| *** DanyC has joined #openstack-ansible | 00:08 | |
| *** DanyC has quit IRC | 00:12 | |
| *** eumel8 has quit IRC | 00:21 | |
| *** hwoarang has quit IRC | 00:22 | |
| *** hwoarang has joined #openstack-ansible | 00:23 | |
| *** macza has quit IRC | 00:23 | |
| *** TxGirlGeek has quit IRC | 00:29 | |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: Ignore deactivate not existing error https://review.openstack.org/631680 | 00:54 |
|---|---|---|
| dmsimard | cloudnull, evrardjp: is there any integration job running python3 ? Getting errors from a python2 interpreter in "openstack-ansible-deploy-aio_metal-ubuntu-bionic" when trying to install ara 1.0 packages: https://review.openstack.org/#/c/631262/ | 01:06 |
| *** macza has joined #openstack-ansible | 01:14 | |
| guilhermesp | chandan_kumar: any ideas on this http://logs.openstack.org/82/627782/4/check/openstack-ansible-upgrade-aio_lxc-ubuntu-bionic/5430fd4/logs/ara-report/result/93303b7a-2400-409a-96cc-090af9623497/ ? | 01:19 |
| *** macza has quit IRC | 01:27 | |
| guilhermesp | you can note that strangely it runs the test discovery with the old venv... | 01:36 |
| *** aedc has quit IRC | 01:43 | |
| *** kopecmartin|off has quit IRC | 01:50 | |
| *** kopecmartin has joined #openstack-ansible | 01:51 | |
| *** kopecmartin has quit IRC | 01:57 | |
| *** tridde has joined #openstack-ansible | 01:57 | |
| *** kopecmartin has joined #openstack-ansible | 01:59 | |
| *** partlycloudy has quit IRC | 02:00 | |
| *** jbadiapa has quit IRC | 02:00 | |
| *** sm806 has quit IRC | 02:00 | |
| *** errr has quit IRC | 02:00 | |
| *** trident has quit IRC | 02:00 | |
| *** sm806 has joined #openstack-ansible | 02:01 | |
| *** openstackgerrit has quit IRC | 02:02 | |
| *** errr has joined #openstack-ansible | 02:07 | |
| *** jbadiapa has joined #openstack-ansible | 02:07 | |
| *** kopecmartin has quit IRC | 02:09 | |
| *** strobelight has joined #openstack-ansible | 02:14 | |
| *** Soopaman has joined #openstack-ansible | 02:32 | |
| *** hamzy has joined #openstack-ansible | 02:33 | |
| *** strobelight_ has joined #openstack-ansible | 02:37 | |
| *** strobelight is now known as Guest3611 | 02:37 | |
| *** strobelight_ is now known as strobelight | 02:38 | |
| *** hwoarang has quit IRC | 02:40 | |
| *** Guest3611 has quit IRC | 02:40 | |
| *** hwoarang has joined #openstack-ansible | 02:41 | |
| *** Soopaman has quit IRC | 03:06 | |
| *** strobelight has quit IRC | 03:06 | |
| *** tridde is now known as trident | 03:06 | |
| *** Soopaman has joined #openstack-ansible | 03:21 | |
| *** macza has joined #openstack-ansible | 03:30 | |
| *** TxGirlGeek has joined #openstack-ansible | 03:35 | |
| *** macza has quit IRC | 03:36 | |
| *** hwoarang has quit IRC | 03:39 | |
| *** hwoarang has joined #openstack-ansible | 03:39 | |
| *** shyamb has joined #openstack-ansible | 03:41 | |
| *** vollman has joined #openstack-ansible | 03:42 | |
| *** udesale has joined #openstack-ansible | 03:47 | |
| *** sawblade6 has quit IRC | 03:54 | |
| *** sawblade6 has joined #openstack-ansible | 03:55 | |
| cloudnull | dmsimard I dont think our integrated repo is running py3 by default anywhere ? | 04:15 |
| cloudnull | mnaser ? | 04:15 |
| *** hwoarang has quit IRC | 04:16 | |
| *** hwoarang has joined #openstack-ansible | 04:18 | |
| *** macza has joined #openstack-ansible | 04:19 | |
| *** udesale has quit IRC | 04:26 | |
| *** udesale has joined #openstack-ansible | 04:32 | |
| *** udesale has quit IRC | 04:38 | |
| *** Soopaman has quit IRC | 04:45 | |
| *** hwoarang has quit IRC | 04:54 | |
| *** hwoarang has joined #openstack-ansible | 04:55 | |
| *** shyamb has quit IRC | 04:58 | |
| *** shyamb has joined #openstack-ansible | 04:59 | |
| *** shyamb has quit IRC | 05:01 | |
| *** shyamb has joined #openstack-ansible | 05:01 | |
| *** CeeMac has quit IRC | 05:12 | |
| *** macza has quit IRC | 05:17 | |
| *** hwoarang has quit IRC | 05:29 | |
| *** hwoarang has joined #openstack-ansible | 05:29 | |
| *** macza has joined #openstack-ansible | 05:32 | |
| *** radeks__ has joined #openstack-ansible | 05:36 | |
| *** radeks__ has quit IRC | 05:41 | |
| *** radeks__ has joined #openstack-ansible | 05:45 | |
| *** TxGirlGeek has quit IRC | 05:46 | |
| *** shyam89 has joined #openstack-ansible | 05:49 | |
| *** shyamb has quit IRC | 05:51 | |
| *** fnpanic has quit IRC | 05:52 | |
| *** macza has quit IRC | 06:01 | |
| chandan_kumar | jrosser: Hello | 06:04 |
| chandan_kumar | jrosser: is there a way to retrive https://github.com/openstack/openstack-ansible-os_tempest/blob/e1d2177898ef3427adb219805354053cccb84e10/tasks/tempestconf.yml#L45 keystone_service_internaluri from os_cloud? | 06:06 |
| *** TxGirlGeek has joined #openstack-ansible | 06:09 | |
| *** TxGirlGeek has quit IRC | 06:18 | |
| *** kopecmartin has joined #openstack-ansible | 06:34 | |
| *** hwoarang has quit IRC | 06:36 | |
| *** hwoarang has joined #openstack-ansible | 06:37 | |
| *** pcaruana has joined #openstack-ansible | 07:25 | |
| *** shyam89 has quit IRC | 07:39 | |
| *** shyamb has joined #openstack-ansible | 07:39 | |
| *** macza has joined #openstack-ansible | 07:44 | |
| *** udesale has joined #openstack-ansible | 07:47 | |
| *** macza has quit IRC | 07:49 | |
| *** pcaruana has quit IRC | 07:55 | |
| *** pcaruana has joined #openstack-ansible | 07:55 | |
| *** rgogunskiy has joined #openstack-ansible | 08:00 | |
