| *** ianychoi has joined #openstack-ansible | 00:01 | |
| snadge | and from what I can see, vmware (vsphere) doesn't appear to support QinQ or double vlan tagging.. i didn't know that would be required? | 00:02 |
|---|---|---|
| *** hamzy has joined #openstack-ansible | 00:14 | |
| kplant | no, i was thinking outloud if it did not support trunks | 00:14 |
| *** marst has joined #openstack-ansible | 00:15 | |
| *** gyee has quit IRC | 00:15 | |
| mnaser | you wouldn't necessarily require it | 00:23 |
| mnaser | if you do vxlan networks only, that might be ok | 00:23 |
| mnaser | but you won't ever be able to access your VMs in there | 00:23 |
| mnaser | :P | 00:23 |
| kplant | you could do vxlan with an sdn | 00:37 |
| kplant | then plop a bastion on the vxlan(s) and a real l2/l3 interface | 00:38 |
| kplant | that was a joke, please don't do that | 00:51 |
| *** marst has quit IRC | 00:59 | |
| *** cshen has joined #openstack-ansible | 01:31 | |
| *** cshen has quit IRC | 01:35 | |
| snadge | im slowly learning about vlan, vxlan, qinq etc.. im a systems guy who feels horribly overwhelmed by all these new networking concepts | 02:00 |
| snadge | i mean.. I obviously understand vlans to the extent that I need to.. but it appears the rabbit hole goes very deep | 02:00 |
| *** markvoelker has joined #openstack-ansible | 02:00 | |
| snadge | so far neutron seems to cause me mentally the most amount of problems.. i feel fairly comfortable with everything else.. so i just need to get over this aversion, or do a course on it or something | 02:03 |
| kplant | i would maybe start with ccent/jncia | 02:06 |
| kplant | you'll want a solid base understanding of routing and switching | 02:07 |
| kplant | not even getting the certifications, they mean nothing, but the material for them if you learn it | 02:07 |
| *** jsquare has quit IRC | 02:19 | |
| *** jra has quit IRC | 02:19 | |
| *** jsquare has joined #openstack-ansible | 02:20 | |
| *** nurdie has joined #openstack-ansible | 02:20 | |
| *** nurdie has quit IRC | 02:24 | |
| *** markvoelker has quit IRC | 02:35 | |
| *** irclogbot_2 has quit IRC | 03:01 | |
| *** irclogbot_2 has joined #openstack-ansible | 03:01 | |
| *** kplant has quit IRC | 03:12 | |
| *** nurdie has joined #openstack-ansible | 03:14 | |
| *** kplant has joined #openstack-ansible | 03:16 | |
| *** nurdie has quit IRC | 03:20 | |
| *** partlycloudy has quit IRC | 03:21 | |
| *** raukadah is now known as chandankumar | 03:25 | |
| *** nicolasbock has quit IRC | 03:25 | |
| *** cshen has joined #openstack-ansible | 03:31 | |
| *** markvoelker has joined #openstack-ansible | 03:32 | |
| *** cshen has quit IRC | 03:40 | |
| *** hwoarang has quit IRC | 03:59 | |
| *** hwoarang has joined #openstack-ansible | 04:00 | |
| *** markvoelker has quit IRC | 04:04 | |
| *** udesale has joined #openstack-ansible | 04:13 | |
| *** marst has joined #openstack-ansible | 04:16 | |
| *** _d34dh0r53_ has joined #openstack-ansible | 04:29 | |
| *** marst has quit IRC | 04:35 | |
| *** prometheanfire has joined #openstack-ansible | 04:37 | |
| *** cloudkiller has joined #openstack-ansible | 04:57 | |
| *** markvoelker has joined #openstack-ansible | 05:02 | |
| *** markvoelker has quit IRC | 05:34 | |
| *** cshen has joined #openstack-ansible | 05:36 | |
| *** cshen has quit IRC | 05:40 | |
| *** cshen has joined #openstack-ansible | 06:21 | |
| *** markvoelker has joined #openstack-ansible | 06:31 | |
| fnpanic | good morning | 06:33 |
| *** pcaruana has joined #openstack-ansible | 06:41 | |
| *** ivve has joined #openstack-ansible | 06:42 | |
| *** DanyC has joined #openstack-ansible | 06:42 | |
| *** DanyC has quit IRC | 06:50 | |
| spotz | Morning | 06:51 |
| *** luksky has joined #openstack-ansible | 06:58 | |
| *** hamzy has quit IRC | 07:05 | |
| *** markvoelker has quit IRC | 07:05 | |
| *** hamzy has joined #openstack-ansible | 07:05 | |
| *** mnaser has quit IRC | 07:06 | |
| *** gundalow has quit IRC | 07:06 | |
| *** gundalow has joined #openstack-ansible | 07:06 | |
| *** mnaser has joined #openstack-ansible | 07:06 | |
| *** tosky has joined #openstack-ansible | 07:43 | |
| *** phasespace has joined #openstack-ansible | 07:46 | |
| *** hamzaachi has joined #openstack-ansible | 07:53 | |
| evrardjp | mnaser: as promised: https://review.openstack.org/#/c/651998/1 | 07:54 |
| evrardjp | now dealing with EM | 07:54 |
| *** ivve has quit IRC | 07:56 | |
| CeeMac | morning | 07:58 |
| *** markvoelker has joined #openstack-ansible | 08:02 | |
| evrardjp | morning | 08:03 |
| evrardjp | (but I am on holidays, so it's just a courtesy greeting, sorry :p ) | 08:03 |
| CeeMac | :) | 08:05 |
| *** ivve has joined #openstack-ansible | 08:11 | |
| *** luksky has quit IRC | 08:12 | |
| fnpanic | does anyone know if it is possible to not store the password on openrc as plaintext? | 08:18 |
| fnpanic | or another solution because having a service using openstack apis is a common scenario | 08:19 |
| fnpanic | because of the ad backend putting the password there is not a good idea | 08:19 |
| spotz | fnpanic You can not have it there but would then be prompted for it, not sure that helps | 08:20 |
| fnpanic | spotz: thanks but this does not help :-) | 08:22 |
| *** dxiri has joined #openstack-ansible | 08:23 | |
| spotz | Could you pass in the password from a script as you go? | 08:27 |
| *** dxiri has quit IRC | 08:28 | |
| evrardjp | fnpanic: you can use a passwords file to remove the password from your clouds.yaml, but not 100% sure what you can do to encrypt it from that on. Have you checked openstack client documentation ? | 08:28 |
| evrardjp | fnpanic: is there a reason you are afraid of having this in plain text in some *trusted* host? | 08:28 |
| spotz | fnpanic evrardjp: Yeah I'm not sure we can encrypt decrypt with Keystone... I don't think Barbican would help either but that might be a suggestion | 08:30 |
| evrardjp | I think I misunderstood the question | 08:30 |
