |
| armaan | morning folks, i was wondering if anybody here used keystone with kubernetes? | 08:12 |
| *** holmsten has joined #openstack-ansible | 08:18 | |
| evrardjp | I haven't | 08:21 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_ceilometer stable/newton: Initial OSA zuul v3 role jobs https://review.openstack.org/513999 | 08:24 |
| evrardjp | armaan: would be interested though :) | 08:24 |
| armaan | evrardjp: I just found this https://github.com/openstack/stackube | 08:25 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_ceilometer stable/newton: Initial OSA zuul v3 role jobs https://review.openstack.org/513999 | 08:25 |
| evrardjp | oh yeah heard about it | 08:26 |
| evrardjp | They thought about including it in the list of official projects, but was temporarily not added | 08:26 |
| evrardjp | Although to me, it looked like an ops nightmare | 08:26 |
| *** gkadam has joined #openstack-ansible | 08:28 | |
| pradiprwt | evrardjp: Hi , I am getting problem in installation of haproxy, can you please help me. Galera is getting failed because if haproxy vip is not reachable | 08:31 |
| evrardjp | pradiprwt: sure | 08:31 |
| evrardjp | I will ask the usual question: Is your external_lb_vip_address the same as internal_lb_vip_address? | 08:32 |
| pradiprwt | no | 08:32 |
| evrardjp | or resolves to the same IP? | 08:32 |
| *** pmannidi has quit IRC | 08:32 | |
| pradiprwt | internal is from management network and external is from flat network | 08:33 |
| pradiprwt | evrardjp: /etc/openstack_deploy/openstack_user_config.yml http://paste.openstack.org/show/627190/ | 08:35 |
| armaan | evrardjp: Yeah, i agree. | 08:39 |
| armaan | evrardjp: Found one more effort in this direction https://github.com/dims/k8s-keystone-auth | 08:39 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_zaqar stable/ocata: [WIP] Initial Zuul v3 role jobs https://review.openstack.org/514299 | 08:40 |
| pradiprwt | evrardjp : now it is working | 08:41 |
| evrardjp | you should ask dims then :) | 08:41 |
| evrardjp | pradiprwt: I didn't change anything :) | 08:41 |
| pradiprwt | evrardjp: I did some changes in user variable file :) | 08:41 |
| evrardjp | ... what was it? | 08:42 |
| pradiprwt | I was providing keepalived without vip only subnet , and by mistake I kept some user variable values in openstack user config file... after removing it started working | 08:43 |
| *** nshetty|lunch is now known as nshetty | 08:48 | |
| openstackgerrit | Gökhan proposed openstack/openstack-ansible-os_aodh stable/pike: Change auth config to keystone v3 according to openstack pike aodh docs[1], we modified auth config under the "keystone_authtoken" and "service_credentials" groups. [1] https://docs.openstack.org/aodh/pike/install/ \ install-rdo.html#install-and-configure https://review.openstack.org/522480 | 08:50 |
| *** flaviosr has quit IRC | 08:52 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Disable glance v1 API by default https://review.openstack.org/522427 | 08:54 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_zaqar stable/ocata: [WIP] Initial Zuul v3 role jobs https://review.openstack.org/514299 | 08:57 |
| *** flaviosr has joined #openstack-ansible | 09:01 | |
| pradiprwt | evrardjp: HAproxy is working but still I am getting issue with galera, here is console output >> http://paste.openstack.org/show/627193/ | 09:01 |
| andymccr | evrardjp: o/ | 09:01 |
| *** jvidal has quit IRC | 09:02 | |
| evrardjp | andymccr: woot! | 09:02 |
| evrardjp | andymccr: may I get a vote on those: https://review.openstack.org/#/c/521566/ https://review.openstack.org/#/c/522314/ https://review.openstack.org/#/c/522386/ https://review.openstack.org/#/c/514289/ https://review.openstack.org/#/c/513999/ ? They clutter my inbox. | 09:03 |
| andymccr | evrardjp: sure will go through those this morning | 09:03 |
| evrardjp | just be fairly straightforward | 09:04 |
| evrardjp | pradiprwt: Failed to establish a new connection: [Errno 113] No route to host | 09:04 |
| andymccr | everytime i see that ceph_client role i hate it more :P | 09:04 |
| evrardjp | can't do much for you that's a networking issue. | 09:04 |
| evrardjp | well it's just to prevent the recursion depth issue :p | 09:05 |
| evrardjp | But yes. | 09:05 |
| andymccr | yeah sure but the way that was designed | 09:05 |
| pradiprwt | I have verified the network everything is working fine | 09:05 |
| evrardjp | You're now an expert, maybe we can do better :D | 09:05 |
| evrardjp | pradiprwt: it is not fine | 09:05 |
| andymccr | evrardjp: on that rally patch - thats going to be a pain to maintain. i guess there is no other way at all? | 09:05 |
| evrardjp | ? | 09:05 |
| evrardjp | oh yeah | 09:06 |
| evrardjp | no | 09:06 |
| evrardjp | no other way | 09:06 |
| evrardjp | I don't want to have a fight with infra and kolla ansible about putting ansible in our requirements and therefore in global requirements :p | 09:06 |
| evrardjp | there are many different ways to do this, and trust me, this one will be the easiest :) | 09:07 |
| andymccr | that would only be dev mode though right? | 09:08 |
| * hwoarang wonders what is up with teh xenial tempest failures | 09:08 | |
| andymccr | why does rally need ansible as a pip install? | 09:08 |
| pradiprwt | evrardjp: Yeah, there is issue with network, host is able to access the url but container is not. Do you have any idea what I need to verify ? | 09:08 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Integrate Congress with OSA. https://review.openstack.org/503971 | 09:09 |
| evrardjp | andymccr: that's part of their dependency | 09:09 |
| andymccr | ahh ok | 09:09 |
| evrardjp | andymccr: basically they have os-idontknowthename | 09:10 |
| evrardjp | that thing implements an ansible wrapper | 09:10 |
| evrardjp | it has in its requirements, ansible | 09:10 |
| evrardjp | so installing rally gets the whole chain which installs ansible | 09:10 |
| evrardjp | so ansible gets pulled unconstrained, and magic happens. | 09:10 |
| evrardjp | pradiprwt: the bridges on the different hosts? | 09:11 |
| andymccr | thats terribad. ok well thats gonna be a fun one when we move to ansible 2.4 etc. cant think of a better way though! | 09:12 |
| odyssey4me | andymccr it goes into the rally venv, so it doesn't affect our installation | 09:13 |
| andymccr | odyssey4me: sure but we're setting a version inside that role, meaning we will have to remember to update that at some point in the future | 09:13 |
| andymccr | or tests will fail and weirdness will happen | 09:13 |
| pradiprwt | evrardjp :br-mgmt, br-vlan, br-vxlan and br-storage I am using --- br-mgmt is on all hosts with IP and containers are also in same range | 09:14 |
| andymccr | if rally themselves wanted to propose a version of ansible required, since they require it, we could get that from their repo instead | 09:14 |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client stable/pike: Remove one layer of includes https://review.openstack.org/521566 | 09:14 |
| *** jvidal has joined #openstack-ansible | 09:15 | |
| odyssey4me | oh yeah - well, meh :p | 09:15 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 09:17 |
| *** pbandark has joined #openstack-ansible | 09:19 | |
| evrardjp | pradiprwt: proper ip ranges? proper routes? | 09:19 |
| evrardjp | odyssey4me: we have weird thing happening in gates | 09:20 |
| evrardjp | I have things that work on my machine that don't work in gates | 09:20 |
| evrardjp | ceilometer for example (https://review.openstack.org/#/c/513999/ ) works on my trusty, but doesn't in gates | 09:20 |
| evrardjp | for a lxc_container_create issue IIRC | 09:20 |
| evrardjp | zaqar patchset 6 here: https://review.openstack.org/#/c/514299/6/ worked too on my machine. Which makes me wonder what's the difference. | 09:21 |
| evrardjp | For zaqar I thought it maybe due to the user that runs the stuff (root vs unprivileged user) | 09:21 |