| *** rgogunskiy has quit IRC | 08:07 | |
| *** shyamb has quit IRC | 08:10 | |
| *** mbi has joined #openstack-ansible | 08:28 | |
| *** tosky has joined #openstack-ansible | 08:38 | |
| chandan_kumar | jrosser: tosky we runned tempest in tripleo-ci http://logs.openstack.org/00/627500/53/check/tripleo-ci-centos-7-standalone-os-tempest/a32ce5e/job-output.txt.gz#_2019-01-18_07_49_40_340221 we need to figure out the correct loggin path stuff now | 08:46 |
| tosky | the correct logging path? | 08:55 |
| chandan_kumar | tosky: I mean to dump stestr.html files at proper place so that it will be shown in logs | 09:01 |
| chandan_kumar | tosky: current stestr files are here http://logs.openstack.org/00/627500/53/check/tripleo-ci-centos-7-standalone-os-tempest/a32ce5e/logs/undercloud/var/log/tempest/stestr_results.html.gz | 09:03 |
| *** shyamb has joined #openstack-ansible | 09:03 | |
| chandan_kumar | tosky: stackviz is missing i think | 09:03 |
| tosky | chandan_kumar: arx is probably more expert on that part, didn't he add code for stackviz support? | 09:06 |
| *** DanyC has joined #openstack-ansible | 09:16 | |
| *** aludwar has quit IRC | 09:17 | |
| *** hamzaachi has joined #openstack-ansible | 09:18 | |
| *** DanyC has quit IRC | 09:20 | |
| *** shyamb has quit IRC | 09:34 | |
| *** rgogunskiy has joined #openstack-ansible | 09:39 | |
| *** shyamb has joined #openstack-ansible | 09:42 | |
| *** rgogunskiy has quit IRC | 10:00 | |
| *** aedc has joined #openstack-ansible | 10:00 | |
| *** hamzaachi_ has joined #openstack-ansible | 10:04 | |
| *** hamzaachi has quit IRC | 10:04 | |
| *** DanyC has joined #openstack-ansible | 10:07 | |
| *** eumel8 has joined #openstack-ansible | 10:22 | |
| *** openstackgerrit has joined #openstack-ansible | 10:27 | |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: Convert test network setup to use the networkd role https://review.openstack.org/592596 | 10:27 |
| *** CeeMac has joined #openstack-ansible | 10:29 | |
| *** dcdamien has joined #openstack-ansible | 10:36 | |
| dcdamien | morning | 10:37 |
| dcdamien | would anybody like to review this backport? https://review.openstack.org/#/c/630371/ | 10:37 |
| *** hamzaachi has joined #openstack-ansible | 10:40 | |
| *** hamzaachi_ has quit IRC | 10:43 | |
| *** hamzaachi has quit IRC | 10:44 | |
| *** hamzaachi has joined #openstack-ansible | 10:45 | |
| CeeMac | morning all | 10:52 |
| *** shyamb has quit IRC | 10:58 | |
| *** electrofelix has joined #openstack-ansible | 11:01 | |
| *** kopecmartin has quit IRC | 11:07 | |
| CeeMac | quick question. In the openstack_user_config.yml host_bind_override is set to "eth12" and there is ifupdown configuration for using pre-up to configure a veth-pair for non-containerized neutron agents | 11:12 |
| CeeMac | I've moved on to netplan, and this doesn't support pre-up from what I can see, does the host_bind_override need to be eth12 or can it be a physical interface | 11:13 |
| CeeMac | and if it can be a physical interface, is the veth-pair still required? | 11:13 |
| chandan_kumar | tosky: I think I am missing something on tripleo code | 11:16 |
| chandan_kumar | let me fix that part | 11:16 |
| *** shyamb has joined #openstack-ansible | 11:16 | |
| *** macza has joined #openstack-ansible | 11:22 | |
| *** macza has quit IRC | 11:26 | |
| jamesdenton | host_bind_override would be defined on a compute node (usually) and is expected to be a physical interface of some kind that will be used in lieu of a bridge (like inside neutron_agents container). It's what ultimately is used for the Neutron provider bridge mapping (e.g. physnet1:eth1 instead of physnet1:br-vlan) | 11:27 |
| openstackgerrit | Chandan Kumar proposed openstack/ansible-role-python_venv_build master: Use usr/local/share/ansible/roles for data_files https://review.openstack.org/631777 | 11:28 |
| CeeMac | hi jamesdenton | 11:29 |
| CeeMac | compute node, right, that makes sense | 11:29 |
| CeeMac | hmm | 11:30 |
| CeeMac | so, if i was to stick with the default br-vlan in the example configuration | 11:31 |
| CeeMac | it requires the veth-pair? | 11:31 |
| jamesdenton | like, for a production environment? | 11:31 |
| CeeMac | yeah | 11:31 |
| *** odyssey4me has joined #openstack-ansible | 11:31 | |
| jamesdenton | what page are you looking at? | 11:31 |
| nowster | horizon is crashing because it doesn't have an orchestration policy file. | 11:32 |
| CeeMac | https://docs.openstack.org/project-deploy-guide/openstack-ansible/draft/app-config-prod.html | 11:32 |
| odyssey4me | hi folks - apologies for my absence on IRC for the last 2 days... I mistakenly deleted my ZNC bouncer and only just got it back up again | 11:32 |
| CeeMac | i'll only be using the vlan for external/provider networks, which is why i thought network node rather than compute node | 11:33 |
| jamesdenton | CeeMac Unless you *need* a flat network, you can get rid of the flat network block. | 11:33 |
| jamesdenton | for the vlan network block, you can set host_bind_override to whatever physical interface you'll use for Neutron (eth1, bond1, etc) | 11:33 |
| jamesdenton | no veth needed | 11:33 |
| jamesdenton | The eth12 or eth11 interfaces specified there are what would be used inside the neutron_agents container. But in Queens+, that isn't likely to exist. | 11:34 |
| CeeMac | ah, cool | 11:34 |
| CeeMac | i remember i had issues previously with linux bridge agent not starting due to eth12 not existing | 11:35 |