| spotz | evrardjp: No I think you got it:) He doesn't want OS_PASSWORD to be plain text in openrc | 08:31 |
| evrardjp | but in which context for openstack-ansible? | 08:31 |
| spotz | I was just thinking in his deployment not OSA specific? | 08:32 |
| evrardjp | that's where I am confused | 08:32 |
| evrardjp | ok | 08:32 |
| spotz | might be wrong though | 08:32 |
| *** markvoelker has quit IRC | 08:34 | |
| evrardjp | fnpanic: unix file permissions is not enough for you? | 08:36 |
| evrardjp | (just asking) | 08:36 |
| spotz | odyssey4me and anyone else if you're coming to summit let me know and your google hangout info on channel or PM | 08:36 |
| evrardjp | the usual one for me :) | 08:36 |
| *** electrofelix has joined #openstack-ansible | 08:39 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_designate master: Updated from OpenStack Ansible Tests https://review.openstack.org/649001 | 08:49 |
| CeeMac | spotz, people still use google hangouts? | 08:50 |
| * CeeMac goes to put it back on his phone | 08:51 | |
| chandankumar | odyssey4me: hello | 08:52 |
| chandankumar | odyssey4me: Hello | 08:53 |
| spotz | CeeMac: just at summit for communication between the team for like dinner and such | 08:53 |
| CeeMac | spotz, i just presumed they'd shut that down along with google+ | 08:53 |
| chandankumar | odyssey4me: please have a look at this error http://logs.openstack.org/54/650054/7/check/openstack-ansible-functional-tempestconf-centos-7/d0f58e9/logs/ara-report/result/3bc8be06-8924-4ace-9add-c3403be12ae5/ when free, thanks! | 08:54 |
| CeeMac | So, I'm building a follow on environment from the RC I've been testing in and I'm planning on implementing untagged management vlan for host management | 08:56 |
| CeeMac | presumably, the deployment node still needs to be homed on the br-mgmt vlan so it can talk direct to the containers? | 08:57 |
| CeeMac | host=physical host | 08:57 |
| jrosser | CeeMac: deploy node does not need to be on br-mgmt | 08:58 |
| CeeMac | as long as their is a route to br-mgmt network right? | 08:58 |
| CeeMac | and hi jrosser :) | 08:58 |
| jrosser | It needs to be able to ssh to the ip of the physical hosts you put in user_config | 08:58 |
| jrosser | Nothing to do with mgmt net really at all | 08:58 |
| CeeMac | oh, so when it runs plays against the containers it communicates with them via the underlying host not direct? | 08:59 |
| jrosser | Once on the host the ansible connection plugin does its thing, without ssh being involved | 08:59 |
| CeeMac | theres an ansible connection plugin? | 08:59 |
| CeeMac | i missed that one | 08:59 |
| jrosser | Oh yes | 08:59 |
| CeeMac | presumably thats a connection plugin for lxc then? | 09:00 |
| jrosser | And nspawn too | 09:00 |
| CeeMac | right ==containers | 09:00 |
| CeeMac | thats clever :D | 09:00 |
| jrosser | Look in the plugins repo :) | 09:00 |
| * CeeMac goes to look | 09:01 | |
| * jrosser away for the rest of today.... | 09:01 | |
| spotz | CeeMac: nope still working:) | 09:01 |
| *** jsquare has quit IRC | 09:05 | |
| *** jsquare has joined #openstack-ansible | 09:05 | |
| *** ivve has quit IRC | 09:10 | |
| *** luksky has joined #openstack-ansible | 09:13 | |
| CeeMac | i seem to recall an issues around multiple repo nodes, was that in rocky? Is it better just to deploy 1? | 09:17 |
| odyssey4me | CeeMac That's stein, for now. | 09:20 |
| *** ivve has joined #openstack-ansible | 09:21 | |
| CeeMac | odyssey4me, ah ok | 09:21 |
| CeeMac | i'll try with 3 then and see what happens :0 | 09:22 |
| CeeMac | :) | 09:22 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/openstack-ansible-os_tempest master: Switch to import_task in os_tempest https://review.openstack.org/650054 | 09:27 |
| *** hamzaachi has quit IRC | 09:27 | |
| *** hamzaachi has joined #openstack-ansible | 09:28 | |
| *** markvoelker has joined #openstack-ansible | 09:31 | |
| CeeMac | hmm | 09:34 |
| CeeMac | its only the compute and network nodes that need br-vlan and br-vxlan right? | 09:34 |
| *** hamzaachi has quit IRC | 09:42 | |
| *** hamzaachi has joined #openstack-ansible | 09:43 | |
| *** hamzaachi has quit IRC | 09:51 | |
| *** hamzaachi has joined #openstack-ansible | 09:52 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_ironic master: Replace usage of netloc filters to urlsplit https://review.openstack.org/648190 | 09:58 |
| openstackgerrit | Merged openstack/openstack-ansible-os_nova master: Replace usage of netloc filters to urlsplit https://review.openstack.org/648189 | 10:00 |
| *** markvoelker has quit IRC | 10:04 | |
| *** Kurlee has joined #openstack-ansible | 10:10 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: Replace usage of netloc filters to urlsplit https://review.openstack.org/648188 | 10:11 |
| *** hamzaachi has quit IRC | 10:19 | |
| *** hamzaachi has joined #openstack-ansible | 10:22 | |
| *** pcaruana has quit IRC | 10:31 | |
| *** nicolasbock has joined #openstack-ansible | 10:44 | |
| *** markvoelker has joined #openstack-ansible | 11:02 | |
| *** udesale has quit IRC | 11:05 | |
| *** dxiri has joined #openstack-ansible | 11:06 | |
| *** dxiri has quit IRC | 11:11 | |
| *** ivve has quit IRC | 11:21 | |
| *** cshen has quit IRC | 11:25 | |
| *** cshen has joined #openstack-ansible | 11:25 | |
| *** ivve has joined #openstack-ansible | 11:30 | |
| *** markvoelker has quit IRC | 11:35 | |
| *** pcaruana has joined #openstack-ansible | 11:36 | |
| *** hamzaachi has quit IRC | 11:44 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_designate master: Updated from OpenStack Ansible Tests https://review.openstack.org/649001 | 11:58 |
| openstackgerrit | Vadim Kuznetsov proposed openstack/openstack-ansible-os_octavia master: Add support for using distribution packages for OpenStack services https://review.openstack.org/652049 | 11:59 |
| *** dxiri has joined #openstack-ansible | 12:01 | |
| openstackgerrit | Vadim Kuznetsov proposed openstack/openstack-ansible-os_octavia master: Add support for using distribution packages for OpenStack services https://review.openstack.org/652049 | 12:03 |