| evrardjp | oh no ceilometer was a service that can't restart, my bad. the lxc_container_create issue is for zaqar | 09:22 |
| odyssey4me | well, different base image - different mirrors | 09:22 |
| *** pbandark has quit IRC | 09:22 | |
| evrardjp | still, different behavior | 09:22 |
| evrardjp | mmm. | 09:23 |
| evrardjp | for ceilometer, I tried on rax public cloud and a vagrant image | 09:23 |
| evrardjp | those trusty worked. | 09:23 |
| odyssey4me | ok, but the zaqar one is failing lint tests | 09:23 |
| evrardjp | I was wondering if that couldn't be a clone issue | 09:23 |
| *** hamza has quit IRC | 09:23 | |
| evrardjp | odyssey4me: the zaqar patchset 6 was not failing for that. It is now, because I try to mess up with it. | 09:24 |
| *** hamza has joined #openstack-ansible | 09:24 | |
| evrardjp | odyssey4me: have a look at the run Patch Set 7: Build failed (check pipeline). For information on how to proceed, see http://docs.openstack.org/infra/manual/developers.html#automated-testing - build-openstack-sphinx-docs http://logs.openstack.org/99/513999/7/check/build-openstack-sphinx-docs/4fd658c/html/ : SUCCESS in 1m 46s - build-openstack-releasenotes http://logs.openstack.org/99/513999/7/chec | 09:24 |
| evrardjp | k/build-openstack-releasenotes/0b87538/html/ : SUCCESS in 2m 38s - openstack-ansible-linters http://logs.openstack.org/99/513999/7/check/openstack-ansible-linters/685eb08/ : SUCCESS in 4m 21s - openstack-ansible-functional-ubuntu-xenial http://logs.openstack.org/99/513999/7/check/openstack-ansible-functional-ubuntu-xenial/02e6673/ : SUCCESS in 30m 31s - openstack-ansible-functional-ubuntu-trusty h | 09:24 |
| evrardjp | ttp://logs.openstack.org/99/513999/7/check/openstack-ansible-functional-ubuntu-trusty/4b78804/ : FAILURE in 25m 30s22/11 14:13 | 09:24 |
| evrardjp | wow. | 09:24 |
| evrardjp | sorry for the wrong paste. | 09:24 |
| evrardjp | well I guess it works :p | 09:24 |
| odyssey4me | oh yes, something else which affects the older branches is the use of conflicting ip's in the inventory | 09:25 |
| odyssey4me | ip's that conflict with some of the providers | 09:25 |
| odyssey4me | or are inconsistent between the tests repo and the inventory | 09:26 |
| evrardjp | ok so I was heading the right direction on my latest patchsets then | 09:26 |
| evrardjp | I am glad I asked you | 09:26 |
| odyssey4me | heh, that's a fun one: http://logs.openstack.org/99/514299/6/check/openstack-ansible-functional-ubuntu-xenial/053d858/job-output.txt.gz#_2017-11-22_20_40_41_152023 | 09:26 |
| odyssey4me | when dropping the network interfaces config, so yeah - seems like that might relate | 09:27 |
| evrardjp | yeah is that the same patch? | 09:29 |
| evrardjp | why do you have different numbers? | 09:29 |
| odyssey4me | different numbers? | 09:30 |
| evrardjp | damn I mispasted again. | 09:30 |
| evrardjp | yeah whatever. | 09:30 |
| evrardjp | wrong morning I guess. | 09:30 |
| evrardjp | well it's already 10:30 but still! | 09:31 |
| evrardjp | yeah that's what I meant, a weird thing going on with the dropping the network interfaces | 09:31 |
| evrardjp | as you can see it's directly an ansible failure | 09:31 |
| evrardjp | so nothing directly related to our code | 09:31 |
| evrardjp | it's 100% on the runtime environement | 09:31 |
| evrardjp | which makes me wonder what else do we have right now | 09:31 |
| evrardjp | so ceilometer failed due to a runtime thing, now zaqar... | 09:32 |
| pradiprwt | evrardjp :br-mgmt, br-vlan, br-vxlan and br-storage, I am using --- br-mgmt is on all hosts with IP and containers are also in same range | 09:34 |
| evrardjp | pradiprwt: can you, from one container, ping another container? | 09:34 |
| pradiprwt | let me check | 09:34 |
| pradiprwt | yes I am able to ping from one container, ping another container | 09:36 |
| pradiprwt | but not able to ping host IP | 09:37 |
| openstackgerrit | Merged openstack/openstack-ansible-os_rally master: Constraint ansible installation https://review.openstack.org/522314 | 09:39 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_rally stable/pike: Constraint ansible installation https://review.openstack.org/522501 | 09:39 |
| pradiprwt | evrardjp : br-mgmt doesn't have gateway, this is dummy virtual network using vlan | 09:40 |
| evrardjp | I don't know what you're doing and what you want to achieve, but I can tell you that you have a networking issue for what you are asking ansible to do. | 09:40 |
| *** nshetty is now known as nshetty|BMGR | 09:41 | |
| pradiprwt | evrardjp : I want to install openstack with HA using ansible, I am following official documentation, but I am facing some issue ... Here is my interface file http://paste.openstack.org/show/627196/ | 09:44 |
| evrardjp | I think vxlan is missing an IP, first | 09:45 |
| evrardjp | not that it's your problem right now | 09:45 |
| pradiprwt | evrardjp : this is infra node interface file, do I need to keep here ? | 09:46 |
| evrardjp | br-mgmt doesn't seem to have internet access like that, don't you need to do other things? | 09:46 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer stable/newton: Initial OSA zuul v3 role jobs https://review.openstack.org/513999 | 09:47 |
| evrardjp | you could do nat and all on your node, maybe? openstack-ansible is barely using the network connectivity on the hosts (through their bridges) and then install openstack on top of it. But I can suggest you to have a look at our limited connectivity guide if you want to follow that route | 09:48 |
| pradiprwt | For internet I have provided the interface enp0s3, and for openstack provider network there is interface enp0s9 | 09:49 |
| evrardjp | yeah that looks great | 09:49 |
| evrardjp | I think the only part missing is that your node would have to do nat for the mgmt network | 09:49 |
| evrardjp | because the management network isn't bridged to anything | 09:50 |
| evrardjp | if you would have brought br-mgmt to the same network as what's in enp0s3 (using a bridge port there) it could have been different | 09:51 |
| evrardjp | It all depends on your use case | 09:51 |
| pradiprwt | I don't have much IPs in enp0s3 range | 09:51 |
| pradiprwt | This is just POC before production :) | 09:52 |
| evrardjp | then you should nat :) | 09:52 |
| pradiprwt | ok | 09:52 |
| evrardjp | doesn't it makes sense? | 09:53 |
| *** hw_wutianwei has quit IRC | 09:53 | |
| pradiprwt | I think nating is fro providing internet access to that range or interface | 09:53 |
| pradiprwt | In my environment host and containers both have internet acces | 09:54 |
| evrardjp | do they? | 09:54 |
| odyssey4me | br-mgmt isn't used for internet access, that will be accessed through the lxcbr0 interface which will nat through the host | 09:54 |
| evrardjp | oh yeah that's true, that's what we do. | 09:55 |
| evrardjp | i always forget this one. | 09:55 |
| pradiprwt | yes, containers have internet because they use lxcbr0 for internet | 09:55 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/ocata: Depth argument is prome to failure for ansible<2.3 https://review.openstack.org/522386 | 09:56 |
| evrardjp | so pradiprwt I don't know. You have clearly an issue with networking though. "No route to host" is clear to me. | 09:56 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-tests stable/newton: Depth argument is prome to failure for ansible<2.3 https://review.openstack.org/522508 | 09:56 |
| pradiprwt | I will try to trace the packet where it is getting drooped | 09:56 |
| mardim | hwoarang, evrardjp odyssey4me cloudnull the 'Do not merge' barrier is lifted https://review.openstack.org/#/c/517259/ :) | 09:58 |
| evrardjp | haha cool | 09:58 |
| mardim | :) | 09:58 |
| evrardjp | wow and it passes tests? | 09:58 |
| evrardjp | :) | 09:58 |
| mardim | Yes and i created a =n extra gate | 09:59 |
| mardim | :) | 09:59 |
| mardim | an* | 09:59 |
| evrardjp | ok so I need an extra tough review then :p | 10:00 |
| mardim | haha yes :) | 10:01 |