| CeeMac | but if i remove the flat block then it won't try and use that anyways | 11:35 |
| jamesdenton | right | 11:35 |
| jamesdenton | brb in a few | 11:35 |
| CeeMac | are the neutron provider bridge mappings set in a configuration file somewhere during neutron deployment? or is that tied up in the post-deployment config | 11:36 |
| *** kopecmartin has joined #openstack-ansible | 11:38 | |
| jamesdenton | they're actually created dynamically based on what you're defining in openstack_user_config.yml | 11:39 |
| CeeMac | based on the network blocks? | 11:40 |
| CeeMac | *facepalm* | 11:40 |
| CeeMac | right under where it says "provider_networks" right? | 11:40 |
| nowster | orchestration == heat, but the default horizon container seems to be missing that policy.conf | 11:42 |
| CeeMac | follow up question | 11:42 |
| *** shyamb has quit IRC | 11:42 | |
| *** shyamb has joined #openstack-ansible | 11:42 | |
| CeeMac | i remember we talked around this issue briefly before | 11:42 |
| CeeMac | on the network node, i have a dedicated interface (trunk) configured for external/provider vlans | 11:43 |
| CeeMac | should this be attached to br-vlan, or is there a config override to tell neutron which nic to use | 11:44 |
| jamesdenton | the interface differs on the network node vs the compute node? | 11:44 |
| CeeMac | yes | 11:45 |
| jamesdenton | are you planning to use vlans for provider and tenant networks? Or were you planning to use vxlan for tenant nets? | 11:45 |
| CeeMac | vxlan for tenant | 11:46 |
| jamesdenton | oh ok | 11:46 |
| CeeMac | i've got that covered | 11:46 |
| CeeMac | (i think/hope) | 11:46 |
| jamesdenton | in that case, define what you are using on the network node | 11:46 |
| jamesdenton | since you won't be using vlan networks on the compute | 11:46 |
| CeeMac | its the br-vlan parts that are getting me | 11:46 |
| jamesdenton | yeah, that could probably be explained a little better | 11:46 |
| CeeMac | would the linux-bridge-agent on the compute node not still need to be able to bind to something? | 11:47 |
| CeeMac | thats where i seem to run into problems | 11:47 |
| jamesdenton | Yes - but ONLY if you spin up a VM on the vlan network will it try | 11:47 |
| CeeMac | right, so on agent service starup it wouldn't automatically attempt to connect to an interface/bridge? | 11:48 |
| jamesdenton | if you spin up a VM in a vxlan network, it should be fine. You would attach the neutron router to the vlan network and the vxlan network, and since thats on the network node, it would work | 11:48 |
| jamesdenton | that is correct | 11:48 |
| CeeMac | cool | 11:48 |
| CeeMac | so | 11:48 |
| CeeMac | on the network node, i configure br-vlan and bind it to the dedicated nic I have configured for the provider vlans | 11:48 |
| CeeMac | ? | 11:48 |
| chandan_kumar | odyssey4me: Hello | 11:49 |
| jamesdenton | Is this queens? or rocky? | 11:49 |
| odyssey4me | o/ chandan_kumar how're things? | 11:49 |
| CeeMac | stable/rocky | 11:49 |
| chandan_kumar | odyssey4me: http://logs.openstack.org/00/627500/53/check/tripleo-ci-centos-7-standalone-os-tempest/a32ce5e/job-output.txt.gz#_2019-01-18_07_49_40_340221 | 11:49 |
| chandan_kumar | odyssey4me: tempest runned there | 11:49 |
| CeeMac | with ubunto bionic | 11:49 |
| jamesdenton | CeeMac Ok. In that case, configure br-vlan but don't put your physical interface in it. Set host_bind_override to that interface and proceed with the install | 11:50 |
| chandan_kumar | odyssey4me: stackviz not created that I fixed right now waiting for next run | 11:50 |
| jamesdenton | brb | 11:50 |
| *** shyamb has quit IRC | 11:52 | |
| odyssey4me | chandan_kumar awesome :) | 11:52 |
| odyssey4me | chandan_kumar how did you get past the restrictions on implementing action plugins? | 11:53 |
| chandan_kumar | odyssey4me: But I need some help here https://review.openstack.org/#/c/631708/ | 11:53 |
| CeeMac | jamesdenton, with netplan, if i have empty interfaces: [] under the bridge will it still come up? | 11:53 |
| *** shyamb has joined #openstack-ansible | 11:53 | |
| chandan_kumar | odyssey4me: for action plugins https://review.openstack.org/#/c/628421/11/ansible.cfg I made the changes here | 11:53 |
| jamesdenton | CeeMac good question. i can't recall. it shouldn't matter, though, in this case | 11:54 |
| odyssey4me | chandan_kumar ah ok, that makes sense | 11:54 |
| CeeMac | ok, i'll give it a whirl and see what happens then :) | 11:54 |
| chandan_kumar | odyssey4me: and for devstack side I made this one https://review.openstack.org/#/c/627482/57/playbooks/python-tempestconf-config-template.yaml but not sure it will work because I am checking the right physcial network provider in devstack | 11:54 |
| odyssey4me | we'll see | 11:55 |
| CeeMac | just finishing prepping the rest of the environment. I gave up on the virtual environment and have managed to provision up some blades so hopefully i wont run into the container networking issue this time! | 11:55 |
| odyssey4me | chandan_kumar how can I help with https://review.openstack.org/#/c/631708/ ? | 11:55 |
| chandan_kumar | odyssey4me: yes I donot want user to set keystone_service_internaluri these values | 11:55 |
| chandan_kumar | odyssey4me: otherwise I am getting undefined var | 11:56 |
| chandan_kumar | odyssey4me: is there a way to retrive from clouds.yaml file and set it there? | 11:56 |