| *** starborn has joined #openstack-ansible | 12:07 | |
| *** dhellmann has left #openstack-ansible | 12:10 | |
| *** partlycloudy has joined #openstack-ansible | 12:28 | |
| *** darkomenz has joined #openstack-ansible | 12:31 | |
| darkomenz | Hello, is there any way to reduce memory consumption when running a multi node deploy. I have a machine that has used 8gb's of physical ram and and 12gb's of swap and is our of memory. As I understand 8GB should be the minimum. | 12:33 |
| darkomenz | ouot of memory * | 12:33 |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/openstack-ansible-os_tempest master: Use tempest_run_smoke for running smoke tests https://review.openstack.org/652060 | 12:40 |
| *** sm806 has quit IRC | 12:45 | |
| *** DanyC has joined #openstack-ansible | 12:45 | |
| *** timburke has quit IRC | 12:46 | |
| *** cjloader has quit IRC | 12:46 | |
| *** timburke has joined #openstack-ansible | 12:48 | |
| *** DanyC has quit IRC | 12:50 | |
| openstackgerrit | Chandan Kumar (raukadah) proposed openstack/openstack-ansible-os_tempest master: Use tempest_run_smoke for running smoke tests https://review.openstack.org/652060 | 12:51 |
| *** dxiri has quit IRC | 13:03 | |
| *** altlogbot_3 has joined #openstack-ansible | 13:03 | |
| *** dave-mccowan has joined #openstack-ansible | 13:07 | |
| jamesdenton | Ansible pros... what would cause Ansible to hang when gathering facts or perform any task against a host involving SSH, yet, SSH works fine from the CLI? | 13:10 |
| *** vnogin has joined #openstack-ansible | 13:11 | |
| ironfoot | is it ssh'ing into the right user? Use -vvvv to see all the debug information from ansible | 13:12 |
| *** dave-mccowan has quit IRC | 13:17 | |
| odyssey4me | jamesdenton it's always the MTU ;) | 13:17 |
| jamesdenton | :) | 13:17 |
| ironfoot | +inf | 13:17 |
| ironfoot | It's like the selinux of networking | 13:18 |
| *** hamzaachi has joined #openstack-ansible | 13:19 | |
| jamesdenton | odyssey4me you were not wrong | 13:22 |
| *** marst has joined #openstack-ansible | 13:23 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-nspawn_hosts master: debian: add support https://review.openstack.org/652071 | 13:23 |
| kplant | someone needs to write a windowing standard for ethernet | 13:24 |
| kplant | mtu/mfs mismatches are super annoying :-( | 13:24 |
| jamesdenton | ugh, i'm all over the place and forgot to set jumbo on the physical NIC. yikes. thanks everyone. | 13:27 |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-nspawn_container_create master: debian: add support https://review.openstack.org/652075 | 13:30 |
| openstackgerrit | Vadim Kuznetsov proposed openstack/openstack-ansible-os_octavia master: Add support for using distribution packages for OpenStack services https://review.openstack.org/652049 | 13:31 |
| *** altlogbot_3 has quit IRC | 13:32 | |
| *** altlogbot_2 has joined #openstack-ansible | 13:33 | |
| *** eglute has quit IRC | 13:33 | |
| *** cloudnull has quit IRC | 13:34 | |
| *** cloudkiller is now known as cloudnull | 13:34 | |
| *** d34dh0r53 has quit IRC | 13:35 | |
| *** phasespace has quit IRC | 13:38 | |
| *** altlogbot_2 has quit IRC | 13:38 | |
| *** altlogbot_2 has joined #openstack-ansible | 13:38 | |
| guilhermesp | cloudnull: around? | 13:50 |
| *** hamzaachi has quit IRC | 13:51 | |
| *** hamzaachi has joined #openstack-ansible | 13:52 | |
| CeeMac | jamesdenton, hi :) | 13:52 |
| CeeMac | is there any reason why you couldnt wire a bond.vlan interface to an openvswitch bridge? | 13:53 |
| jamesdenton | If you're asking what I think you're asking, it would limit you to a single flat network from neutron's perspective, since you're already tagging a vlan on that interface | 13:57 |
| *** altlogbot_2 has quit IRC | 14:00 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-nspawn_hosts master: debian: add support https://review.openstack.org/652071 | 14:04 |
| jamesdenton | CeeMac Does that make sense? | 14:06 |
| CeeMac | jamesdenton, it does | 14:07 |
| mnaser | morning all | 14:07 |
| CeeMac | i was thinking from the br-vxlan perspective | 14:08 |
| mnaser | evrardjp: I left some reviews, I can update your openstack/releases change, or you can if you want :) | 14:08 |
| CeeMac | although i'm still unclear on if/which the other bridges are required alongside openvswitch | 14:08 |
| CeeMac | as in all of the docs it only seems to reference br-provider | 14:08 |
| CeeMac | but there is br-int and br-tun - they get autocreated maybe? | 14:09 |
| jamesdenton | yes, br-int and br-tun are automatically created | 14:09 |
| jamesdenton | you simply need to create a provider bridge - and newish versions of OSA may create that for you, too | 14:09 |
| CeeMac | hmm | 14:10 |
| jamesdenton | br-vxlan is a vestige of containerized network agents, but it serves as an interface on top of which the VTEP address (for vxlan) is configured | 14:10 |
| CeeMac | and does that need to ovs as well for tenant networks? | 14:11 |
| CeeMac | oh wait, you're saying if the network agents are running on metal (which they do in rocky) you don't necessarily need br-vxlan? | 14:12 |
| CeeMac | in user_config br-vxlan binds to the neutron_linuxbridge_agent group by default | 14:13 |
| odyssey4me | jamesdenton that.is.hilarious. | 14:15 |
| jamesdenton | odyssey4me you took a stab in the dark, and were right. lol | 14:15 |
| jamesdenton | CeeMac There's an expectation that br-vxlan exists, as the provider definition for the vxlan network type is used in a playbook to dynamically fetch the VTEP IP from inventory, IIRC. Even for OVS. | 14:18 |
| CeeMac | right | 14:18 |
| CeeMac | if i understand things correctly, with ovs you still create seperate bridges, each with their own port | 14:19 |
| CeeMac | so i could bind bond.vlan to the br-vxlan bridge, and a dedicated/seperate bond for br-provider? | 14:19 |
| CeeMac | they're the only 2 bridges requied in ovs i think? the others (br-mgmt br-storage) would remain lxb? | 14:20 |
