| *** irclogbot_2 has joined #openstack-ansible | 00:22 | |
| *** irclogbot_2 has quit IRC | 00:39 | |
| *** irclogbot_2 has joined #openstack-ansible | 01:26 | |
| *** irclogbot_2 has quit IRC | 01:48 | |
| openstackgerrit | James Denton proposed openstack/openstack-ansible-os_neutron master: Implement uWSGI for neutron-api https://review.opendev.org/486156 | 02:32 |
|---|---|---|
| *** irclogbot_2 has joined #openstack-ansible | 02:46 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Pass nova_service_region to service_setup tasks https://review.opendev.org/697834 | 03:49 |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift master: Pass _service_region to service_setup tasks https://review.opendev.org/697867 | 04:19 |
| *** udesale has joined #openstack-ansible | 04:25 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Pass _service_region to service_setup tasks https://review.opendev.org/697866 | 04:25 |
| openstackgerrit | Merged openstack/openstack-ansible master: Set neutron_service_region in group_vars https://review.opendev.org/697846 | 04:48 |
| *** gokhani has joined #openstack-ansible | 05:01 | |
| *** cshen has joined #openstack-ansible | 05:28 | |
| *** cshen has quit IRC | 05:33 | |
| *** raukadah is now known as chkumar|ruck | 06:04 | |
| *** elenalindq has joined #openstack-ansible | 06:05 | |
| *** cshen has joined #openstack-ansible | 07:38 | |
| snadge | jrosser: i got the router pingable from everywhere (all the vms via the br-vlan interface) .. but not within the utility container that tries to ping it | 08:08 |
| snadge | within the container there is an eth0 which is attached to lxc-br0 (or whatever its called im remote now and can't remember) it is the default route though. and br-mgmt is accessible via eth2 | 08:09 |
| snadge | so im guessing whats supposed to happen is that the lxc container bridge.. is supposed to allow that traffic to pass to br-vlan.. but im not sure | 08:10 |
| *** sshnaidm|off is now known as sshnaidm | 08:11 | |
| jrosser | lxcbr0 is created by lxc and it should NAT via the host ip stack | 08:11 |
| ChosSimbaOne | morning. Does openStack-ansible store any confing data outside /etc/openstack_deploy? I want to deploy a additional cluster from the same deployment host, and I have /etc/openstack_deploy as an git repository, so would just do a branch and do the customization within that branch? That should work riight? | 08:12 |
| ChosSimbaOne | ansible fact etc. is not stored elsewhere, right? | 08:12 |
| jrosser | ChosSimbaOne: I think an env var was added that allows you to point to alternate directories for the config data | 08:12 |
| jrosser | snadge: can you ping the router ip from the host of the utility container? | 08:13 |
| snadge | ahh okay.. i'll look into lxcbr0 tomorrow, i bet its attached to the wrong interface or something silly like that | 08:13 |
| snadge | yes definitely.. and the other vms | 08:13 |
| snadge | so it looks like br-vlan is working properly | 08:13 |
| snadge | but inside the container, the lxc bridge isn't letting us ping ips on that interface | 08:14 |
| jrosser | is this centos? | 08:14 |
| snadge | yes | 08:14 |
| jrosser | hmm | 08:14 |
| snadge | an obvious option is to expose br-vlan directly to the container via another interface.. eg.. eth3 or whatever | 08:14 |
| jrosser | we had a very odd bug a while back where forwarding gets disabled on bridges | 08:15 |
| snadge | but it doesn't look like the AIO works that way, it appears to use the lxc bridge | 08:15 |
| ChosSimbaOne | jrosser: thank you, will look into that :-) | 08:16 |
| snadge | i was just going to give up and turn tempest install off, but im so close now that I want to know why this doesn't work.. and i suspect being able to access that network via the containers is a useful thing anyway | 08:16 |
| jrosser | fwiw I put an extra bridge and interface on my utility container for this | 08:16 |