| openstackgerrit | Merged openstack/openstack-ansible-os_glance master: Use glance_services to get group name https://review.openstack.org/520989 | 10:03 |
| * hwoarang sends mardim a chocolate cookie and a -1 | 10:04 | |
| *** electrofelix has joined #openstack-ansible | 10:05 | |
| hwoarang | odyssey4me: mhayden: evrardjp: a centos fix please https://review.openstack.org/#/c/521882/ :) | 10:05 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic stable/ocata: Initial Zuul v3 role jobs https://review.openstack.org/514289 | 10:06 |
| odyssey4me | hwoarang hmm, nice catch | 10:06 |
| odyssey4me | not quite sure I like having it done there, but don't have a better suggestion right now | 10:07 |
| odyssey4me | the only thing I can think of is that it should probably be included in a bunch of the role distro packages as it's clearly a requirement for each of the services | 10:08 |
| * mardim mardim will eat the chocolate and he will be furious for the -1 :P | 10:08 | |
| mardim | hwoarang, ^ | 10:08 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Do not implement reload for novnc/spice service https://review.openstack.org/522510 | 10:08 |
| hwoarang | odyssey4me: yeah but pip_install is a meta dependency :p | 10:09 |
| odyssey4me | sure, but one day it won't be | 10:10 |
| *** ThomasS has joined #openstack-ansible | 10:10 | |
| hwoarang | yeah i know :) | 10:10 |
| ThomasS | hi | 10:10 |
| ThomasS | i am struggeling with a deployment of 7 nodes with ceph integration | 10:11 |
| ThomasS | i am using this guide | 10:11 |
| ThomasS | https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/app-config-prod-ceph.html | 10:11 |
| ThomasS | tag 16.0.3 | 10:12 |
| ThomasS | everything deployed fine so far | 10:12 |
| ThomasS | the running of the openstack-ansible setup-openstack.yml | 10:13 |
| ThomasS | error happens at TASK [os_cinder : Perform a cinder DB sync] | 10:13 |
| ThomasS | the error looks strange: | 10:13 |
| ThomasS | https://pastebin.com/p2Bnb0M0 | 10:13 |
| *** nshetty|BMGR is now known as nshetty | 10:13 | |
| ThomasS | maybe you can give me hint | 10:14 |
| odyssey4me | ThomasS I'm not sure why our pike docs haven't been updated with these merges, but these fixes to the docs are important to know for that example: https://github.com/openstack/openstack-ansible/commit/6249a8c9e5a334ba2e4bd9880a8351257802627f & https://github.com/openstack/openstack-ansible/commit/83e2b5415b50119eecc651fee4839ad25c5b810d | 10:14 |
| evrardjp | mardim: reviewed! | 10:14 |
| mardim | evrardjp, thanks :) | 10:15 |
| *** mrch has joined #openstack-ansible | 10:15 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_cinder master: Use cinder_services dict to identify cinder-api service name https://review.openstack.org/522511 | 10:15 |
| odyssey4me | ThomasS wow, that's not normal... I wonder what went wrong there | 10:15 |
| evrardjp | mardim: I didn't send a chocolate with my -1, but you get the idea that I wanted to send it. | 10:15 |
| evrardjp | oh no, my bad, I only send chocolates when +2 ! | 10:16 |
| evrardjp | :p | 10:16 |
| *** armaan has quit IRC | 10:16 | |
| ThomasS | also the queens docs are not up to date | 10:16 |
| ThomasS | https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/app-config-prod-ceph.html | 10:16 |
| mardim | evrardjp, the +2 chocolate is sweeter :) | 10:16 |
| ThomasS | the fixes are not there you mentioned | 10:16 |
| *** armaan has joined #openstack-ansible | 10:17 | |
| odyssey4me | ThomasS for queens only https://github.com/openstack/openstack-ansible/commit/33836d3d5ef7ec20b77ffd8082feb0f145eb294f is relevant | 10:17 |
| ThomasS | do not get me wrong | 10:17 |
| odyssey4me | so those are up to date it seems | 10:17 |
| evrardjp | odyssey4me: do we have a docs publication issue again? | 10:18 |
| ThomasS | i was using the queens config example because pike has none | 10:18 |
| ThomasS | my tag is 16.03 | 10:18 |
| ThomasS | so installing and following the pike deployment guide with 16.0.3 and using the queens ceph example | 10:18 |
| odyssey4me | yes, so for pike you need the ceph_mons bit until we can merge a patch into pike which makes it work like queens | 10:19 |
| openstackgerrit | Merged openstack/openstack-ansible-pip_install master: vars: Explicitly pull the openssl package https://review.openstack.org/521882 | 10:20 |
| odyssey4me | evrardjp FYI our pike docs don't seem to have been published in a while - I'm following up with infra | 10:20 |
| odyssey4me | master seems up to date | 10:20 |
| evrardjp | Oh that's great. | 10:21 |
| ThomasS | so i rerun | 10:21 |
| ThomasS | https://pastebin.com/nHhuiC2z | 10:21 |
| evrardjp | I mean it's great you're dealing it with infra. | 10:21 |
| ThomasS | now with -vvv | 10:21 |
| ThomasS | any hint what i can do? | 10:21 |
| ThomasS | worked really good soar | 10:21 |
| evrardjp | I had a lot of trouble understanding all this all fits together when I was fixing master with Andreas. But I still haven't finished my reading. | 10:22 |
| odyssey4me | ThomasS hmm, it seems that something's trying to use the host name to figure out some config, and failing | 10:22 |
| evrardjp | ThomasS: looks like a configuration issue, breaking something in your cinder install | 10:23 |
| odyssey4me | odd that we aren't seeing that same issue in our tests | 10:23 |
| ThomasS | can i rerun starting with setup hosts? | 10:23 |
| odyssey4me | no, that won't help | 10:23 |
| ThomasS | could it be that my hosts file is messed up? i have a static one | 10:24 |
| evrardjp | odyssey4me: could you put spotz and me in the conversations/reviews for the docs changes? This way we know how to deal with other branches if that happens. | 10:24 |
| odyssey4me | ThomasS can you pastebin/gist the resulting /etc/cinder/cinder.conf from inside that container? | 10:24 |
| ThomasS | and seen that it was changed | 10:24 |
| odyssey4me | evrardjp sure | 10:24 |
| ThomasS | will do | 10:24 |
| evrardjp | ThomasS: odyssey4me probably worth knowing what are overrides too. | 10:24 |
| ThomasS | i also post my config | 10:24 |
| evrardjp | great | 10:24 |
| ThomasS | hold a sec | 10:24 |
| odyssey4me | ThomasS you have a manually set hosts file? What's in there? Because we add a bunch of stuff into /etc/hosts too. | 10:25 |
| * evrardjp hears the lift music while waiting | 10:25 | |
| odyssey4me | haha | 10:25 |
| evrardjp | :) | 10:25 |
| ThomasS | uh damn | 10:25 |
| ThomasS | maybe this messed it up | 10:25 |
| ThomasS | can i regenerate it` | 10:26 |
| ThomasS | ? | 10:26 |
| ThomasS | my user_varaibles | 10:26 |
| ThomasS | https://pastebin.com/mKbfWzi7 | 10:26 |
| odyssey4me | ThomasS ok, you definitely need to add https://github.com/openstack/openstack-ansible/commit/83e2b5415b50119eecc651fee4839ad25c5b810d to your user_variables file | 10:27 |
| odyssey4me | but that's unrelated I think | 10:27 |
| ThomasS | my openstack user variables | 10:27 |
| ThomasS | https://pastebin.com/dPr93nJN | 10:27 |
| ThomasS | i will add it | 10:27 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Remove uw_nginx jobs https://review.openstack.org/522040 | 10:27 |
| ThomasS | getting the cinder.conf | 10:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_keystone stable/pike: Remove uw_nginx jobs https://review.openstack.org/522514 | 10:28 |
| evrardjp | odyssey4me: I don't know if you've seen this: https://review.openstack.org/#/c/522427/ | 10:29 |
| evrardjp | just fyi | 10:29 |
| evrardjp | because you have a good memory, I plan to use that information later. | 10:29 |
| evrardjp | :D | 10:29 |
| odyssey4me | ah yes, that's a good call | 10:30 |
| odyssey4me | thanks jmccrory | 10:30 |
| ThomasS | my cinder.conf | 10:30 |
| ThomasS | https://pastebin.com/PbdyaU7s | 10:30 |
| *** armaan has quit IRC | 10:31 | |
| *** armaan has joined #openstack-ansible | 10:31 | |
| evrardjp | yeah :) | 10:31 |
| ThomasS | how can i regenerate the hosts files? | 10:31 |
| ThomasS | they have been replaced by my ansible run :-( | 10:32 |
| ThomasS | i was just copying the file :-( | 10:32 |