| jamesdenton | With OSA OVS, br-vxlan would also be a linux bridge containing bond.vlan. You only need to create br-provider as an OVS bridge | 14:21 |
| *** fdegir has joined #openstack-ansible | 14:22 | |
| CeeMac | oh | 14:22 |
| CeeMac | well | 14:22 |
| CeeMac | that makes things straightforward :) | 14:23 |
| CeeMac | and br-vlan would not then be required? I'm pure vxlan for tenant networks | 14:24 |
| jamesdenton | You wouldn't need br-vlan as a linux bridge, no. But you will need a provider bridge for OVS, and that can be called br-provider or br-vlan or br-ex or br-whatever | 14:24 |
| *** cshen has quit IRC | 14:25 | |
| CeeMac | right | 14:25 |
| jamesdenton | But it wouldn't be necessary on a compute, per say, if you're only doing overlay | 14:25 |
| CeeMac | unless dvr? | 14:25 |
| jamesdenton | However, because the docs bind that network to neutron_openvswitch_agent group, it might want it there regardless. | 14:25 |
| jamesdenton | Right, DVR | 14:25 |
| *** altlogbot_0 has joined #openstack-ansible | 14:26 | |
| jamesdenton | You *can* limit the group to 'network_hosts' instead of 'neutron_openvswitch_agent', in which case it ought to only be configured on the infras/network nodes | 14:26 |
| CeeMac | is it possible to inobtrusivley implement DVR at a future point if i start with ovs and legacy routers? | 14:26 |
| CeeMac | sorry. jumping around a bit i know | 14:27 |
| jamesdenton | There may be one or two overrides for it, IIRC. You may be better off rolling it out initially, then you can set an override to force tenant routers to legacy or HA when created by non-admins | 14:28 |
| CeeMac | makes sense | 14:29 |
| *** altlogbot_0 has quit IRC | 14:29 | |
| CeeMac | incidentally, i've abandoned netplan for these new server builds and am trying pure systemd-networkd | 14:29 |
| *** altlogbot_1 has joined #openstack-ansible | 14:30 | |
| jamesdenton | i went pure networkd, too. they all suck. | 14:30 |
| CeeMac | haha, true dat | 14:30 |
| CeeMac | its not as bad as i initially thought | 14:30 |
| CeeMac | although it took me a while to work out why my bond.vlan interface wasn't being created | 14:31 |
| jamesdenton | death by micro-segmentation | 14:31 |
| *** nurdie has joined #openstack-ansible | 14:31 | |
| *** altlogbot_1 has quit IRC | 14:33 | |
| CeeMac | jamesdenton, if i wanted to provision 2 provider switches, is that configurable through osa? | 14:39 |
| jamesdenton | yes | 14:39 |
| jamesdenton | just define two different 'vlan' blocks with their respective attributes | 14:40 |
| CeeMac | cool | 14:40 |
| CeeMac | want to seperate out internet facing traffic to transport/mpls traffic | 14:41 |
| *** vnogin has quit IRC | 14:41 | |
| jamesdenton | http://paste.openstack.org/show/749243/ | 14:41 |
| CeeMac | thanks | 14:43 |
| CeeMac | and that is the same regardless of lxb/ovs right? | 14:45 |
| CeeMac | and can be on the same host group | 14:46 |
| jamesdenton | yeah, the group bind could change (from neutron_linuxbridge_agent to neutron_openvswitch_agent) but in my example, those networks apply to 'compute_hosts' regardless of plugin | 14:46 |
| CeeMac | awesome, thanks | 14:47 |
| *** altlogbot_1 has joined #openstack-ansible | 14:48 | |
| CeeMac | out of interest are you using and sdn controller with ovs? | 14:49 |
| jamesdenton | i am not, no. | 14:51 |
| *** altlogbot_1 has quit IRC | 14:51 | |
| *** kplant has quit IRC | 14:51 | |
| *** altlogbot_0 has joined #openstack-ansible | 14:52 | |
| *** altlogbot_0 has quit IRC | 14:55 | |
| *** altlogbot_0 has joined #openstack-ansible | 14:56 | |
| *** altlogbot_0 has quit IRC | 14:56 | |
| *** kplant has joined #openstack-ansible | 14:57 | |
| CeeMac | hmm | 14:58 |
| CeeMac | if the dvrs exist in ovs | 14:58 |
| *** altlogbot_3 has joined #openstack-ansible | 14:58 | |
| CeeMac | how does the tenant vxlan network integrate with the dvr and break out via the lxb br-vxlan | 14:59 |
| CeeMac | think that's a monday question | 15:01 |
| CeeMac | ok, i'm out. have a great weekend osa people :) | 15:04 |
| noonedeadpunk | Thanks, you too CeeMac :) | 15:08 |
| jamesdenton | CeeMac It doesn't have to break out of br-vxlan in that way. OVS will use the IP address that is configured on br-vxlan, and thus traffic leaves bond.vlan connected to br-vxlan. There is no 'plumbing connection' between the lxb and ovs bridge, per say | 15:09 |
| chandankumar | guilhermesp: after turning all smoke tests few tests failed, we will fix it on monday! | 15:10 |
| *** tosky has quit IRC | 15:11 | |
| *** tosky has joined #openstack-ansible | 15:12 | |
| *** ivve has quit IRC | 15:20 | |
| *** electrofelix has quit IRC | 15:23 | |
| *** luksky has quit IRC | 15:24 | |
| dmsimard | odyssey4me, mnaser: heads up, things like http://git.openstack.org/cgit/openstack/openstack-ansible/tree/scripts/scripts-library.sh#n158 and http://git.openstack.org/cgit/openstack/openstack-ansible-tests/tree/test-ansible-env-prep.sh#n191 are eventually going to fail with the incoming opendev move | 15:28 |
| mnaser | dmsimard: ugh, that is going to be pretty bad | 15:30 |
| mnaser | I worry we will break existing releases | 15:30 |
| dmsimard | mnaser: src/git.openstack.org/openstack/ara should only be used in Zuul itself, though, right ? | 15:31 |
| mnaser | dmsimard: im wondering if there is other places we might depend on things | 15:31 |
| *** chandankumar is now known as raukadah | 15:36 | |
| logan- | for external dependencies the redirects should work fine | 15:38 |
| logan- | in zuul where we hard code src paths we'll have to fix that, but that can only work in the gate anyway, so we should never break a release due to it | 15:38 |
| dmsimard | mnaser: re-reading that code, it should probably not break after all | 15:45 |
| dmsimard | it checks if the directory exists (depends-on) and if it doesn't it installs from pypi | 15:45 |
| noonedeadpunk | Is there any rabbitmq expert? Preferably among cores:) | 15:47 |