| snadge | yeah that makes sense.. honestly, I know very little about lxc, and i just copy pasted the config for the lxc bridge from the AIO example | 08:17 |
| jrosser | but I do that because I know it won’t work otherwise because of the way my networks are segregated | 08:17 |
| snadge | so i should probably take a closer look at it | 08:17 |
| jrosser | for me even NAT via the host won’t work | 08:18 |
| snadge | the way i have it set up.. it *should* work | 08:18 |
| snadge | as long as i can get connectivity between br-vlan and lxcbr0 between the host and the containers on it | 08:19 |
| jrosser | I would be checking if any kind of forwarding is working on that bridge | 08:19 |
| *** miloa has joined #openstack-ansible | 08:19 | |
| jrosser | tcpdump inside and outside the container at the same time | 08:19 |
| jrosser | like I say centos has exhibited this before | 08:20 |
| jrosser | snadge: also simple check like looking at the routing table in the utility container to make sure that the route to your external network looks like it should b me via eth0 | 08:23 |
| snadge | yeah i've checked that it routes via eth0 which is the default route | 08:24 |
| *** tosky has joined #openstack-ansible | 08:25 | |
| jrosser | I’ll take a look back through the history on the neutron/tempest roles - this just feels familiar | 08:26 |
| snadge | if i ping and dump the lxcbr0 interface.. i can see echo requests coming from 10.255.255.39 which looks like the lxc interface | 08:27 |
| snadge | from the host | 08:28 |
| snadge | but no replies | 08:28 |
| jrosser | iirc that 10. range is only local to the host | 08:30 |
| jrosser | does your host have an ip on br-vlan? | 08:30 |
| snadge | it has two of them for some reason | 08:31 |
| jrosser | in the range of the ‘external’ neutron network | 08:31 |
| snadge | where is that defined? | 08:33 |
| *** ebbex has quit IRC | 09:00 | |
| *** udesale has quit IRC | 09:01 | |
| *** DanyC has joined #openstack-ansible | 09:06 | |
| *** DanyC has quit IRC | 09:07 | |
| *** DanyC has joined #openstack-ansible | 09:09 | |
| *** yolanda has joined #openstack-ansible | 09:09 | |
| *** rpittau|afk is now known as rpittau | 09:15 | |
| *** rgogunskiy has joined #openstack-ansible | 09:21 | |
| *** kopecmartin|off is now known as kopecmartin | 09:27 | |
| *** jonher has quit IRC | 09:40 | |
| *** jonher has joined #openstack-ansible | 09:41 | |
| CeeMac | morning | 09:50 |
| *** DanyC has quit IRC | 09:58 | |
| *** udesale has joined #openstack-ansible | 10:08 | |
| *** csmart has quit IRC | 10:16 | |
| *** masterpe has quit IRC | 10:16 | |
| *** DanyC has joined #openstack-ansible | 10:16 | |
| ChosSimbaOne | Hi. I am setting up infra noes for my production deployment of openstack. Previous I have only tired to deploy single infra node setup, but now I do 3 for the cluster resilience. So ealier i just did a interface with the public ip on the single host, but now i gues that it will be a "floating" ip with keepalived, so i should not hardcode the address to an interface, and just have the dns record fort ehe external_lb_vip_address | 10:29 |
| *** DanyC has quit IRC | 10:29 | |
| *** udesale has quit IRC | 10:31 | |
| *** DanyC has joined #openstack-ansible | 10:33 | |
| ChosSimbaOne | ahh, I relalized what I needed to do. Forgot that the interface is specified in the user_variables.yml | 10:36 |
| *** sum12 has joined #openstack-ansible | 10:42 | |
| *** cshen has quit IRC | 10:46 | |
| *** mensis has joined #openstack-ansible | 10:54 | |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/openstack-ansible-os_tempest master: Replace ostempest job with plain tripleo standalone https://review.opendev.org/697973 | 10:54 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/ansible-role-python_venv_build master: Replace ostempest job with plain tripleo standalone https://review.opendev.org/697974 | 10:57 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/ansible-config_template master: Replace ostempest job with plain tripleo standalone https://review.opendev.org/697975 | 11:00 |