| openstackgerrit | Merged openstack/ansible-role-python_venv_build master: Use usr/local/share/ansible/roles for data_files https://review.openstack.org/631777 | 11:56 |
| CeeMac | jamesdenton, to confirm, i can use host_bind_interface in the "vlan" type br-vlan network? | 11:57 |
| *** shyamb has quit IRC | 11:58 | |
| nowster | https://github.com/openstack/horizon/tree/master/openstack_dashboard/conf is missing orchestration_policy.json ? | 11:59 |
| odyssey4me | chandan_kumar it looks like clouds.yaml is on compute1 and not on tempest1: http://logs.openstack.org/08/631708/1/check/openstack-ansible-functional-tempestconf-centos-7/86a6cc8/logs/ara-report/result/33dc1e46-5ffd-40e3-94e6-ccca70f0d466/ | 12:00 |
| odyssey4me | chandan_kumar ok, so it seems we should add https://github.com/openstack/openstack-ansible-tests/blob/fe6c8344d1cdf23add574a461af280d4033b8428/test-prepare-host.yml#L76-L85 to https://github.com/openstack/openstack-ansible-tests/blob/fe6c8344d1cdf23add574a461af280d4033b8428/test-install-tempest.yml | 12:04 |
| *** vnogin has joined #openstack-ansible | 12:04 | |
| chandan_kumar | odyssey4me: checking | 12:05 |
| *** shyamb has joined #openstack-ansible | 12:05 | |
| *** vnogin has quit IRC | 12:09 | |
| jamesdenton | CeeMac Absolutely. Let's say you have bond1 in the br-vlan bridge. And you don't have host_bind_override specified. When Neutron creates a linuxbridge for your network (ex. brq-XXXXXX) it will create a vlan-tagged subinterface off br-vlan (ex. br-vlan.100) and put that in the brq-XXXXX bridge. That definitely won't work for a 'flat' network type, as Linux won't put a bridge into a bridge (untagged br-vlan into | 12:20 |
| jamesdenton | brq-XXXXX). That's where host_bind_override comes from, and why it's only listed on the flat network block. It gives the operator a way to specify a physical (non-bridge) interface for flat networks. BUT, I also like to use it for vlan networks, because you avoid the extra hop/latency introduced by the br-vlan and subsequent tagged subinterface. | 12:20 |
| CeeMac | ok, cool, thanks :) | 12:22 |
| CeeMac | so by using the host_bind_override it will create the vlan-tagged interface directly off the specified adapter instead? | 12:23 |
| jamesdenton | exactly | 12:23 |
| CeeMac | does br-vlan actually get used then, or does it just need to exist "because" (as in specfied in netplan) | 12:24 |
| jamesdenton | in that case it would probably not be actively used. But there's an assumption it's there, or rather, i/we don't know the affects of not having it there at this time | 12:25 |
| CeeMac | sure | 12:25 |
| CeeMac | no harm in having it and not using right | 12:25 |
| jamesdenton | the bridges are/were there as a means to connect LXC containers and various interfaces to the network. now that neutron agents run on baremetal by default, they're less important, but still part of the refarch, so... | 12:26 |
| CeeMac | yeah | 12:26 |
| CeeMac | well, I'll see if it works in a little bit! | 12:27 |
| jamesdenton | good luck! | 12:27 |
| CeeMac | thanks! | 12:28 |
| CeeMac | just got to take a lvm snapshot on all of the nodes, then I can kick off the deployment | 12:28 |
| *** vnogin has joined #openstack-ansible | 12:29 | |
| *** shyamb has quit IRC | 12:32 | |
| CeeMac | oh, btg, can you remember the config override that sets the domain name in the environment? I think it uses openstack.local or something by default? | 12:39 |
| jamesdenton | not offhand, sorry | 12:39 |
| CeeMac | no worries, i'll see if i can dig it out of there reference docs | 12:40 |
| *** ansmith has joined #openstack-ansible | 12:42 | |
| *** strobelight has joined #openstack-ansible | 12:48 | |
| *** vnogin has quit IRC | 12:48 | |
| *** strobelight_ has joined #openstack-ansible | 12:49 | |
| *** strobelight is now known as Guest31101 | 12:49 | |
| *** strobelight_ is now known as strobelight | 12:50 | |
| *** vnogin has joined #openstack-ansible | 12:52 | |
| *** Guest31101 has quit IRC | 12:53 | |
| *** udesale has quit IRC | 13:07 | |
| *** udesale has joined #openstack-ansible | 13:08 | |
| *** canori01 has joined #openstack-ansible | 13:08 | |
| canori01 | hey guys, I'm having a problem where when I run the neutron playbook, it resets the rabbit password. The password in the config matches what's on the user_secrets.yml, but neutron gets login failed until I manually reset the password. Anyt thoughts as to what could be doiung that? | 13:12 |
| *** strobelight_ has joined #openstack-ansible | 13:16 | |
| *** strobelight is now known as Guest19750 | 13:16 | |
| *** strobelight_ is now known as strobelight | 13:17 | |
| *** Guest19750 has quit IRC | 13:20 | |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-tests master: Setup clouds.yaml on tempest node https://review.openstack.org/631794 | 13:29 |
| chandan_kumar | odyssey4me: ^^ | 13:30 |
| *** vnogin has quit IRC | 13:40 | |
| *** vnogin has joined #openstack-ansible | 13:41 | |
| chandan_kumar | odyssey4me: I think there is some problem with stackviz when tempest is installed from distro | 13:47 |
| chandan_kumar | odyssey4me: I need to fix that | 13:47 |
| chandan_kumar | sorry may be I am doing soemthing wrong | 13:49 |
| chandan_kumar | odyssey4me: at here http://logs.openstack.org/00/627500/54/check/tripleo-ci-centos-7-standalone-os-tempest/a26d2d0/logs/ara_oooq/result/d382b8ac-526d-4022-97c3-0d7ab48ea672/ after subunit generation it got terminated | 13:52 |