| noonedeadpunk | Just need advice about some idea | 15:48 |
| *** hamzaachi has quit IRC | 15:49 | |
| *** hamzaachi has joined #openstack-ansible | 15:50 | |
| noonedeadpunk | ok, whatever. So yestarday phasespace pointed that we don't have ha enabled for all rabbitmq vhosts - only for /. It seems ,that the policies are not inherited from / iirc. Than my thought was to move rabbitmq_openstack_policies to all group_vars. | 15:52 |
| noonedeadpunk | And apply policies like https://github.com/openstack/openstack-ansible-rabbitmq_server/blob/master/tasks/rabbitmq_post_install.yml#L87-L98 | 15:52 |
| noonedeadpunk | For each role which is uses rabbit (almost every actually). So I was going to patch one role (probably ceilometer) during this weekends as an exmaple and to check the concept. | 15:54 |
| noonedeadpunk | So does anyone have any thoughts about it? | 15:54 |
| noonedeadpunk | And, probably, this all should be backported at least to stein? | 15:55 |
| *** cmart has joined #openstack-ansible | 15:58 | |
| openstackgerrit | Vadim Kuznetsov proposed openstack/openstack-ansible-os_octavia master: Add support for using distribution packages for OpenStack services https://review.openstack.org/652049 | 16:05 |
| *** gyee has joined #openstack-ansible | 16:15 | |
| *** cshen has joined #openstack-ansible | 16:19 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/ansible-role-python_venv_build master: Ensure venv_wheel_build_enable is evaluated as boolean https://review.openstack.org/652108 | 16:22 |
| *** cshen has quit IRC | 16:23 | |
| *** hamzaachi has quit IRC | 16:29 | |
| *** hamzaachi has joined #openstack-ansible | 16:29 | |
| mnaser | noonedeadpunk: I think it would be good to actually move the policy setup inside the role rather than in rabbitmq_server IMHO | 16:33 |
| noonedeadpunk | mnaser: That excatly what I was going to achieve - set policy inside every affected role | 16:35 |
| noonedeadpunk | But as the policy will be the same I though about adding rabbitmq_openstack_policies into gloabl group_vars | 16:35 |
| noonedeadpunk | somewhere here https://github.com/openstack/openstack-ansible-os_ceilometer/blob/master/tasks/mq_setup.yml#L35 | 16:36 |
| noonedeadpunk | and probably we don't need policy for / at all (not sure wether smth utilize / vhost) | 16:38 |
| *** tosky has quit IRC | 16:38 | |
| partlycloudy | Hello everyone, i got a rookie question on using Clos network with openstack. (continued from yesterday) | 16:39 |
| partlycloudy | now I understand the subnetting for br-mgmt, br-vxlan, br-storage networks for every leaf. | 16:39 |
| partlycloudy | what i don't know is how to bring the external network (which sits in a dedicate leaf) to every other leaf. | 16:39 |
| partlycloudy | sorry if that sounds like i duno what i'm doing (cos that exactly is the case here...) | 16:40 |
| *** bgmccollum has quit IRC | 16:41 | |
| partlycloudy | how can i give every leaf the access to the default gateway for the external network which is located on a different leaf? | 16:42 |
| jrosser | partlycloudy: hello :) | 16:43 |
| partlycloudy | jrosser: hello there! | 16:44 |
| jrosser | so i have a set of leaves that the external network comes to, running whichever routing protocol is needed for the upstream router | 16:45 |
| jrosser | and thats as far as the external networks go. network nodes and so on have an interface on those leaves to allow floating IP | 16:45 |
| jrosser | if you want an effective L2 network over your L3 fabric you'll need an overlay of some sort | 16:46 |
| jrosser | or you can split you external network into smaller subnets and make a "segmented provider network" which may be excellent/terrible fit for your use case | 16:49 |
| jrosser | really depends what you want to achieve | 16:49 |
| partlycloudy | jrosser: if i understand it correctly, the first solution is to connect a set of leaves (which contain network nodes or whatever requires external access) to the same L2 layer as the external router? | 16:50 |
| *** bgmccollum has joined #openstack-ansible | 16:51 | |
| jrosser | if all you are interested in is floating ip via neutron l3 agent then that would do it | 16:51 |
| jrosser | but if you want to attach instances directly to the external net, it wouldn't do it | 16:52 |
| partlycloudy | jrosser: yes, i actually wish to be able to connect instances to external network directly (bypass the network nodes). | 16:53 |
| *** cjloader has joined #openstack-ansible | 16:53 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/ansible-role-python_venv_build master: Ensure the build host is the first repo server https://review.openstack.org/652110 | 16:54 |
| jrosser | ok so that is fundamentally hard if you also want to build a L3 fabric, it's a sort of two orthoganol things | 16:54 |
| kplant | that's kind of scary isn't it? | 16:54 |
| odyssey4me | mnaser jrosser ^ that should solve the build host selection for a homogenous env... I'll try to figure something out for multi-arch/multi-distro envs | 16:54 |
| partlycloudy | jrosser: would something like EVPN/VXLAN help to connect every leaf together on the same L2 level with the external router? | 16:56 |
| noonedeadpunk | odyssey4me but in my case lsyncd seems to be configured on the last host.... | 16:57 |
| odyssey4me | noonedeadpunk oh? how the heck did that happen? | 16:57 |
| odyssey4me | note that it's the first host from the ansible inventory standpoint, not alphanumerically sorted | 16:58 |
| noonedeadpunk | so I have the following http://paste.openstack.org/show/749252/ and lsyncd is running on the uacloud-mgmt03 | 16:58 |
| odyssey4me | noonedeadpunk can you paste the output of: ansible -m ping repo_all | 16:58 |
| noonedeadpunk | Yep, you're right, they got reversed in inventory http://paste.openstack.org/show/749253/ | 16:59 |
| noonedeadpunk | So disregard:) | 16:59 |
| odyssey4me | noonedeadpunk yeah, that's the dynamic inventory as far as I know | 17:00 |
| *** cmart has quit IRC | 17:00 | |
| mnaser | odyssey4me: cloudnull was trying to look into that afaik | 17:02 |