| ThomasS | my fault | 10:32 |
| evrardjp | which hosts file? | 10:33 |
| odyssey4me | ThomasS not sure what you mean? | 10:33 |
| evrardjp | /etc/hosts from a container? | 10:33 |
| evrardjp | yeah I didn't understand that too | 10:33 |
| odyssey4me | but it looks like this wasn't an explicit cinder.conf setting, but rather a default that uses the host name to figure out something like the ip address... and for some reason that broke | 10:33 |
| ThomasS | no on the hardware nodes | 10:33 |
| ThomasS | i have a job to copy one hosts file to all base os nodes | 10:34 |
| odyssey4me | if you're mucking with the hosts file and changing things that relate to the container names or the host names, then that might explain it | 10:34 |
| ThomasS | ok so can i fix ist? | 10:34 |
| ThomasS | it | 10:34 |
| odyssey4me | ok, so you'll need to replace that with an ansible lineinfile task rather so that it adds to the hosts file instead of replacing it | 10:34 |
| ThomasS | will do this | 10:35 |
| ThomasS | so re-run all playbooks starting with setup-hosts? | 10:35 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_zaqar stable/ocata: [WIP] Initial Zuul v3 role jobs https://review.openstack.org/514299 | 10:36 |
| odyssey4me | ThomasS well, first you need to stop the hosts file copy thing and make it work *with* our hosts file changes | 10:36 |
| ThomasS | already done | 10:37 |
| odyssey4me | so, as suggested, rather than just copying the file from one host to others - implement an ansible play that uses lineinfil to get those right | 10:37 |
| ThomasS | will do this | 10:37 |
| odyssey4me | and then, yes, re-run setup-hosts | 10:37 |
| odyssey4me | is this ia new/test environment? | 10:37 |
| ThomasS | a new enviroment for a poc | 10:38 |
| odyssey4me | ok, it might make sense to destroy all the containers before re-executing | 10:39 |
| odyssey4me | openstack-ansible lxc-containers-destroy.yml | 10:39 |
| ThomasS | ok | 10:39 |
| odyssey4me | if it's a new test environment, none of the data is important so the containers can be wiped | 10:39 |
| ThomasS | yes the data can go :-) | 10:40 |
| ThomasS | they are gone. let's see | 10:44 |
| ThomasS | thanks for your help! | 10:44 |
| ThomasS | let's get fingers crossed | 10:46 |
| gokhan | evrardjp there is problem on ceilometer gates. It can not find policy.json and api.paste.ini files. I looked at it. Indeed there no policy.json and api-paste.ini file. | 10:46 |
| gokhan | http://logs.openstack.org/57/521857/1/check/openstack-ansible-functional-ubuntu-xenial/f181b69/job-output.txt.gz#_2017-11-23_08_30_22_695303 | 10:47 |
| gokhan | https://git.openstack.org/cgit/openstack/ceilometer/plain/etc/ceilometer/ | 10:47 |
| odyssey4me | gokhan ah, it's possible that they are no longer relevant for the service | 10:47 |
| odyssey4me | I think they may have removed the API service entirely, so the role will have to be updated to remove all the stuff related to the API that's been removed | 10:48 |
| evrardjp | I think I removed that. gokhan | 10:51 |
| evrardjp | which branch?? | 10:51 |
| evrardjp | could you deal with that? | 10:51 |
| gokhan | evrardjp master branch. odyssey4me ceilometer api and client retired and on queens branch also panko role needed | 10:53 |
| *** jamesdenton has quit IRC | 10:53 | |
| SmearedBeard | when I test with Vagrant and make a change and run `vagrant provision` it seems to use the older version of the file.. is that expected? | 10:56 |
| SmearedBeard | so I should run `vagrant destroy && vagrant up centos7` when testing a new change? | 10:56 |
| hwoarang | SmearedBeard: centos? | 10:57 |
| SmearedBeard | you mean my OS? | 10:58 |
| hwoarang | the VM | 10:58 |
| hwoarang | ah you said centos7 didn't see it | 10:59 |
| hwoarang | so centos7 will only sync the working directory once during provision | 10:59 |
| hwoarang | you either need to rsync it again yourself for your new changes to appear to destroy and provision again | 10:59 |
| SmearedBeard | hwoarang ah, hence my confusion :D OK, cool so this is expected.. is this the same for ubuntu and opensus then? | 11:00 |
| hwoarang | no ubuntu and opensuse mount the directory using vboxfs | 11:00 |
| hwoarang | so your changes are always updated in vm | 11:00 |
| SmearedBeard | cool, that explains a lot then | 11:00 |
| openstackgerrit | Dimitrios Markou proposed openstack/openstack-ansible-os_neutron master: Add OvS-NSH support https://review.openstack.org/517259 | 11:01 |
| gokhan | evrardjp , on ocata and pike ,ceilometer,aodh and gnocchi roles are mature. I will modify some auth config deprecated variables on these branchs. Then I will struggle to upgrade ceilometer role on queens branch . | 11:01 |
| *** udesale__ has joined #openstack-ansible | 11:03 | |
| odyssey4me | gokhan thanks for pushing on with it | 11:04 |
| odyssey4me | if panko's needed, then are you up to the challenge of developing a new role for it? | 11:04 |
| *** stuartgr has joined #openstack-ansible | 11:05 | |
| openstackgerrit | Gökhan proposed openstack/openstack-ansible-os_aodh stable/pike: Change auth config to keystone v3 https://review.openstack.org/522480 | 11:05 |
| *** udesale has quit IRC | 11:05 | |
| *** armaan has quit IRC | 11:06 | |
| *** armaan has joined #openstack-ansible | 11:07 | |
| gokhan | odyssey4me, ok but I need time. I am junior and at learning stage :) Firstly I will try install new ceilometer, panko on devstack then start to write role | 11:11 |
| odyssey4me | gokhan thank you - we appreciate your time and effort! | 11:12 |
| andymccr | gokhan: that's awesome. | 11:14 |
| *** armaan has quit IRC | 11:14 | |
| gokhan | odyssey4me, thank you too :) | 11:15 |
| *** armaan has joined #openstack-ansible | 11:15 | |
| openstackgerrit | Guoqiang Ding proposed openstack/openstack-ansible-os_neutron master: Update the doc link https://review.openstack.org/522520 | 11:16 |
| gokhan | odyssey4me, andymccr, evrardjp if anyone need cloudkitty service on ocata branch, I integrated it on openstack-ansible. https://github.com/b3lab/openstack-ansible/commit/eb0005f5dad94879f10f5502353d5f7c9835a3c1 | 11:17 |
| gokhan | anyone use this commit :) | 11:17 |
| gokhan | it is working successfully | 11:18 |
| *** yifei has quit IRC | 11:19 | |
| *** yifei has joined #openstack-ansible | 11:20 | |
| odyssey4me | hmm, I see we forgot to create a pike branch for that repo | 11:23 |
| odyssey4me | do all those changes match up to what's in https://github.com/openstack/openstack-ansible-os_cloudkitty/tree/stable/ocata/extras ? | 11:24 |
| odyssey4me | if not, could you please push patches to those files to fix them? | 11:24 |
| *** armaan has quit IRC | 11:24 | |
| *** armaan has joined #openstack-ansible | 11:25 | |
| pradiprwt | evrardjp : That problem got solved by enabling promiscous mode to allow all vms, I have one more question how to disable aodh, ceilometer and gnochhi projects? | 11:26 |
| odyssey4me | pradiprwt as discussed yesterday, it is not enabled by default! | 11:41 |
| odyssey4me | if it is in your test setup, then either you've enabled it yourself or you're using a very, very old build | 11:42 |
| odyssey4me | pradiprwt do you have an entry for metering-alarm_host in any one of the files in /etc/openstack_deploy ? | 11:43 |
| odyssey4me | apologies, that should have been 'metering-alarm_hosts' (I missed the trailing s the first time) | 11:43 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Do not implement reload for novnc/spice service https://review.openstack.org/522526 | 11:44 |
| odyssey4me | also, do you have entries for 'metering-compute_hosts', 'metering-infra_hosts', 'metrics_hosts' | 11:44 |
| pradiprwt | odyssey4me: yes it is there in openstack_user_config.yml file | 11:46 |
| pradiprwt | do I need to remove it | 11:46 |
| odyssey4me | pradiprwt and that is why ou have those services enabled | 11:46 |
| pradiprwt | ohhh | 11:47 |
| odyssey4me | those entries tell ansible that you want to enable them | 11:47 |