| chandan_kumar | odyssey4me: I think it is terminating it https://github.com/openstack/openstack-ansible-os_tempest/blob/master/tasks/tempest_run.yml#L91 is it possible to generate stackviz irrescpective of the tests fails? | 13:55 |
| *** udesale has quit IRC | 14:01 | |
| *** udesale has joined #openstack-ansible | 14:02 | |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_tempest master: [DNM] testing os_cloud https://review.openstack.org/631708 | 14:13 |
| *** zul has joined #openstack-ansible | 14:21 | |
| *** priteau has joined #openstack-ansible | 14:25 | |
| *** aludwar has joined #openstack-ansible | 14:27 | |
| cloudnull | good mornings | 14:31 |
| kaiokmo | mornings cloudnull | 14:33 |
| cloudnull | o/ | 14:33 |
| nowster | I'm troubleshooting a new install (from source): https://github.com/openstack/horizon/tree/master/openstack_dashboard/conf is missing orchestration_policy.json ? | 14:34 |
| cloudnull | nowster is that a new install from master? | 14:36 |
| nowster | yes | 14:40 |
| nowster | I'm getting crashes in horizon because the file doesn't exist and it's trying to load a None value. | 14:40 |
| nowster | http://paste.openstack.org/show/742960/ | 14:42 |
| nowster | the "looking up policy for service..." bits are my debugging output | 14:42 |
| cloudnull | interesting. there's been a lot of change in the policy front recently. | 14:47 |
| cloudnull | maybe something we just need to include? | 14:47 |
| cloudnull | I'm assuming "orchestration_policy" is from heat | 14:47 |
| cloudnull | https://github.com/openstack/horizon/tree/master/openstack_dashboard/conf | 14:48 |
| cloudnull | oh.. ha wrong link. | 14:48 |
| cloudnull | https://github.com/openstack/openstack-ansible-os_horizon/blob/f32909cb7f20a702ac483276aa55747d7a54d7eb/tasks/horizon_install_source.yml#L122-L136 | 14:50 |
| cloudnull | nowster mind seeing you have a heat_policy.json file in your horizon container ? | 14:53 |
| cloudnull | maybe the name has changed? | 14:53 |
| cloudnull | https://github.com/openstack/openstack-ansible-os_horizon/blob/0d44c7dca53449002c394fbaae920acf44736815/templates/horizon_local_settings.py.j2#L470-L488 | 14:53 |
| nowster | # ls /openstack/venvs/horizon-19.0.0.0b1/lib/python2.7/site-packages/openstack_dashboard/conf | 14:53 |
| nowster | cinder_policy.d glance_policy.json neutron_policy.json nova_policy.json | 14:53 |
| nowster | cinder_policy.json keystone_policy.json nova_policy.d | 14:53 |
| cloudnull | looks like we dont normally set the policy map to specific names. | 14:54 |
| nowster | no heat_policy | 14:54 |
| cloudnull | ok. | 14:54 |
| cloudnull | anything in /etc/horizon || /etc/openstack-dashboard ? | 14:54 |
| nowster | # ls /etc/horizon/ | 14:55 |
| nowster | local_settings.py | 14:55 |
| nowster | no /etc/openstack-dashboard | 14:55 |
| jrosser | cloudnull: this is interesting http://git.openstack.org/cgit/openstack/openstack-ansible-os_horizon/tree/templates/horizon_local_settings.py.j2#n471 | 14:56 |
| jrosser | big fat warning about keeping horizon and keystone policy identical | 14:56 |
| cloudnull | yes I thought we had something that would reach-out and pull back policy files when set | 14:57 |
| jrosser | imho there are a few things folks have mentioned all pointing at this - guilhermesp & cinder, non-admin can't change passwords for me | 14:57 |
| cloudnull | however I could see that being broken given the how projects are pushing policy into code, instead of using the default file | 14:58 |
| cloudnull | I wonder if that means we're going to need to generate default + user defined policy for horizon? | 14:59 |
| cloudnull | in the absents of a default policy file ? | 14:59 |
| *** zul has quit IRC | 15:00 | |
| jrosser | i wonder what the horizon folks are expecting to happen.... | 15:00 |
| cloudnull | nowster if you add something like https://github.com/openstack/heat/blob/stable/pike/etc/heat/policy.json to your install does horizon being working as expected? | 15:01 |
| nowster | I'll try that now. | 15:01 |
| cloudnull | `wget https://raw.githubusercontent.com/openstack/heat/stable/pike/etc/heat/policy.json -O /openstack/venvs/horizon-19.0.0.0b1/lib/python2.7/site-packages/openstack_dashboard/conf/orchestration_policy.json` should put the file in place as I assume horizon would expect it | 15:02 |
| nowster | rats! wrong filename | 15:02 |
| cloudnull | that would need to be done on all of your horizon nodes. | 15:02 |
| nowster | I just have one at the moment | 15:03 |
| odyssey4me | I wonder how that's all supposed to work with policy-in-code - perhaps generate the default policy using the service binary, then copy it over? | 15:03 |
| jrosser | the master horizon docs still say "The implementation in Horizon is based on copies of policy files found in the service’s source code." | 15:03 |
| jrosser | i wonder if thats actually right or just historical | 15:04 |
| nowster | I think it should be called heat_policy.json | 15:05 |
| * cloudnull asked in the #openstack-horizon channel | 15:06 | |
| cloudnull | oh sorry nowster . thats my fault | 15:06 |
| nowster | That now doesn't like the file. | 15:07 |
| cloudnull | for folks not following in the horizon channel - https://pasted.tech/pastes/de55306f63f2c721913bd0b899d405a373a65b6b.raw | 15:07 |
| cloudnull | we'll see what they say :) | 15:07 |