| mnaser | sos just cc on that message :) | 17:02 |
| *** openstacking_123 has joined #openstack-ansible | 17:03 | |
| openstacking_123 | Anyone have an issue with the latest stable openstack-ansible rocky deploy having issues deleting images or restoring snapshots? | 17:03 |
| openstacking_123 | specifically seeing ile "rbd.pyx", line 2456, in rbd.Image.unprotect_snap . PermissionError: [errno 1] error unprotecting snapshot | 17:05 |
| openstacking_123 | In the glance api log | 17:05 |
| *** bgmccollum has quit IRC | 17:05 | |
| jrosser | partlycloudy: you should look at two things "routed provider networks" and like you say an alternative is EVPN or similar to make a virtual L2. Both of these have different pros/cons - also double check you really do need a L3 underlay for this :) | 17:05 |
| *** openstacking_123 has quit IRC | 17:06 | |
| *** hamzaachi has quit IRC | 17:08 | |
| partlycloudy | jrosser: thank you so much for showing me the way! i'll do my homework now and come back for help if i get jammed :-) | 17:10 |
| *** noonedeadpunk has quit IRC | 17:11 | |
| odyssey4me | good call on https://review.openstack.org/651598 by the way mnaser - just need another core to +2+w | 17:12 |
| mnaser | odyssey4me: slowly getting my hang on python_venv_build :) | 17:12 |
| mnaser | those are stuff that bubbled up to me from our stein deploy | 17:12 |
| odyssey4me | mnaser yeah, I do see that we're not using the global pins anywhere - so that will need to be instrumented in, otherwise we lose out on the pip/setuptools/wheel pins | 17:13 |
| *** cmart has joined #openstack-ansible | 17:14 | |
| odyssey4me | we're also not actually building the wheels for pip/setuptools/wheel anywhere, so we need to sort that out to help speed things up | 17:15 |
| raukadah | guilhermesp: what is the etherpad link for osa train planning? | 17:20 |
| raukadah | odyssey4me: is there a way to skip the python_venv_build handler here http://logs.openstack.org/54/650054/8/check/openstack-ansible-functional-tempestconf-centos-7/5cc3faa/job-output.txt.gz#_2019-04-12_10_31_43_727975? | 17:22 |
| raukadah | it is constantly failing | 17:23 |
| odyssey4me | raukadah it is failing because of something else there, not because of the handler | 17:23 |
| odyssey4me | either a var is not properly imported, or not evaluating properly | 17:23 |
| raukadah | odyssey4me: or is this one https://github.com/openstack/openstack-ansible-os_tempest/blob/master/tests/test-tempest-functional.yml#L32? | 17:24 |
| partlycloudy | jrosser: sorry to bother again. for the solution on "subneting provider networks", that would give each leaf a different external gateway, right? | 17:24 |
| odyssey4me | raukadah I dunno, but I haven't looked deeply into the specific test details and don't have the time to either. | 17:26 |
| raukadah | odyssey4me: I will check tomorrow! | 17:27 |
| *** ivve has joined #openstack-ansible | 17:27 | |
| *** dxiri has joined #openstack-ansible | 17:37 | |
| *** hamzaachi has joined #openstack-ansible | 17:38 | |
| *** darkomenz has quit IRC | 17:39 | |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-lxc_hosts stable/stein: Use pkill for lxc-dnsmasq systemd unit file https://review.openstack.org/652119 | 17:42 |
| *** openstacking_123 has joined #openstack-ansible | 17:45 | |
| openstacking_123 | Another thing I notice. Is all images are being created as protected in Ceph and a snapshot is made | 17:45 |
| jrosser | partlycloudy: for a routed provider network leaf would have its own gateway | 17:47 |
| logan- | openstacking_123: yeah, that is standard since ceph requires a protected snapshot to do copy on write image layering. are there any children of the snapshot existing still? that would prevent unprotecting it i'd think | 17:47 |
| openstacking_123 | Well I just uploaded a non snap shot brand new image and hit the same issue. | 17:49 |
| openstacking_123 | Unproctecting it in ceph allows it to delete | 17:49 |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-lxc_hosts stable/queens: Use pkill for lxc-dnsmasq systemd unit file https://review.openstack.org/652120 | 17:49 |
| partlycloudy | jrosser: i see. i'll dig into it. thank you so much for helping! | 17:49 |
| logan- | openstacking_123: are you using the same cephx key to do your manual unprotect as openstack is using? or are you using the admin key? i wonder if the openstack key lacks some permissions | 17:52 |
| openstacking_123 | Yeah using cephx with glance key from a glance container | 17:53 |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-lxc_hosts stable/pike: Use pkill for lxc-dnsmasq systemd unit file https://review.openstack.org/652123 | 17:55 |
| openstacking_123 | logan https://gist.github.com/fritzstauff/9e0d64cdeb09d504b935e6965f30bc6c | 17:55 |
| openstacking_123 | Is the the error ^ | 17:55 |
| openstacking_123 | But as an admin on ceph no problem | 17:56 |
| *** Kurlee has quit IRC | 17:56 | |
| openstacking_123 | Seems like it is trying to parse some rados gw buckets first. Which I would assume is unrelated since the snapshots are rbd based | 17:57 |
| logan- | just to verify your 'ceph auth list' looks something like https://pasted.tech/pastes/f90aa96ea1f109b2fb291c6e3a8d4024408c337b for the glance key right? | 17:59 |
| *** spatel has joined #openstack-ansible | 18:01 | |
| openstacking_123 | loga-- I think yours uses the older model? I have https://gist.github.com/fritzstauff/9e0d64cdeb09d504b935e6965f30bc6c . Which when compared to the ceph docs looks right | 18:02 |
| openstacking_123 | at http://docs.ceph.com/docs/mimic/rbd/rbd-openstack/ | 18:04 |
| logan- | ya profile rbd should be fine | 18:04 |
| openstacking_123 | So we have another Openstack cloud deployed on rocky from a few months ago and same thing. | 18:05 |
| *** openstacking_123 has quit IRC | 18:06 | |
| *** openstacking_123 has joined #openstack-ansible | 18:07 | |
| guilhermesp | raukadah: https://etherpad.openstack.org/p/osa-train-ptg | 18:08 |