| *** cshen has joined #openstack-ansible | 11:03 | |
| *** cshen has quit IRC | 11:07 | |
| noonedeadpunk | mornigns | 11:11 |
| noonedeadpunk | jrosser: regarding region and group_vars - I think this is smth we can/should move out from group vars into role defaults. | 11:11 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Set neutron_service_region in group_vars https://review.opendev.org/697981 | 11:13 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer stable/train: Pass _service_region to service_setup tasks https://review.opendev.org/697983 | 11:13 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_swift stable/train: Pass _service_region to service_setup tasks https://review.opendev.org/697984 | 11:13 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_nova stable/train: Pass nova_service_region to service_setup tasks https://review.opendev.org/697985 | 11:13 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_neutron stable/train: Pass neutron_service_region to service_setup tasks https://review.opendev.org/697986 | 11:14 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_keystone stable/train: Drop deprecated options https://review.opendev.org/697987 | 11:14 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Add reno regarding buster support https://review.opendev.org/697988 | 11:14 |
| noonedeadpunk | jrosser: btw, I'd love to hear your opinion regarding https://review.opendev.org/#/c/697830/ For me falling controller node used to bring a lot of pain as services API were timouting to response haproxy, which made horizon barely usable. | 11:16 |
| noonedeadpunk | but probably it's better to discuss on the meeting... | 11:17 |
| elenalindq | noonedeadpunk: Morning! The Friday change on murano.yml did the magic, yay for that! :) The upgrade still fails now because there is no placement container when we try to remove pip.conf. http://paste.openstack.org/show/k6QzoWvKKZdn8cEHt8Ob/ | 11:17 |
| noonedeadpunk | elenalindq: ok, so I think that re-running upgrade script (if it failed before finishing openstack_hosts) will probably fail, as we're adding end.v for placement but container itself was not created | 11:19 |
| noonedeadpunk | I feel like we probably do not really need this task nowadays and we can just drop it out | 11:19 |
| elenalindq | Yes, it fails fast when rerunning the upgrade. | 11:20 |
| elenalindq | noonedeadpunk: and yes, we don't get as far as to running the setup playbook so the container is not created. | 11:22 |
| noonedeadpunk | so what I can offer right now is either drop /etc/openstack_deploy/conf.d/placement.yml and remove group and host from inventory or launch the rest of the commands manually | 11:23 |
| jrosser | seems we didnt correctly fix the missing placement containers, that happened to us too | 11:24 |
| elenalindq | noonedeadpunk I commented out the remove pip.conf line in the run-upgrade.sh script and it continues with the upgrade. | 11:26 |
| noonedeadpunk | yeah, this is an option as well:) | 11:27 |
| noonedeadpunk | jrosser: I'm not sure we can drop pip-conf-removal.yml as it was re-added pretty recently https://opendev.org/openstack/openstack-ansible/commit/ebf844c4e9c857191030961c8f4355fa1e8dce46 | 11:28 |
| openstackgerrit | kourosh vivan proposed openstack/openstack-ansible-os_tempest master: Fix stackwiz venv pip install args https://review.opendev.org/694979 | 11:32 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Make upgrade script respect OSA_CONFIG_DIR https://review.opendev.org/697994 | 11:33 |
| *** cshen has joined #openstack-ansible | 11:34 | |
| *** csmart has joined #openstack-ansible | 11:39 | |