| pradiprwt | Thanks now I will remove all entries from config | 11:47 |
| odyssey4me | ok, once that's done you'll need to destroy the containers, then remove them from your inventory - that is if you already have them setup | 11:47 |
| pradiprwt | yeah, thanks odyssey4me | 11:48 |
| odyssey4me | to remove the containers, execute this: openstack-ansible lxc-containers-destroy --limit aodh_all,ceilometer_all,gnocchi_all | 11:49 |
| odyssey4me | then use inventory-manage.py to remove the container entries from your inventory | 11:49 |
| odyssey4me | once that's done, they're gone - but the services were configured to use it... so you'll have to re-run setup-openstack.yml to reconfigure the services to no longer use it | 11:50 |
| odyssey4me | hwoarang evrardjp found a fairly significant bug and fixed it with https://review.openstack.org/522526 | 11:50 |
| *** gkadam_ has joined #openstack-ansible | 11:55 | |
| openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-lxc_hosts master: tasks: lxc_net: Do not source non-existing network script on CentOS https://review.openstack.org/522530 | 11:57 |
| hwoarang | cloudnull: mhayden^ | 11:57 |
| hwoarang | odyssey4me: let me check | 11:57 |
| *** gkadam has quit IRC | 11:57 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Do not implement reload for novnc/spice service https://review.openstack.org/522526 | 11:58 |
| odyssey4me | nice catch there hwoarang | 11:58 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_nova master: Do not implement reload for novnc/spice service https://review.openstack.org/522510 | 12:02 |
| hwoarang | odyssey4me: i am not sure how it managed to pass the gates... | 12:05 |
| hwoarang | maybe i am using an older centos and it's fatal there whereas on newer ones it is not | 12:05 |
| odyssey4me | no idea | 12:05 |
| mardim | evrardjp, hwoarang odyssey4me I am hitting an error in the latest build in CI which I think is transient and is timeout related could you take a look? | 12:24 |
| mardim | http://logs.openstack.org/59/517259/18/check/openstack-ansible-ovs-nsh-ubuntu-xenial/774da47/job-output.txt.gz | 12:24 |
| mardim | Also I was not hitting that error in the previous build | 12:25 |
| hwoarang | mardim: we know about it. a failure during tempest | 12:25 |
| hwoarang | ah no not tempest | 12:25 |
| hwoarang | a different one | 12:25 |
| hwoarang | i dont know what to say. another failure with linux-headers | 12:26 |
| mardim | Yes but the linux headers are getting installed also in server1 and server2 and localhost | 12:26 |
| mardim | but I am not seeing any error there | 12:26 |
| mardim | the error is only in agents1 agents2 | 12:26 |
| mardim | Also in the previous run the error was not there | 12:27 |
| mardim | that's why I think is transient | 12:27 |
| hwoarang | it's not easy for me to decipher apt error messages | 12:27 |
| hwoarang | so i dont know | 12:27 |
| mardim | hwoarang, ok thanks maybe the other guys have some ideas :) | 12:27 |
| mardim | odyssey4me, evrardjp ^ | 12:27 |
| mardim | andymccr, ^ | 12:28 |
| odyssey4me | mardim either it's transient, or it's not - if the error happens all the time, then it's not transient - if it happens sometimes, then it's usually best to try and find the pattern | 12:30 |
| *** jwitko_ has quit IRC | 12:30 | |
| odyssey4me | so first decide whether it's transient or not - does it always fail in the same place in the same way, does it work for you in your tests, but not in the gating tests? | 12:30 |
| mardim | odyssey4me, do you think is realted to this https://github.com/ansible/ansible-modules-core/issues/2951 ? | 12:30 |
| mardim | odyssey4me, it works for me wehn I run tests locally | 12:30 |
| mardim | odyssey4me, It works in a real deployment | 12:31 |
| mardim | odyssey4me, And some time passes in Ci some time has the error | 12:31 |
| odyssey4me | ok, then find the pattern for the failure from the logs - does it happen in one nodepool provider, more than one, or all of them? | 12:31 |
| mardim | what is a nodepool | 12:31 |
| mardim | ? | 12:31 |
| odyssey4me | all openstack tests are run in various cloud providers, each test node is provided in pools (one per provider/region) | 12:33 |
| odyssey4me | so when the test executes, it could be in rackspace cloud iad in the first test, then in ovh in europe for the next | 12:33 |
| mardim | odyssey4me, Ah ok thanks I understood | 12:33 |
| odyssey4me | sometimes one of the regions may have a bad cache/mirror, or something being done in the test that fetches something from the internet fails because of latency or some other issue | 12:34 |
| odyssey4me | also some of the VM's being tested on have nested virt, and some do not | 12:34 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_zaqar stable/ocata: Initial Zuul v3 role jobs https://review.openstack.org/514299 | 12:34 |
| odyssey4me | in your test I can see from http://logs.openstack.org/59/517259/18/check/openstack-ansible-ovs-nsh-ubuntu-xenial/774da47/zuul-info/inventory.yaml that the test ran in citycloud-kna1 (the provider, under nodepool) | 12:34 |
| odyssey4me | so check where you see other fails of the same type, and see if the fails are particular to a specific provider or more than one | 12:35 |
| mardim | odyssey4me, How did you see that ? | 12:35 |
| mardim | odyssey4me, I mean how did you find out which is the provider ? | 12:36 |
| odyssey4me | mardimlook at the file I linked and you'll see | 12:36 |
| odyssey4me | mardim ^ | 12:36 |
| mardim | odyssey4me, ah ok yes thanks | 12:36 |
| mardim | I will look into that | 12:36 |
| odyssey4me | you can also use logstash.openstack.org to search job logs and find patterns | 12:38 |
| odyssey4me | the logs get ingested when they can, so a job that just ran will not appear there until a few hours later usually | 12:39 |
| mardim | a ok I will check it | 12:40 |
| mardim | never heard of it though | 12:40 |
| hwoarang | i think it's important to understand what the error is trying to tell you. then you will find it easier to look at the right place | 12:40 |
| mardim | hwoarang, the problem is that I cannot replicate | 12:41 |
| hwoarang | it doesn't matter | 12:41 |
| hwoarang | your host is not the same as the ones running in nodepool | 12:41 |
| hwoarang | and the log has valid info in it | 12:41 |
| odyssey4me | it looks like this is the actual fail: dpkg: error processing archive /var/cache/apt/archives/linux-headers-4.4.0-101-generic_4.4.0-101.124_amd64.deb (--unpack):\r\n unable to install new version of '/lib/modules/4.4.0-101-generic/build': No such file or directory | 12:42 |
| mardim | hwoarang, odyssey4me Yes that is happening because the package is not installed | 12:42 |
| mardim | hwoarang, odyssey4me But I do not know why | 12:43 |
| openstackgerrit | Aaron Mehar proposed openstack/openstack-ansible-os_trove master: Adds CentOS 7 support https://review.openstack.org/522541 | 12:54 |
| *** dmsimard|afk is now known as dmsimard | 12:55 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Create variable override tests https://review.openstack.org/522542 | 13:03 |
| hwoarang | mardim: why do you need the linux-headers anyway? | 13:06 |
| hwoarang | can you not simply pull the dkms package? | 13:06 |
| *** jvidal has quit IRC | 13:06 | |
| mardim | hwoarang, the dkms need the headers to compile the ovs-nsh kernel modules | 13:10 |
| hwoarang | so it will pull them itself | 13:10 |
| hwoarang | why are you being explicit? | 13:10 |
| mardim | what do you mean it will pull them itself | 13:11 |
| mardim | you mean like | 13:11 |
| hwoarang | i dont see any other dkms ubuntu package to pull linux-headers | 13:11 |
| hwoarang | the dkms packages are already compiled. so you shouldn't need the headers. you are not building the dkms on the role | 13:11 |
| mardim | no but I am compiling the kernel modules for ovs-nsh | 13:12 |
| hwoarang | where | 13:12 |
| mardim | to be more explicit | 13:12 |
| mardim | when I install | 13:12 |
| mardim | the | 13:12 |