| *** irclogbot_2 has quit IRC | 18:08 | |
| *** irclogbot_2 has joined #openstack-ansible | 18:10 | |
| *** dave-mccowan has joined #openstack-ansible | 18:12 | |
| *** openstac_ has joined #openstack-ansible | 18:12 | |
| openstac_ | logan- Seems like this issue to me http://lists.ceph.com/pipermail/ceph-users-ceph.com/2017-January/015682.html | 18:13 |
| openstac_ | Will let you know | 18:13 |
| logan- | yeah I was just looking at that thread too. kind of weird their docs seem to differ | 18:13 |
| openstac_ | Indeed | 18:13 |
| logan- | i checked another cluster which uses the 'profile rbd' caps | 18:14 |
| logan- | https://pasted.tech/pastes/fad4d37056604932b961c3e9c66cc2539f82f1db | 18:14 |
| logan- | the difference from yours is that my caps apply 'profile rbd' to mgr | 18:15 |
| *** openstacking_123 has quit IRC | 18:15 | |
| openstac_ | Excellent will test that now | 18:16 |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-nspawn_hosts master: debian: add support https://review.openstack.org/652071 | 18:18 |
| spatel | stein - should i put that in production ? | 18:18 |
| *** tosky has joined #openstack-ansible | 18:18 | |
| *** cshen has joined #openstack-ansible | 18:19 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-nspawn_hosts master: debian: add support https://review.openstack.org/652071 | 18:20 |
| *** cshen has quit IRC | 18:23 | |
| raukadah | guilhermesp: thanks! | 18:24 |
| openstac_ | logan- looks like same issue | 18:25 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/ansible-role-python_venv_build master: Ensure the build host is the first repo server https://review.openstack.org/652110 | 18:28 |
| openstac_ | logan- so it seems like all rgw buckets had issues. We added them as rbd profiles for now. We don't know what the proper fix is yet. But this will do for now | 18:40 |
| openstac_ | On thing openstack ansible makes the rgw buckets toward the end. So there maybe some differnt procedure in place for those. | 18:41 |
| logan- | yeah that's weird. i have rgw running on the 'profile rbd' cluster and no issues like this :/ | 18:42 |
| openstac_ | interesting will check that | 18:44 |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-nspawn_hosts master: debian: add support https://review.openstack.org/652071 | 18:46 |
| *** openstac_ has quit IRC | 18:54 | |
| *** openstacking_123 has joined #openstack-ansible | 18:57 | |
| openstacking_123 | logan- ceph auth caps client.glance mon "profile rbd" osd "allow class-read object_prefix rbd_children, profile rbd pool=volumes, profile rbd pool=images" | 18:58 |
| openstacking_123 | Looks like maybe the long term fix for mimic | 18:59 |
| logan- | nice | 18:59 |
| openstacking_123 | Anyways that is working for us. Thanks again for all the help | 18:59 |
| logan- | np, good to know for my luminous->mimic notes. thanks | 19:00 |
| kplant | no matter how many different ways i try, my multinode deployment of stable/rocky ends up corrupting the database during setup-openstack.yml: http://paste.openstack.org/show/749260/ i cleaned up the output with jq | 19:00 |
| kplant | not sure where i'm going wrong here | 19:00 |
| openstackgerrit | Merged openstack/ansible-role-python_venv_build master: Delete constraints and requirements files on build fail https://review.openstack.org/651598 | 19:02 |
| *** openstacking_123 has quit IRC | 19:04 | |
| openstackgerrit | Vadim Kuznetsov proposed openstack/openstack-ansible-os_octavia master: Add support for using distribution packages for OpenStack services https://review.openstack.org/652049 | 19:07 |
| *** openstacking_123 has joined #openstack-ansible | 19:08 | |
| kplant | os is centos 7, using lxc containers and an install from source | 19:09 |
| openstackgerrit | Merged openstack/ansible-role-python_venv_build master: Ensure venv_wheel_build_enable is evaluated as boolean https://review.openstack.org/652108 | 19:17 |
| openstacking_123 | kplant you doing any bonding. Can you describe your network setup between nodes. | 19:17 |
| kplant | no bonding, it's a very simple poc environment. there are 10 kvm guests (1 deploy, 3 infra, 3 ceph, 3 compute)' | 19:18 |
| kplant | each machine has a nic mapped to a specific network: external, mgmt, vxlan, vlan and storage | 19:18 |
| openstacking_123 | kplant what tag are you on? | 19:19 |
| openstacking_123 | git tag | 19:19 |
| kplant | 18.1.6-6-g7aff566 | 19:20 |
| openstacking_123 | kplant I would guess that is the issue. I would run 18.1.5 per the docs | 19:21 |
| openstacking_123 | Unless you are doing development you are going to run into trouble doing antying besides stable 18.1.5 | 19:22 |
| openstacking_123 | https://docs.openstack.org/project-deploy-guide/openstack-ansible/rocky/deploymenthost.html | 19:22 |
| kplant | okay, i can definitely give that a try | 19:23 |
| *** phasespace has joined #openstack-ansible | 19:23 | |
| kplant | does it make sense for something in a stable branch to purposefully not be stable though? | 19:23 |
| *** nurdie has quit IRC | 19:24 | |
| openstacking_123 | kplant speaking from experince just stable will give you issues | 19:24 |
| kplant | understood. i'm definitely going to give it a try, something i haven't done yet | 19:25 |
| mnaser | openstacking_123: let's not spread fud | 19:28 |
| mnaser | stable branches are fine. they're tested. tagged releases are nothing but stable branches that are tagged | 19:28 |
| mnaser | you can track stable with no problems | 19:28 |
| *** cshen has joined #openstack-ansible | 19:29 | |
| kplant | ^ that's what i was thinking | 19:29 |
| *** spatel has quit IRC | 19:30 | |
| openstacking_123 | mnaser not meaning to but. I have done a lot of deploys and always hit issues if I am not using the one in doc. | 19:31 |
| mnaser | right. I'd be happy to look into those issues, but there's no reason to call stable problematic, tags have failed in the past due to third parties also failing | 19:32 |
| kplant | mnaser: do you have any guesses as to what keeps blowing up the database? | 19:34 |