| *** masterpe has joined #openstack-ansible | 11:39 | |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible-os_nova stable/train: Remove deprecated filters https://review.opendev.org/697995 | 11:41 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Make upgrade script respect OSA_CONFIG_DIR https://review.opendev.org/697994 | 11:42 |
| jrosser | noonedeadpunk: but it was backport straight to stein? https://review.opendev.org/#/c/658145/ | 11:43 |
| jrosser | i think R->S was the big rework of the repo server | 11:43 |
| noonedeadpunk | oh... | 11:44 |
| noonedeadpunk | I think you can be right - didn't thought about such possibility | 11:45 |
| *** pcaruana has joined #openstack-ansible | 11:46 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Drop pip-conf-removal task https://review.opendev.org/697996 | 11:49 |
| noonedeadpunk | every time I think we're good to release some stuff raise | 11:50 |
| noonedeadpunk | and thanks elenalindq for testing this out and alerting us :) | 11:50 |
| *** DanyC_ has joined #openstack-ansible | 12:05 | |
| *** DanyC has quit IRC | 12:06 | |
| elenalindq | noonedeadpunk: Very happy to help! :) | 12:09 |
| *** DanyC_ has quit IRC | 12:10 | |
| *** dave-mccowan has joined #openstack-ansible | 12:19 | |
| *** ansmith has quit IRC | 12:25 | |
| *** dave-mccowan has quit IRC | 12:25 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_horizon master: opensuse: adjust for Python 3.x switch https://review.opendev.org/691833 | 12:28 |
| *** udesale has joined #openstack-ansible | 12:33 | |
| *** sum12 has quit IRC | 12:42 | |
| *** DanyC has joined #openstack-ansible | 12:47 | |
| *** DanyC_ has joined #openstack-ansible | 12:47 | |
| *** DanyC has quit IRC | 12:51 | |
| *** DanyC_ has quit IRC | 12:57 | |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible stable/stein: Calico scenario accidentally removed https://review.opendev.org/698009 | 12:58 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_neutron stable/stein: l3 agent on network_hosts do dvr_snat, anywhere else dvr https://review.opendev.org/693031 | 13:00 |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible-os_neutron stable/stein: l3 agent on network_hosts do dvr_snat, anywhere else dvr https://review.opendev.org/693031 | 13:00 |
| gshippey | You're too fast for me @noonedeadpunk ! | 13:02 |
| noonedeadpunk | gshippey: sorry for that:) | 13:03 |
| *** shyamb has joined #openstack-ansible | 13:08 | |
| *** DanyC has joined #openstack-ansible | 13:08 | |
| guilhermesp | mornings | 13:15 |
| noonedeadpunk | o/ | 13:18 |
| *** goldyfruit_ has quit IRC | 13:19 | |
| *** shyamb has quit IRC | 13:20 | |
| mgariepy | anyone had issue with ephemeral storage when upgrading to 18.04 ? | 13:20 |
| mgariepy | the _base image for the ephemeral part is formated with differents options, so the backing file for the disk.eph0 is not quite right and will not work for the vm being migrated. | 13:21 |
| chkumar|ruck | noonedeadpunk: jrosser mnaser please have a look https://review.opendev.org/#/q/topic:move_to_plain_standalone+(status:open+OR+status:merged) when free, first 3 patches | 13:26 |
| chkumar|ruck | thanks! | 13:26 |
| *** ansmith has joined #openstack-ansible | 13:27 | |
| *** zbr has quit IRC | 13:27 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/train: Add reno regarding buster support https://review.opendev.org/697988 | 13:29 |
| *** zbr has joined #openstack-ansible | 13:32 | |
| *** jamesden_ has joined #openstack-ansible | 13:35 | |
| *** jamesdenton has quit IRC | 13:36 | |
| *** miloa has quit IRC | 13:57 | |
| *** sum12 has joined #openstack-ansible | 14:03 | |
| openstackgerrit | Georgina Shippey proposed openstack/openstack-ansible stable/stein: Collect etcd logs https://review.opendev.org/693809 | 14:20 |