| nowster | because it's got html all over it. | 15:07 |
| cloudnull | https://raw.githubusercontent.com/openstack/heat/stable/pike/etc/heat/policy.json ? | 15:08 |
| nowster | I forgot the raw | 15:08 |
| cloudnull | ah | 15:08 |
| cloudnull | :) | 15:08 |
| nowster | MUCH happier. | 15:09 |
| cloudnull | nice! | 15:10 |
| *** vnogin has quit IRC | 15:29 | |
| cloudnull | nowster have any other policy issues cropped up? | 15:31 |
| nowster | Not so far. | 15:32 |
| nowster | I'm doing preparations for bringing up an instance. Currently doing network topology. | 15:33 |
| prometheanfire | cloudnull: time to rebase :D | 15:41 |
| cloudnull | ? | 15:41 |
| guilhermesp | yeah, also, I'm experiencing issues with magnum today in rocky deployments. http://paste.openstack.org/show/742965/ | 15:42 |
| nowster | cloudnull: miscellaneous errors in the "Project / Orchestration" pages | 15:43 |
| prometheanfire | cloudnull: all my patches | 15:45 |
| prometheanfire | because of networkd (and mainly time) | 15:45 |
| nowster | and installing an image file doesn't seem to be working | 15:45 |
| nowster | comes up with an error and stays in state "queued" | 15:45 |
| *** asettle has joined #openstack-ansible | 15:46 | |
| asettle | spotz, there we go :) | 15:46 |
| odyssey4me | OH EM GEE | 15:47 |
| *** vnogin has joined #openstack-ansible | 15:47 | |
| odyssey4me | o/ asettle, welcome back to the rabble | 15:47 |
| spotz | hehe I wasn't going to let you not be here asettle! | 15:48 |
| nowster | urgh! No openstack cli installed by default. :( :( | 15:48 |
| *** udesale has quit IRC | 15:48 | |
| spotz | nowster: It should be on both the main host and in the utility container at least in an AIO | 15:49 |
| odyssey4me | nowster no, only where it's needed - but you'll find it in the utility container, and in the ansible runtime venv | 15:49 |
| cloudnull | ohai asettle! | 15:49 |
| spotz | I know for sure the utility container | 15:49 |
| nowster | they're in the utility container | 15:49 |
| nowster | Found it myself. Thanks. | 15:49 |
| spotz | :) | 15:49 |
| nowster | This isn't an AIO. My last deployment was. | 15:49 |
| odyssey4me | nowster in some cases, the service venv may also have it, but we're trying to slim down the requirements to only those necessary to help speed up the deployment | 15:50 |
| jrosser | nowster: are you using master? | 15:51 |
| nowster | yes | 15:51 |
| jrosser | that'll be unreleased versions of everything, if thats ok. is it for the aarch64 stuff? | 15:54 |
| nowster | yes | 15:54 |
| nowster | Having got the AIO working (to some extent) the next stage is to have a multiple host deployment | 15:54 |
| jrosser | it would be worth going through all the outsanding reviews and making sure they all have the same topic set, then we can look at them as a set | 15:54 |
| jrosser | i bunch of it may well class a bugs and could be backported, but seeing the whole set of patches together would help getting that all reviewed | 15:55 |
| nowster | One problem is that Cirros doesn't build from source any more. | 15:55 |
| jrosser | have you done anything with upstream cirros to get the aarch64 image fixed? | 15:56 |
| nowster | so I had to deconstruct the image and replace kernel and grub-uefi components | 15:56 |
| jrosser | it would be worth reaching out the cirros people to see if it can be fixed upstream, not having a working image available really hinders testing the patches | 15:58 |
| asettle | odyssey4me, cloudnull - oh hai folks! :D | 15:58 |
| nowster | jrosser: there doesn't seem to have been any activity on cirros for about 18 months | 15:59 |
| cloudnull | https://media1.tenor.com/images/f54932e6b9553a5538f31a5ddd78a9f3/tenor.gif?itemid=5039368 | 15:59 |
| *** pcaruana has quit IRC | 16:00 | |
| asettle | cloudnull, <3 | 16:01 |
| asettle | How's things back at the ranch? | 16:01 |
| chandan_kumar | odyssey4me: need some help here http://logs.openstack.org/82/627482/58/check/python-tempestconf-tempest-os-tempest/8b68cd5/ara-report/result/83c58a6c-b149-4f76-8b0f-833d5af21fbf/ | 16:01 |
| cloudnull | asettle https://twitter.com/cloudnull/status/1086291047389503488 - so the same | 16:01 |
| chandan_kumar | odyssey4me: what should i do here to always generate stackviz if tests fails or not https://github.com/openstack/openstack-ansible-os_tempest/blob/master/tasks/tempest_run.yml#L91 ? | 16:02 |
| prometheanfire | asettle: ohai | 16:03 |
| cloudnull | asettle how have you been? hows life on the outside? | 16:03 |
| chandan_kumar | cloudnull: https://review.openstack.org/#/c/631794/ thanks :-) | 16:04 |
| asettle | LOL cloudnull some things don't change aye ;) | 16:05 |
| asettle | prometheanfire, hey pal! | 16:05 |
| asettle | cloudnull, life on the outside is nice. Greener, actually ;) | 16:05 |
| asettle | (and she laughed, and laughed, and laughed) | 16:06 |
| cloudnull | so the chameleons are in fact greener on the other side!? | 16:06 |
| spotz | From what I understand life is good on the green side | 16:07 |
| asettle | cloudnull, they really, really hard. | 16:07 |
| asettle | are* | 16:07 |
| asettle | Wtf alex where did *hard* come from | 16:07 |
| asettle | spotz, so far, so good! | 16:08 |
| cloudnull | https://i.snag.gy/Cvjdzg.jpg -- from my desk | 16:09 |
| * cloudnull loves me some suse | 16:09 | |
| chandan_kumar | cloudnull: can I get one ? | 16:10 |
| * cloudnull only has this one | 16:11 | |