| mardim | openvswitch-nsh-dkms.deb | 13:12 |
| hwoarang | something is not right | 13:13 |
| mardim | the dkms will run the ./configure command | 13:13 |
| mardim | and to do so needs the headers | 13:13 |
| hwoarang | i think your dkms package is not right | 13:13 |
| mardim | it is because it's working if it finds the headers | 13:13 |
| mardim | we already tested that | 13:14 |
| hwoarang | i am just looking at https://packages.ubuntu.com/xenial/virtualbox-guest-dkms | 13:14 |
| hwoarang | and it does not need linux-headers | 13:14 |
| mardim | hwoarang, Yes maybe it assumes that the heders are already installed | 13:15 |
| hwoarang | the fact that your package works doesn't mean it's packaged correctly | 13:15 |
| mardim | headers | 13:15 |
| hwoarang | no there are no such assumptions | 13:15 |
| hwoarang | packages are explicit on what dependencies they need | 13:15 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Create variable override tests https://review.openstack.org/522542 | 13:16 |
| mardim | i am not expert in packaging I just followed the procedure we did so far in other installers | 13:16 |
| mardim | like fuel | 13:16 |
| hwoarang | i suggest you compare your dkms package with an existing one on ubuntu | 13:16 |
| mardim | hwoarang, ok sure | 13:17 |
| *** jvidal has joined #openstack-ansible | 13:18 | |
| evrardjp | odyssey4me: question on https://review.openstack.org/#/c/522526/2 | 13:25 |
| *** chyka has joined #openstack-ansible | 13:25 | |
| evrardjp | hwoarang: IIRC virtualbox-guest-dkms needs linux-headers, it just doesn't ship with it. | 13:26 |
| hwoarang | evrardjp: it's not listed as dependency | 13:27 |
| evrardjp | yes | 13:27 |
| hwoarang | no ubuntu dkms package pulls linux-headers | 13:27 |
| evrardjp | ok | 13:27 |
| hwoarang | if they need them they should pull them | 13:27 |
| hwoarang | something is fishy... | 13:27 |
| mardim | hwoarang, the same here | 13:27 |
| evrardjp | well I suspected this to be a licensing issue | 13:27 |
| mardim | hwoarang, https://bbs.archlinux.org/viewtopic.php?id=210378 | 13:27 |
| hwoarang | evrardjp: hmm why? Adding a package in the 'Dependency' list should have anything to do with licenses | 13:28 |
| evrardjp | mardim: exactly, on my hosts I already had issues in the past that required me to install linux-headers for installing virtualbox | 13:28 |
| evrardjp | but I don't know anything about all that. | 13:29 |
| hwoarang | *shouldn't | 13:29 |
| evrardjp | hwoarang: I don't know. | 13:29 |
| mardim | evrardjp, same here I just tried to install the packages with dkms | 13:29 |
| evrardjp | consider I haven't said anything. | 13:29 |
| mardim | and error occured you need linux headers | 13:29 |
| *** chyka has quit IRC | 13:29 | |
| hwoarang | i have no clue what dkms requires but i would be surprised if installing a dkms package requires the headers to be present | 13:29 |
| evrardjp | I will trust hwoarang in those matters. | 13:29 |
| hwoarang | maybe ubuntu doesn't seem modules, but recipies to build modules | 13:30 |
| hwoarang | *ship | 13:30 |
| hwoarang | whereas centos/suse give you the .ko files directly | 13:30 |
| mardim | hwoarang, I think the idea with the dkms is that | 13:30 |
| evrardjp | oh yeah I don't think those have .ko. I think it's just the sources? | 13:31 |
| mardim | you do not need to have the kernel modules precompiled for any version of the kernel | 13:31 |
| evrardjp | sorry I shouldn't talk when i don't know. | 13:31 |
| mardim | but you can compile them on the fly if you have the linux heraders of the current kernel | 13:31 |
| hwoarang | ok then i dont know what to say | 13:34 |
| mardim | me neither I just follow instructions here :) | 13:35 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: tasks: lxc_net: Do not source non-existing network script on CentOS https://review.openstack.org/522530 | 13:36 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: [DNM] Making sure this test vars will fail when not properly done https://review.openstack.org/522552 | 13:36 |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_tempest stable/pike: Disable glance v1 API by default https://review.openstack.org/522555 | 13:46 |
| evrardjp | wow I didn't know glance image api v1 isn't supported for THAT long ago. | 13:46 |
| odyssey4me | uh, evrardjp why're you disabling it for pike? our defaults still have it enabled in pike | 13:47 |
| openstackgerrit | Flavio Percoco proposed openstack/openstack-ansible-plugins master: Preserve comments in INI files https://review.openstack.org/522557 | 13:47 |
| gokhan | folks, I am testing magnum installation, but I see a problem on magnum | 13:47 |
| evrardjp | yeah I wanted to open the discussion of things to do/not do | 13:47 |
| gokhan | it is installing developer version | 13:48 |
| evrardjp | odyssey4me: I have -1ed it already, but this way I remember I need to think about all these deprecations. | 13:48 |
| gokhan | like magum-4.1.4.dev | 13:48 |
| evrardjp | flaper87: will review this one :) | 13:48 |
| evrardjp | I* | 13:48 |
| flaper87 | evrardjp: thanks :D | 13:49 |
| flaper87 | evrardjp: it's a bit of a hack but it seems to work beautifully | 13:49 |
| flaper87 | (last famous words) | 13:49 |
| evrardjp | lol | 13:50 |
| evrardjp | looks ok to me | 13:50 |
| evrardjp | will wait for the pep8 decision | 13:50 |
| odyssey4me | gokhan yes, we install from source at arbitrary SHA's, so it almost always ends up being a dev version | 13:50 |
| gokhan | odyssey4me, you say it is stable version. I looked at heat and sahara but on them I didn't see .dev installation, so I doubt of this situation on magnum | 13:52 |
| odyssey4me | gokhan that depends on when the service last committed a change, and when they last released | 13:53 |
| odyssey4me | evidently magnum has commits beyond their last release | 13:53 |
| odyssey4me | we do not use their released tags, we test at a sha regardless of whether it's tagged or not | 13:53 |
| evrardjp | odyssey4me: I am thinking more and more to use only tags tbh. | 13:55 |
| gokhan | odyssey4me, ok thanks. I think we should use only tags maybe. | 13:56 |
| odyssey4me | evrardjp oh sure, I have some work in the pipeline to make the use of SHA's optional | 13:58 |
| evrardjp | my sha bump tool takes tags instead of sha | 13:58 |
| evrardjp | so I don't need to do anything for it | 13:58 |
| evrardjp | can take* | 13:58 |
| odyssey4me | oh, that can be done today that way, yes - but that will only work for non-master | 13:58 |
| evrardjp | oh yeah I am talking about non master | 13:59 |
| evrardjp | wow I didn't realize we weren't talking about the same thing | 13:59 |
| *** stuartgr has left #openstack-ansible | 13:59 | |
| *** jvidal has quit IRC | 14:05 | |
| *** nshetty has quit IRC | 14:09 | |
| spotz | y'all are very chatty I'm gonna need to increase my scroll!:) | 14:13 |
| spotz | Ah cool jmccrory's glance patch went through finally. Fell asleep waiting for it to run:) | 14:16 |
| mardim | and the service was up | 14:17 |
| mardim | sorry ^ | 14:18 |
| *** jvidal has joined #openstack-ansible | 14:18 | |
| *** armaan_ has joined #openstack-ansible | 14:19 | |
| *** armaan has quit IRC | 14:19 | |
| evrardjp | you mean the tempest one? | 14:24 |
| evrardjp | yeah | 14:24 |
| evrardjp | that's cool | 14:24 |
| evrardjp | he did a great job :) | 14:24 |
| evrardjp | I didn't realize but it's holiday for you guys, right? | 14:25 |
| *** dave-mccowan has joined #openstack-ansible | 14:26 | |
| *** lbragstad has joined #openstack-ansible | 14:29 | |
| *** armaan_ has quit IRC | 14:32 | |
| spotz | evrardjp: Yep and yep | 14:35 |
| evrardjp | enjoy then! | 14:36 |
| bndzor | morning | 14:36 |
| spotz | evrardjp: I had a feeling it was config related but was editing videos before bed. I need to work all weekend to catch up from summit so I'll be about and my scroll is now set to 3000 vs 300 so no more missing stuff! | 14:36 |
| evrardjp | :) | 14:37 |