| elenalindq | noonedeadpunk Just wanted to let you know, the upgrade from Stein to Train failed again. This time when running placement-manage db sync because the table 'placement.resource_providers' does not exist. Ansible log, placement-manage db sync output, and tables show at http://paste.openstack.org/show/QmU9FJwZbY2zyJoH9thK/ . | 14:25 |
| noonedeadpunk | elenalindq: and what if you drop table alembic_version and re-run placement-manage db sync ? | 14:34 |
| jrosser | elenalindq: thats familiar, we're just looking at our notes here | 14:37 |
| elenalindq | noonedeadpunk: after dropping the alembic_version table, 'placement-manage db sync' worked. | 14:38 |
| jrosser | noonedeadpunk: that is the exact same error we got on train upgrade | 14:40 |
| noonedeadpunk | jrosser: and how did you workaround it? | 14:41 |
| jrosser | at that point we hadnt found the .rc file for the placement migration command | 14:42 |
| noonedeadpunk | elenalindq: do you have /etc/placement/placement-migrate-db.rc ? | 14:42 |
| jrosser | so we updated the my.cnf in the placement container to connect as root | 14:42 |
| jrosser | but i have to say we were having a bit of a nightmare with it and lots was broken, so that was really experimentation | 14:43 |
| elenalindq | noonedeadpunk: Yes, I have placement-migrate-db.rc placement.conf in /etc/placement | 14:43 |
| elenalindq | noonedeadpunk: the .rc file looks like this http://paste.openstack.org/show/DMiM9UlImx2WC6blUlTe/ | 14:44 |
| *** goldyfruit has joined #openstack-ansible | 14:44 | |
| noonedeadpunk | elenalindq: do you have https://review.opendev.org/#/c/696865/1/playbooks/os-placement-install.yml in /opt/openstack-ansible/playbooks/os-placement-install.yml ? | 14:46 |
| noonedeadpunk | I think you have... | 14:46 |
| elenalindq | Yes, I do. I deployed today and upgrade to stable/train. | 14:47 |
| elenalindq | s/upgrade/upgraded/ | 14:47 |
| jrosser | so thats a few steps of hacking we did that should not be necessary | 14:48 |
| jrosser | noonedeadpunk: we were thinking this was db permissions related | 14:48 |
| jrosser | if db users don't have access to tables created by other users the SHOW CREATE may fail | 14:48 |
| noonedeadpunk | it's weird that mysql-migrate-db.sh exited with 0 and no required tables were present... | 14:51 |
| noonedeadpunk | I think that once there's versions in alembic_version, placement-manage db sync ignores table creation as suggests that they should be already there | 14:51 |
| jrosser | noonedeadpunk: the fail / exit 0 / no tables suggests an error situation thats not handled at all in the migration script | 14:54 |
| noonedeadpunk | so it returns 1 https://opendev.org/openstack/placement/src/branch/master/placement_db_tools/mysql-migrate-db.sh#L158 but do not exit with 1 | 14:57 |
| noonedeadpunk | nah | 14:57 |
| * noonedeadpunk deploying sandbox | 14:59 | |
| openstackgerrit | Merged openstack/ansible-config_template master: Replace ostempest job with plain tripleo standalone https://review.opendev.org/697975 | 15:07 |
| jrosser | jamesden_: you around? | 15:11 |
| *** miloa has joined #openstack-ansible | 15:13 | |
| *** jamesden_ is now known as jamesdenton | 15:15 | |
| jamesdenton | jrosser present | 15:15 |
| jrosser | we had a wierd thing with this https://github.com/openstack/openstack-ansible-os_neutron/blob/master/templates/plugins/ml2/ml2_conf.ini.j2#L6 | 15:16 |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Replace ostempest job with plain tripleo standalone https://review.opendev.org/697973 | 15:17 |
| jamesdenton | what did you see? | 15:17 |
| jrosser | if you have a cloud with vxlan networks, you don't want 'vlan' appearing in the list of tenant_network_types | 15:17 |
| jamesdenton | that is likely true | 15:17 |
| jrosser | but then lets say you add a new provider network that is a vlan, maybe for SRIOV | 15:17 |
| jrosser | then all of a sudden you get vlan in tenant_network_types | 15:18 |
| *** chkumar|ruck is now known as raukadah | 15:19 | |