| cloudnull | asettle hows your lesser half? | 16:11 |
| cloudnull | tell him we miss him too | 16:11 |
| spotz | chandan_kumar: I'll pick you up one in Denver if I'm there | 16:11 |
| chandan_kumar | spotz: thanks :-) | 16:11 |
| asettle | cloudnull, CUTE | 16:12 |
| asettle | cloudnull, and he's good. Sometimes if you yell andymccr into the ether he appears. | 16:12 |
| asettle | I'm currently in Nuremberg though and he's at home so </3 | 16:12 |
| * cloudnull begins screaming "andymccr" | 16:13 | |
| spotz | asettle: How's the whether, I'll be next door in LUX in a few weeks | 16:13 |
| andymccr | o_O | 16:13 |
| spotz | I know mhayden does if you yell:) | 16:13 |
| spotz | hehehe | 16:13 |
| cloudnull | it worked! | 16:13 |
| * mhayden toots | 16:13 | |
| cloudnull | oh snap! | 16:13 |
| spotz | See! The old guard is here:) | 16:13 |
| cloudnull | its dad joke and meme time ! | 16:14 |
| cloudnull | hows it mhayden!? | 16:14 |
| mhayden | it's not bad | 16:14 |
| * mhayden is working on slides | 16:14 | |
| * cloudnull misses playing footsie with you | 16:15 | |
| *** chandan_kumar is now known as codemonster | 16:17 | |
| *** asettle has quit IRC | 16:20 | |
| *** vnogin has quit IRC | 16:22 | |
| cloudnull | asettle mhayden you both headed to devcon? | 16:25 |
| cloudnull | thats this week right? | 16:25 |
| mhayden | i'll be in Brnoooo | 16:25 |
| mhayden | i fly out Wednesday | 16:25 |
| cloudnull | ah | 16:25 |
| cloudnull | jrosser on https://review.openstack.org/#/c/631668/1/elk_metrics_6x/env.d/elk.yml -- maybe we use something like "physical_hosts"? | 16:26 |
| jrosser | cloudnull: perhaps i just don't follow why those things are not installed on "hosts" via the integration inventory | 16:29 |
| *** macza has joined #openstack-ansible | 16:30 | |
| cloudnull | ah, fair point. i'll just leave that as-is | 16:32 |
| *** zul has joined #openstack-ansible | 16:34 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: modify playbook inventory to make groups more specific https://review.openstack.org/631668 | 16:41 |
| openstackgerrit | Merged openstack/openstack-ansible-tests master: Setup clouds.yaml on tempest node https://review.openstack.org/631794 | 16:47 |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican master: Force force-tlsv12 only https://review.openstack.org/625634 | 16:50 |
| *** TxGirlGeek has joined #openstack-ansible | 16:54 | |
| *** hamzaachi has quit IRC | 16:54 | |
| *** hamzaachi has joined #openstack-ansible | 16:55 | |
| *** kopecmartin is now known as kopecmartin|off | 16:57 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_barbican stable/rocky: Enable overriding the service setup host python interpreter https://review.openstack.org/631231 | 16:58 |
| *** dcdamien has quit IRC | 16:59 | |
| *** DanyC has quit IRC | 17:00 | |
| *** hamzaachi has quit IRC | 17:08 | |
| *** codemonster is now known as chandankumar | 17:08 | |
| CeeMac | anyone know what would cause this error during task lxc_container_create: Gather container facts? | 17:15 |
| CeeMac | http://paste.openstack.org/show/742971/ | 17:15 |
| CeeMac | i'm not sure if it is an issue on the host, or somewhere else. After re-running the playbook, the containers have been deployed but without any IP settings, | 17:16 |
| CeeMac | which is an improvement over them not deploying intially. The other hosts containers are deployed without issue. | 17:17 |
| *** jbadiapa has quit IRC | 17:24 | |
| *** partlycloudy has joined #openstack-ansible | 17:27 | |
| partlycloudy | hi all, not mean to start a debate, what bare metal provisioning tools folks here prefer to use? | 17:29 |
| gundalow | We are excited to announce that AnsibleFest Atlanta will be held September 24-26, 2019. Mark your calendars now and sign up to be notified when our early, discounted registration opens in May ansible.com/ansiblefest | 17:30 |
| *** spatel has joined #openstack-ansible | 17:31 | |
| spatel | Folks :) | 17:31 |
| spatel | how are you doing | 17:31 |
| *** DanyC has joined #openstack-ansible | 17:40 | |
| *** DanyC has quit IRC | 17:44 | |
| *** hamzaachi has joined #openstack-ansible | 17:46 | |
| cloudnull | o/ spatel how are things ? | 17:46 |
| spatel | I am great! was very busy last few weeks, didn't get time to join IRC | 17:47 |
| spatel | currently we are running 446 instances (which include ~300 SR-IOV) | 17:48 |
| spatel | ~190 total compute nodes i have.. | 17:49 |
| spatel | Right now i am chasing down issue related Horizon (its really slow) i wonder may be because of ~400 instance... | 17:50 |
| *** TxGirlGeek has quit IRC | 17:50 | |
| spatel | Anyone has any experience with slowness with many instances? | 17:50 |
| jamesdenton | slowness is relative :) | 18:01 |
| jamesdenton | i would verify that memcached config is good (and working) if possible | 18:02 |
| openstackgerrit | Merged openstack/openstack-ansible master: Make container_interface optional for provider network https://review.openstack.org/630964 | 18:05 |
| openstackgerrit | Merged openstack/openstack-ansible master: Remove tempest service enable vars https://review.openstack.org/630660 | 18:10 |
| jrosser | spatel: agree with jamesdenton last time horizon got all bogged down for me it was insufficient max connections in memcached | 18:13 |
| spatel | insufficient max connections in memcached how do i check and where? | 18:14 |