| evrardjp | I have 10k I am still missing stuff. So I learned to stop caring and stopped reading. If someone really wants, (s)he can ping me when I am back:D | 14:37 |
| spotz | evrardjp: So I know you told odyssey4me to add me to something but not what! | 14:37 |
| evrardjp | oh | 14:38 |
| evrardjp | docs thing | 14:38 |
| evrardjp | ensuring the docs are properly published in all branches | 14:38 |
| spotz | I used to look at the logs but no clue where mudpuppy put the bouncer when he moved it. Luckily he used my original setup so it still works:) | 14:38 |
| spotz | evrardjp: Ok and we're still waiting on the new headers from Jimmy I think but we should be all good there or a quick fix. I've also got a patch ready to go for when newton goes eol to fix our releases page | 14:39 |
| spotz | After Thursday I should be a little freeer so I'll do review sweeps and try to get some stuff moving again | 14:40 |
| ThomasS | hi | 14:41 |
| ThomasS | i am at the reinstall now and setup-infrastructure is failing because on infra1 the containers have no it | 14:42 |
| ThomasS | it looks like the iptables rules are missing on this host | 14:42 |
| ThomasS | how can i setup these? are they not supposed to be setup by setup-hosts playbook? | 14:43 |
| ThomasS | the host files btw look now fine :-) | 14:43 |
| ThomasS | i did iptables-save and restore to copy the rules manually | 14:46 |
| spotz | ThomasS: no the playbppks should work there, if you have an error can you pastebin it so someone can look when they have a chance? | 14:46 |
| ThomasS | sure | 14:47 |
| ThomasS | spotz: all containers on infra1 have no ip address assigned | 14:48 |
| bndzor | check your network config | 14:48 |
| ThomasS | https://pastebin.com/J1Sp7bT3 | 14:48 |
| *** woodard has joined #openstack-ansible | 14:48 | |
| ThomasS | what exactly? | 14:48 |
| bndzor | depends on your network config | 14:48 |
| ThomasS | the other nodes are fine like infra2 and 3 | 14:49 |
| ThomasS | all got ips | 14:49 |
| bndzor | that sounds wierd | 14:49 |
| bndzor | try maybe rebooting the machine and see if that helps | 14:49 |
| ThomasS | it is maybe :-) | 14:49 |
| gun1x | ThomasS: | 14:50 |
| gun1x | did you run setup-hosts? | 14:50 |
| ThomasS | yes | 14:50 |
| ThomasS | completed without errors | 14:50 |
| gun1x | ThomasS: first question, are you on pike? 16.0.2? | 14:50 |
| ThomasS | 16.0.3 | 14:50 |
| ThomasS | yes | 14:50 |
| gun1x | ok | 14:50 |
| ThomasS | as full disclosure i did a destroy of all containers | 14:51 |
| gun1x | OSA (and any other automation solution) should run by its self, without manual intervention. | 14:51 |
| ThomasS | because i messed the hosts file | 14:51 |
| gun1x | if you destroyed containers, they should be automatically created with setup hosts | 14:51 |
| *** dave-mccowan has quit IRC | 14:51 | |
| ThomasS | they are | 14:51 |
| gun1x | i would suggest posting the playbook on bpaste. | 14:51 |
| bndzor | sounds more like some network config or daemon is b0rk3d | 14:51 |
| ThomasS | https://pastebin.com/J1Sp7bT3 | 14:52 |
| spotz | ThomasS: I'm assuming th first time infra1 did get an IP? In shich case when you cleaned up you might have forgotten to remove something | 14:52 |
| ThomasS | yes | 14:53 |
| ThomasS | i guess so | 14:53 |
| ThomasS | i tried to fix the error that the repo container was gone | 14:53 |
| gun1x | so stuff i would check: network config on all hosts (making sure ping and TCP is allowed on all bridges) and playbook (so that there are no mistakes there). also required packages (as suggested by the docu). if all these are ok, playbooks should run wihtout error | 14:53 |
| spotz | ThomasS: You mentioned host files, start there but then check networking | 14:53 |
| ThomasS | and to use the public repo again | 14:53 |
| ThomasS | the playbooks run without error :-) | 14:54 |
| ThomasS | maybe i do a destroy again and retry :-( | 14:54 |
| bndzor | i did a install yesterday using 3 controllers (or infras, whatevah mhm), 3 storages(ceph) and 7 computenodes on stable/pike, worked gucci | 14:54 |
| gun1x | ThomasS: starting from zero usually helps | 14:55 |
| ThomasS | ok | 14:55 |
| bndzor | ^ using ubuntu 16.04 etc | 14:56 |
| *** jwitko_ has joined #openstack-ansible | 14:57 | |
| ThomasS | my too 16.04 | 14:58 |
| ThomasS | i guess i broke something on the infra1 | 14:58 |
| *** jwitko_ has quit IRC | 14:59 | |
| spotz | ThomasS: My guess is something is hanging around from the first installation | 15:00 |
| *** udesale__ has quit IRC | 15:02 | |
| odyssey4me | ThomasS there's a script which puts all the iptables bits in place | 15:07 |
| odyssey4me | lxc-system-manage | 15:07 |
| odyssey4me | that should fix things all up for you, but it's a bit wierd that you got into a state... I wonder how reproducable that is so that we can figure out a solution | 15:08 |
| *** chyka has joined #openstack-ansible | 15:14 | |
| *** chyka has quit IRC | 15:19 | |
| openstackgerrit | Dimitrios Markou proposed openstack/openstack-ansible-os_neutron master: Add OvS-NSH support https://review.openstack.org/517259 | 15:29 |
| *** szaher has quit IRC | 15:34 | |
| *** szaher has joined #openstack-ansible | 15:35 | |
| openstackgerrit | Dimitrios Markou proposed openstack/openstack-ansible-os_neutron master: Add OvS-NSH support https://review.openstack.org/517259 | 15:35 |
| *** gkadam_ has quit IRC | 15:38 | |
| *** gkadam has joined #openstack-ansible | 15:40 | |
| *** lbragstad has quit IRC | 15:53 | |
| *** gkadam is now known as gkadam-dinner | 15:53 | |
| *** pcaruana has quit IRC | 16:09 | |
| *** chyka has joined #openstack-ansible | 16:13 | |
| *** mrch has quit IRC | 16:14 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: ansible-role-requirements: Switch ODL git repo to the github mirror https://review.openstack.org/521792 | 16:17 |
| *** chyka has quit IRC | 16:17 | |
| mgariepy | is it just me or github is super slow today ? | 16:22 |
| mgariepy | cloning at : 20.00 KiB/s \ | 16:22 |
| openstackgerrit | Periyasamy Palanisamy proposed openstack/openstack-ansible-os_neutron master: add bgpvpn ml2 config support https://review.openstack.org/522598 | 16:25 |
| openstackgerrit | Periyasamy Palanisamy proposed openstack/openstack-ansible-os_neutron master: add bgpvpn ml2 config support https://review.openstack.org/522598 | 16:26 |
| openstackgerrit | Aaron Mehar proposed openstack/openstack-ansible-os_trove master: Adds CentOS 7 support https://review.openstack.org/522541 | 16:27 |
| hwoarang | andymccr cloudnull logan-: do you know how often the ceph ansible roles sync from https://github.com/ceph/ceph-ansible ? last sync is 23 days ago so I can't properly test openSUSE in osa :/ | 16:29 |
| hwoarang | i have seen your name on some commits (mhayden too) so i wanted to check with you first :) | 16:30 |
| *** dave-mccowan has joined #openstack-ansible | 16:30 | |
| andymccr | i think logan- had an issue with that | 16:30 |
| andymccr | hwoarang: you mean like the ceph/ansible-ceph-osd roles | 16:31 |
| hwoarang | yeah | 16:31 |
| andymccr | yeah i kinda feel like we should use the ceph-ansible repo itself rather than the individual roles | 16:31 |
| hwoarang | from what i understood these are autogenerated from the main repo | 16:31 |
| andymccr | yeah | 16:31 |
| hwoarang | or something like that | 16:31 |
| andymccr | they are supposed to be | 16:31 |
| andymccr | i think logan- had an objection to using the overarching ceph-ansible repo itself | 16:32 |
| hwoarang | well something is broken for a month and nobody noticed so maybe we are not supposed to use them like you said ;p | 16:32 |
| hwoarang | i will try to talk to upstream again | 16:32 |
| andymccr | the last bump was a result of logan- pinging in the ceph-ansible irc channel i think :) | 16:32 |
| odyssey4me | they appear to be a bit of a second class citizen :/ | 16:32 |
| odyssey4me | ja, I suspect they don't have it automated - but instead it gets done by hand | 16:33 |