| jamesdenton | true. | 15:19 |
| jrosser | this seems to be becasue you have to specify a network type in the type: field in provider_networks | 15:19 |
| jamesdenton | yes, because provider network logic is used for container networks, too (raw vs everything else) | 15:20 |
| jrosser | it's confusing :) | 15:20 |
| jamesdenton | it could be better :D | 15:20 |
| jrosser | ok so theres not something silly we've missed there? | 15:20 |
| *** DanyC has quit IRC | 15:20 | |
| jrosser | the solution was to config override the ml2.tenant_network_types to not have vlan in it | 15:21 |
| jamesdenton | i don't think so. Could definitely improve documentation to call out that ^^ | 15:21 |
| jamesdenton | the idea of tenant network types is weird, because i can't really think of a situation where you would want network types to vary due to the exhaustion of a particular type. it's an ordered list, so the first one is utilized before the next (e.g. vxlan, vlan, etc) | 15:22 |
| jamesdenton | and if tenants can't specify a type, then it doesn't make sense (to me) | 15:22 |
| jrosser | yes, that is an odd model | 15:22 |
| *** DanyC has joined #openstack-ansible | 15:23 | |
| jamesdenton | we could add some logic to force only overlay types as tenant_network_types (vxlan, geneve, gre) and in the absence of those, default to vlan, and let operators override | 15:23 |
| jamesdenton | or just leave it alone | 15:24 |
| jrosser | perhaps really the key should be tenant_type | 15:27 |
| jrosser | and if thats not there it doesnt add it to the list | 15:27 |
| jrosser | but i dunno like you say this is used for too many things all at once right now anyway | 15:27 |
| jamesdenton | just a boolean? | 15:27 |
| jrosser | oh is see, is_tenant: true | 15:28 |
| jamesdenton | something like that | 15:28 |
| cjloader | morning | 15:29 |
| tmartins | Hey guys, I did something really cool this weekend! I deployed OSA-Train on tio of LXD! I mean, only the Compute Nodes and Ceph OSDs are actually LXD Containers! So cool! :-D | 15:30 |
| *** udesale has quit IRC | 15:36 | |
| *** rgogunskiy has quit IRC | 15:39 | |
| *** rgogunskiy has joined #openstack-ansible | 15:43 | |
| jrosser | tmartins: really - very interesting. I use LXD loads for stuff | 15:43 |
| jrosser | could you boot a vm inside the LXD compute node? | 15:44 |
| jamesdenton | jrosser We could add a new key (is_tenant) but leave it backwards compatible (default: false) as to not break upgrades and whatnot | 15:47 |
| *** johanssone has quit IRC | 15:48 | |
| *** rgogunskiy has quit IRC | 15:48 | |
| jrosser | wouldnt that take away the actual tenant networks though? | 15:48 |
| *** johanssone has joined #openstack-ansible | 15:49 | |
| jrosser | this would need to become is_tenant: true https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/openstack_user_config.yml.prod.example#L34-L43 | 15:51 |
| jrosser | maybe i misunderstand which way round the logic would be.....\ | 15:51 |
| jamesdenton | right - add an 'is_tenant' to that block to ensure "vxlan" is added to network_types for either neutron_provider_networks.network_types | 15:53 |
| jamesdenton | where it could get messy is if you DO want vlan as tenant networks, and have multiple vlan blocks there. you would really only have to add it to 1 | 15:54 |
| jrosser | i think theres a jinja filter for unique entries | 15:54 |
| admin0 | boostraping on 19.0.8 gives scripts/scripts-library.sh: line 324: iptables: command not found and Unit nspawn-macvlan.service could not be found. .. shouldn't bootstrap be taking care of that | 15:58 |
| admin0 | and looks like a mistake in the doc ..https://docs.openstack.org/openstack-ansible/stein/admin/upgrades/major-upgrades.html -- # git checkout 19.0.9 -- this tag does not exist yet | 16:00 |
| *** goldyfruit_ has joined #openstack-ansible | 16:01 | |
| *** goldyfruit has quit IRC | 16:04 | |