| *** openstacking_123 has quit IRC | 19:38 | |
| *** openstacking_123 has joined #openstack-ansible | 19:39 | |
| logan- | check the mysql logs in /var/log/mysql_logs on the galera container. also check for system issues like OOM on the node hosting the galera container. from your log, you lost the connection to mysql.. now you have to figure out why that happened | 19:40 |
| jsquare | we're using the nfs driver for cinder, stable/rocky, instances don't boot up, no errors whatsoever, console shows no boot device found, fdisk -l on the image files show no partitions | 19:41 |
| kplant | i did check out the galera error logs but didn't even think to check something as simple as memory | 19:41 |
| jsquare | anyone seen this? | 19:41 |
| logan- | kplant: if you do find that galera oomed or some resource issue like that, you might take a look at https://github.com/openstack/openstack-ansible/blob/master/tests/roles/bootstrap-host/templates/user_variables.aio.yml.j2 for some ideas on tuning for test systems. all of the tuning there allows for an OSA deploy on a 8vcpu/8gb test node in the openstack ci. | 19:43 |
| kplant | i do have each infra guest as 8c/16g but.. with ceph mons and mgrs on there | 19:46 |
| kplant | it might be pushing it over | 19:46 |
| *** openstacking_123 has quit IRC | 19:48 | |
| logan- | it would not surprise me. i've had issues running 16gb infra nodes before. 32gb is the smallest i've run recently, no issues there. | 19:48 |
| openstackgerrit | Merged openstack/openstack-ansible-ceph_client stable/rocky: Remove unnecessary GPG keys setting for YUM repos. https://review.openstack.org/650528 | 19:49 |
| kplant | i already toredown the env but i'm going to rebuild it with 32g infra nodes | 19:50 |
| kplant | that's a good idea, i hope you're right | 19:50 |
| mnaser | jsquare: do you actually see the files there? | 19:53 |
| jsquare | mnaser: yes, the files are there | 19:57 |
| *** openstacking_123 has joined #openstack-ansible | 19:57 | |
| *** starborn has quit IRC | 19:58 | |
| jsquare | if we try to list the partitions on the volume, there are none, I suspect that is the problem, don't know why | 19:59 |
| mnaser | jsquare: so /var/lib/nova/instances/<whatever> is there right? | 19:59 |
| mnaser | wait this is cinder | 19:59 |
| mnaser | so I mean your compute node has that mounted | 19:59 |
| jsquare | yes | 19:59 |
| jsquare | the compute node has it mounted | 19:59 |
| mnaser | so are you doing boot from volume? did you make sure you create a volume from an image? | 19:59 |
| jsquare | yes, launch instance, from image/create volume | 20:00 |
| jsquare | other deployments, non-OSA, don't show this behavior, not saying it's OSA's fault though | 20:01 |
| *** openstacking_123 has quit IRC | 20:05 | |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_container_create master: Updated from OpenStack Ansible Tests https://review.openstack.org/645904 | 20:06 |
| *** luksky has joined #openstack-ansible | 20:07 | |
| *** partlycloudy has quit IRC | 20:23 | |
| *** pcaruana has quit IRC | 20:26 | |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-nspawn_hosts master: debian: add support https://review.openstack.org/652071 | 20:28 |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-nspawn_container_create master: debian: add support https://review.openstack.org/652075 | 20:29 |
| *** Darcidride has joined #openstack-ansible | 20:30 | |
| *** aludwar has quit IRC | 20:37 | |
| mnaser | jsquare: maybe try creating a volume on nfs manually from image | 20:41 |
| mnaser | and see if that imag has the right data there? | 20:41 |
| openstackgerrit | Guilherme Steinmuller Pimentel proposed openstack/openstack-ansible-nspawn_hosts master: debian: add support https://review.openstack.org/652071 | 20:50 |
| *** cmart has quit IRC | 20:51 | |
| *** cshen has quit IRC | 20:52 | |
| openstackgerrit | Vadim Kuznetsov proposed openstack/openstack-ansible-os_octavia master: Add support for using distribution packages for OpenStack services https://review.openstack.org/652049 | 21:02 |
| *** cmart has joined #openstack-ansible | 21:04 | |
| guilhermesp | cores, could I get votes here https://review.openstack.org/#/c/652075/ ? needs to merge this in order to restore job voting here https://review.openstack.org/#/c/652071/ | 21:12 |
| *** cmart has quit IRC | 21:30 | |
| *** nurdie has joined #openstack-ansible | 21:32 | |
| *** cmart has joined #openstack-ansible | 21:33 | |
| *** kplant has quit IRC | 21:36 | |
| *** kplant has joined #openstack-ansible | 21:36 | |
| *** nurdie has quit IRC | 21:37 | |
| *** cmart has quit IRC | 21:38 | |
| *** cmart has joined #openstack-ansible | 21:41 | |
| *** marst has quit IRC | 21:46 | |
| *** hamzaachi has quit IRC | 21:56 | |
| *** Darcidride has quit IRC | 22:06 | |
| *** Darcidride has joined #openstack-ansible | 22:08 | |
| openstackgerrit | Mohammed Naser proposed openstack/ansible-role-python_venv_build stable/stein: Delete constraints and requirements files on build fail https://review.openstack.org/652161 | 22:29 |
| djhankb | How is the rsyslog configured when its deployed? In my deployment I had configured a host/ip in the "log_hosts" section - but I'm not exactly sure where the syslog data is supposed to go. Looking in the rsyslog container I don't have any logs and rsyslog doesn't seem to be configured at all. Are there some variables in user_variables.yml that would be required to get this working?" | 22:35 |
| *** cshen has joined #openstack-ansible | 22:49 | |
| *** cshen has quit IRC | 22:53 | |
| openstackgerrit | Merged openstack/openstack-ansible-nspawn_hosts master: debian: add support https://review.openstack.org/652071 | 22:57 |
| *** dave-mccowan has quit IRC | 23:00 | |
| *** partlycloudy has joined #openstack-ansible | 23:09 | |
| *** tosky has quit IRC | 23:11 | |
| *** luksky has quit IRC | 23:19 | |
| *** nicolasbock has quit IRC | 23:30 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!