| andymccr | honestly, if logan- isn't opposed to it, the ceph-ansible overall repo has worked well for me and it removes that problem | 16:33 |
| andymccr | it also prevents a situation where you move from versions and the roles split out differently | 16:33 |
| andymccr | since it'll just get all of the roles | 16:33 |
| odyssey4me | but it will mean that it's not managed from a-r-r any more right? | 16:33 |
| odyssey4me | we'd have to track the version in something else | 16:34 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: ansible-role-requirements: Switch ODL git repo to the github mirror https://review.openstack.org/521796 | 16:34 |
| andymccr | odyssey4me: i use a-r-r the only diff is that you have to set the role path to include /etc/ansible/roles/ceph-ansible/roles/ | 16:34 |
| hwoarang | well the main repo is tagged to so we can still track releases | 16:35 |
| andymccr | yeah | 16:35 |
| hwoarang | ok i will wait for logan- but that's still an issue because ceph master is not really master right now | 16:36 |
| hwoarang | lets see | 16:36 |
| andymccr | hwoarang: so does suse work now on ceph-ansible master? | 16:41 |
| mardim | Guys FYI the centos gate is failing | 16:42 |
| hwoarang | andymccr: in theory they can't run CI on openSUSE so I will have to run OSA to see what happens | 16:43 |
| odyssey4me | andymccr hmm, that works | 16:44 |
| *** dave-mccowan has quit IRC | 16:45 | |
| andymccr | hwoarang: i'll try play about with it and see if i can test it for you on that stand alone stuff i was doing | 16:46 |
| hwoarang | andymccr: i couldn't use that a lot of things didn't place out of the box for me. you are using 2.X iirc but i wanted master and too many things changed | 16:47 |
| hwoarang | but the suse code is very similar to redhat so i don't think i am far off | 16:47 |
| hwoarang | *s/place/play/ | 16:48 |
| *** chyka has joined #openstack-ansible | 16:55 | |
| *** jvidal has quit IRC | 16:57 | |
| *** michelv has quit IRC | 17:01 | |
| *** stuartgr has joined #openstack-ansible | 17:12 | |
| *** holmsten has quit IRC | 17:17 | |
| *** lbragstad has joined #openstack-ansible | 17:19 | |
| *** gkadam-dinner is now known as gkadam | 17:22 | |
| *** gkadam has quit IRC | 17:27 | |
| *** stuartgr has left #openstack-ansible | 17:29 | |
| openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 17:31 |
| *** lbragstad has quit IRC | 17:36 | |
| *** michelv has joined #openstack-ansible | 17:50 | |
| *** baloo has joined #openstack-ansible | 17:51 | |
| *** baloo is now known as Guest31559 | 17:51 | |
| *** michelv has quit IRC | 17:54 | |
| *** mbuil has quit IRC | 17:56 | |
| *** shardy has quit IRC | 18:09 | |
| *** savvas_ has joined #openstack-ansible | 18:25 | |
| *** savvas_ has quit IRC | 18:28 | |
| *** germs has joined #openstack-ansible | 18:39 | |
| *** germs1 has joined #openstack-ansible | 18:40 | |
| *** viktor_ has joined #openstack-ansible | 18:43 | |
| *** germs has quit IRC | 18:43 | |
| *** m|y|k has quit IRC | 18:44 | |
| *** viktor_ has quit IRC | 18:52 | |
| *** japestinho has quit IRC | 19:11 | |
| *** savvas_ has joined #openstack-ansible | 19:24 | |
| *** savvas_ has quit IRC | 19:26 | |
| *** savvas_ has joined #openstack-ansible | 19:26 | |
| *** electrofelix has quit IRC | 19:30 | |
| *** pradiprwt has quit IRC | 19:49 | |
| *** ThomasS_ has joined #openstack-ansible | 20:01 | |
| ThomasS_ | hi guys | 20:01 |
| ThomasS_ | so the setup-infrastructure went ok expect two errors with the rabbitmq container | 20:02 |
| ThomasS_ | the fiest task which fails is TASK [rabbitmq_server : Read rabbit cookie] | 20:02 |
| ThomasS_ | the second one which brakes it is TASK [rabbitmq_server : Apply rabbitmq policies] | 20:03 |
| ThomasS_ | the error looks like this | 20:03 |
| ThomasS_ | https://pastebin.com/ZxV8dmEy | 20:04 |
| ThomasS_ | the message says hostname problem but the entries are there | 20:04 |
| ThomasS_ | in /etc/hosts | 20:04 |
| ThomasS_ | this is only happening on infra1 | 20:09 |
| *** savvas__ has joined #openstack-ansible | 20:10 | |
| *** savvas__ has quit IRC | 20:11 | |
| *** savvas_ has quit IRC | 20:13 | |
| *** savvas_ has joined #openstack-ansible | 20:14 | |
| *** savvas__ has joined #openstack-ansible | 20:15 | |
| *** savvas__ has quit IRC | 20:17 | |
| *** savvas_ has quit IRC | 20:17 | |
| *** savvas_ has joined #openstack-ansible | 20:17 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Set ceph_mons in group_vars https://review.openstack.org/520588 | 20:22 |
| *** savvas__ has joined #openstack-ansible | 20:23 | |
| *** savvas__ has quit IRC | 20:23 | |
| *** savvas__ has joined #openstack-ansible | 20:24 | |
| *** savvas_ has quit IRC | 20:25 | |
| *** savvas_ has joined #openstack-ansible | 20:26 | |
| *** savvas_ has quit IRC | 20:26 | |
| *** savvas__ has quit IRC | 20:27 | |
| *** savvas_ has joined #openstack-ansible | 20:27 | |
| *** savvas__ has joined #openstack-ansible | 20:29 | |
| *** savvas_ has quit IRC | 20:31 | |
| *** savvas_ has joined #openstack-ansible | 20:32 | |
| *** savvas__ has quit IRC | 20:34 | |
| *** savvas__ has joined #openstack-ansible | 20:35 | |
| *** savvas_ has quit IRC | 20:37 | |
| *** savvas_ has joined #openstack-ansible | 20:38 | |
| *** savvas__ has quit IRC | 20:40 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-lxc_hosts stable/pike: tasks: lxc_net: Do not source non-existing network script on CentOS https://review.openstack.org/522660 | 21:01 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Allow the haproxy configuration to run with limited groups https://review.openstack.org/521331 | 21:03 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible master: Remove the AIO scenario & add new scenarios to maintain coverage https://review.openstack.org/516002 | 21:03 |
| *** savvas__ has joined #openstack-ansible | 21:27 | |
| *** savvas_ has quit IRC | 21:29 | |
| *** savvas__ has quit IRC | 21:34 | |
| *** savvas_ has joined #openstack-ansible | 21:34 | |
| ThomasS_ | i rebuild the rabbit mq container on infra1 | 21:41 |
| ThomasS_ | problem is that the error says that it could not resolve the hostname | 21:42 |
| ThomasS_ | i tried from host and container and the name in the error message resolves perfect | 21:42 |
| ThomasS_ | here is the error https://pastebin.com/FjwJgCev | 21:43 |
| ThomasS_ | i am a bit lost :-( | 21:43 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh stable/pike: Change auth config to keystone v3 https://review.openstack.org/522480 | 21:46 |
| ThomasS_ | i think i got it. the infra1 node had a wrong entry in /etc/hosts | 21:50 |
| ThomasS_ | trying to fix it nox | 21:50 |
| ThomasS_ | now | 21:50 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone stable/pike: Remove uw_nginx jobs https://review.openstack.org/522514 | 21:56 |
| *** newmember has joined #openstack-ansible | 22:01 | |
| ThomasS_ | did not fix it... | 22:01 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon stable/pike: Install sahara-dashboard local_settings https://review.openstack.org/516748 | 22:01 |
| *** mbuil has joined #openstack-ansible | 22:11 | |
| openstackgerrit | Manuel Buil proposed openstack/openstack-ansible master: Add openvswitch_nsh group to inventory https://review.openstack.org/517269 | 22:23 |
| *** hamza has quit IRC | 22:27 | |
| *** Guest31559 has quit IRC | 22:44 | |
| *** mbuil has quit IRC | 22:47 | |
| -openstackstatus- NOTICE: Zuul has been restarted due to an unexpected issue. We're able to re-enqueue changes from check and gate pipelines, please check http://zuulv3.openstack.org/ for more information. | 22:54 | |
| *** ThomasS_ has quit IRC | 23:06 | |
| *** savvas_ has quit IRC | 23:07 | |
| *** savvas_ has joined #openstack-ansible | 23:07 | |
| *** savvas_ has joined #openstack-ansible | 23:41 | |
| *** savvas_ has quit IRC | 23:41 | |
| *** savvas_ has joined #openstack-ansible | 23:41 | |
| *** lbragstad has joined #openstack-ansible | 23:56 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!