| *** cshen has quit IRC | 16:04 | |
| jamesdenton | jrosser I get what you mean. yes, default to true in the absence of the key entirely. | 16:15 |
| *** schwicht has joined #openstack-ansible | 16:18 | |
| *** schwicht has joined #openstack-ansible | 16:18 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift master: Replace git.openstack.org with opendev.org https://review.opendev.org/694352 | 16:30 |
| jrosser | what..... https://c7943b2bdbe378a62304-3c3896142b99a74b22b7b4d2994950df.ssl.cf5.rackcdn.com/691833/6/check/openstack-ansible-deploy-aio_distro_metal-ubuntu-bionic/ac8606f/logs/ara-report/result/5468ef3c-b055-4ddd-9dd6-990c06450513/ | 16:35 |
| jamesdenton | ¯\_(ツ)_/¯ | 16:37 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/train: Remove deprecated filters https://review.opendev.org/697995 | 16:38 |
| openstackgerrit | Merged openstack/ansible-config_template master: Add fallback option when stating the source file https://review.opendev.org/697547 | 16:38 |
| jrosser | "{{ 58 | random(seed='aio1', start=2) }}" always returns 48 | 16:39 |
| jrosser | so i don't know where it gets 'bad minute' from | 16:39 |
| cjloader | maybe chron job? | 16:43 |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer master: Replace git.openstack.org with opendev.org https://review.opendev.org/694292 | 16:44 |
| jrosser | cjloader: yes it's trying to set up a cron but putting aparrently bogus data in, but only sometimes | 16:46 |
| jrosser | i've seen it in two CI job failures now | 16:46 |
| *** rpittau is now known as rpittau|afk | 16:46 | |
| cjloader | hmm | 16:46 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_neutron stable/train: Use python3 for functional test tox environments https://review.opendev.org/698043 | 16:57 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_neutron stable/train: Use python3 for functional test tox environments https://review.opendev.org/698043 | 16:58 |
| openstackgerrit | Merged openstack/openstack-ansible-os_aodh master: Replace git.openstack.org with opendev.org https://review.opendev.org/694279 | 16:58 |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-os_neutron stable/train: Pass neutron_service_region to service_setup tasks https://review.opendev.org/697986 | 17:00 |
| jrosser | admin0: those messages from the scripts library log collection have been there for a very long time - if you can fix them it would be great | 17:04 |
| openstackgerrit | Merged openstack/ansible-config_template master: Replace git.openstack.org with opendev.org https://review.opendev.org/694363 | 17:06 |
| *** miloa has quit IRC | 17:07 | |
| *** gyee has joined #openstack-ansible | 17:16 | |
| admin0 | jrosser, you meant we can proceed with the upgrade even with those messages ? | 17:18 |
| jrosser | absolutely | 17:18 |
| jrosser | admin0: oh wait - they appear during bootstrap? i thought they came from the CI log collection | 17:19 |
| admin0 | they appear in the first step . during bootstrap | 17:19 |
| jrosser | admin0: reason i ask is they are part of the log collection, see here https://zuul.opendev.org/t/openstack/build/127a9d0a272c4adbb1dd5451328cdef2/log/job-output.txt#2485 | 17:20 |
| jrosser | and i don't think that should run as part of a manual deploy or an upgrade | 17:21 |
| openstackgerrit | Merged openstack/ansible-role-systemd_networkd master: Change upper constraints url https://review.opendev.org/694374 | 17:22 |
| admin0 | jrosser, https://pastebin.com/0mqiFVQQ | 17:23 |
| admin0 | here is where it appeared in my case | 17:23 |
| admin0 | Line 1114 and 1115 | 17:23 |
| *** rgogunskiy has joined #openstack-ansible | 17:24 | |
| *** rgogunskiy has quit IRC | 17:29 | |
| jrosser | admin0: but you see your failiure is actually at like 1090? | 17:29 |
| jrosser | *line | 17:29 |
| *** pcaruana has quit IRC | 17:31 | |
| admin0 | i see those as well | 17:32 |
| openstackgerrit | Merged openstack/openstack-ansible-tests stable/train: Use python3 for functional tests where possible https://review.opendev.org/697885 | 17:53 |
| noonedeadpunk | jrosser: ok, I've reproduced placement migration failure | 18:06 |
| jrosser | ah, how did you manage that? | 18:07 |
| admin0 | jrosser, all errors went away at 2nd bootstrap run :) | 18:11 |
| admin0 | oh wait . that was in master :( | 18:11 |
| openstackgerrit | James Denton proposed openstack/openstack-ansible master: Update scripts-library.sh to grab output when command exists https://review.opendev.org/698055 | 18:13 |
| jrosser | jamesdenton: \o/ | 18:15 |
| jrosser | awesome | 18:15 |
| jamesdenton | it's small, but helpful. hopefully | 18:15 |
| *** schwicht has quit IRC | 18:16 | |
| *** pcaruana has joined #openstack-ansible | 18:23 | |
| *** DanyC has quit IRC | 18:31 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/ansible-config_template master: Add interpreter line to action plugin https://review.opendev.org/698059 | 18:37 |
| *** schwicht has joined #openstack-ansible | 18:47 | |
| *** DanyC has joined #openstack-ansible | 19:37 | |
| *** DanyC has quit IRC | 19:47 | |
| *** DanyC has joined #openstack-ansible | 19:48 | |
| admin0 | quick clarification .. does neutron_agents run on_metal now ? or on containers | 19:56 |
| jamesdenton | metal by default, IIRC | 19:56 |
| admin0 | asking because i started ansible long time back and at that time, i had to override it using env.d | 19:56 |
| admin0 | ok | 19:56 |
| admin0 | openstack-ansible "${SCRIPTS_PATH}/upgrade-utilities/deploy-config-changes.yml is making noise on the only .env.d override i have there which is neuton_agents_container is_metal: true | 19:57 |
| *** DanyC has quit IRC | 19:59 | |
| *** andrea15 has quit IRC | 20:07 | |
| *** DanyC has joined #openstack-ansible | 20:21 | |
| *** poopcat has quit IRC | 20:43 | |
| *** DanyC has quit IRC | 20:46 | |
| *** poopcat has joined #openstack-ansible | 20:50 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova stable/train: Pass nova_service_region to service_setup tasks https://review.opendev.org/697985 | 21:15 |
| *** pcaruana has quit IRC | 21:31 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_swift stable/train: Pass _service_region to service_setup tasks https://review.opendev.org/697984 | 21:39 |
| *** ansmith has quit IRC | 21:39 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ceilometer stable/train: Pass _service_region to service_setup tasks https://review.opendev.org/697983 | 21:40 |
| *** nicolasbock has joined #openstack-ansible | 21:41 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/stein: Collect etcd logs https://review.opendev.org/693809 | 21:54 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone stable/train: Drop deprecated options https://review.opendev.org/697987 | 21:54 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron stable/train: Use python3 for functional test tox environments https://review.opendev.org/698043 | 21:57 |
| openstackgerrit | Merged openstack/openstack-ansible master: Drop pip-conf-removal task https://review.opendev.org/697996 | 22:03 |
| *** DanyC has joined #openstack-ansible | 22:10 | |
| *** elenalindq has quit IRC | 22:15 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Fix stackwiz venv pip install args https://review.opendev.org/694979 | 22:18 |
| *** schwicht has quit IRC | 22:21 | |
| *** schwicht has joined #openstack-ansible | 22:55 | |
| *** schwicht has quit IRC | 23:06 | |
| *** schwicht has joined #openstack-ansible | 23:09 | |
| *** ansmith has joined #openstack-ansible | 23:21 | |
| *** schwicht has quit IRC | 23:26 | |
| *** schwicht has joined #openstack-ansible | 23:30 | |
| *** schwicht has quit IRC | 23:42 | |
| *** goldyfruit_ has quit IRC | 23:50 | |
| *** tosky has quit IRC | 23:56 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!