| spatel | jrosser: | 18:14 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add openstack integration to skydive https://review.openstack.org/631858 | 18:14 |
| spatel | jamesdenton: I do have correct memcached setup and working.. | 18:15 |
| spatel | why don't horizon pull data from mysql instead of API :( | 18:15 |
| spatel | it takes 20 second to load instance/ tab (if you search that will more than 30s ) | 18:16 |
| cloudnull | spatel one area Id check on memcached is that the memory settings are large enough for the size of your environment. | 18:16 |
| cloudnull | if you see memcached running up against the memeory limit you might want to up the settings | 18:16 |
| spatel | where is that setting in memcached.. i am looking.. | 18:17 |
| spatel | also i found this error in logs but anyway its not related to performance http://paste.openstack.org/show/742980/ | 18:18 |
| spatel | it was case sensitive issue Roboto vs roboto | 18:18 |
| *** priteau has quit IRC | 18:18 | |
| jrosser | spatel: http://paste.openstack.org/show/742981/ | 18:20 |
| spatel | jrosser: Thanks this is my output http://paste.openstack.org/show/742982/ | 18:21 |
| spatel | Is memcached running in cluster? how they share cache between 3 nodes? | 18:21 |
| spatel | jrosser: your output has little different options, you don't have maxconns 4096 | 18:23 |
| *** priteau has joined #openstack-ansible | 18:23 | |
| jrosser | they are quite different arent they - probably different versions? | 18:24 |
| jrosser | i can see it set to 4096 in the config file anyway | 18:24 |
| jrosser | spatel: it may just be that horizon makes tons and tons of API calls :/ | 18:25 |
| spatel | do you seeing any issue in my memcached stats? | 18:26 |
| jrosser | well the only obvious thing is it doesnt seem to tell you how many connections you have | 18:27 |
| spatel | my log level is DEBUG in /etc/horizon/local_settings.py | 18:28 |
| *** priteau has quit IRC | 18:28 | |
| jrosser | but you could equally do netstat -n | grep 11211 | wc -l | 18:28 |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add openstack integration to skydive https://review.openstack.org/631858 | 18:29 |
| spatel | jrosser: 1337 | 18:29 |
| jrosser | spatel: cloudnull and i had a big adventure with elasticsearch and osprofiler https://twitter.com/thejrosser/status/1007718744297263104 | 18:30 |
| jrosser | but you really really want to try that out in a lab first, it's quite involved to get that going | 18:31 |
| spatel | interesting... | 18:31 |
| jrosser | the horizon integration is gross - i'll warn you now | 18:31 |
| spatel | hmm! i am not going to do that... :) | 18:32 |
| spatel | scared | 18:33 |
| spatel | why there are no alternative for Horizon? | 18:33 |
| *** hamzaachi has quit IRC | 18:36 | |
| jrosser | you could try similar operations with the cli, see if the times to list servers are similar | 18:37 |
| *** hamzaachi has joined #openstack-ansible | 18:37 | |
| spatel | This is cool memcache-top --instance=172.28.1.212 | 18:43 |
| spatel | it gives good stats | 18:43 |
| spatel | jrosser: I have noticed "nova list" and Horizon instance/ taking same time to list | 19:01 |
| *** zul has quit IRC | 19:03 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add openstack integration to skydive https://review.openstack.org/631858 | 19:04 |
| spatel | jrosser: is there an option to limit number of host horizon pulling? | 19:06 |
| spatel | it would be great if we tell only load 10 host | 19:06 |
| *** bgmccollum has quit IRC | 19:08 | |
| *** electrofelix has quit IRC | 19:13 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: Ensure public volume type option is evaluated as a boolean https://review.openstack.org/629815 | 19:15 |
| *** bgmccollum has joined #openstack-ansible | 19:27 | |
| *** bgmccollum has quit IRC | 19:31 | |
| spatel | https://bugs.launchpad.net/nova/+bug/1788149 | 19:33 |
| openstack | Launchpad bug 1788149 in OpenStack Compute (nova) "nova list too slow" [Undecided,Expired] | 19:33 |
| spatel | check this out... very interesting | 19:33 |
| openstackgerrit | Nicolas Bock proposed openstack/openstack-ansible master: Some small spelling fixes https://review.openstack.org/631877 | 19:45 |
| *** Darcidride_ has joined #openstack-ansible | 19:47 | |
| *** bgmccollum has joined #openstack-ansible | 19:48 | |
| *** bgmccollum has quit IRC | 20:01 | |
| broken_one | hello o/ | 20:09 |
| broken_one | may have found a bug with OSA neutron package. had an issue that is could not find the neutron-keepalived-state-change script | 20:12 |
| broken_one | this occured when creating an HA router | 20:12 |
| *** bgmccollum has joined #openstack-ansible | 20:19 | |
| *** bgmccollum has quit IRC | 20:20 | |
| jrosser | broken_one: it would be great if you could see if this is the fix https://review.openstack.org/#/c/630374/ | 20:21 |
| broken_one | so we look at the env path in the venv | 20:23 |
| broken_one | cd /opestack/venvs/neutron* | 20:23 |
| broken_one | source bin/activate | 20:24 |
| broken_one | and the env has the path to the vitrual venv | 20:24 |
| broken_one | and it still could not find that script | 20:24 |
| broken_one | i can test your fix on tues if that works | 20:25 |
| broken_one | for now we had to symlink the script to /usr/local/bin | 20:25 |
| *** hamzaachi has quit IRC | 20:25 | |
| *** vnogin has joined #openstack-ansible | 20:26 | |
| *** vnogin has quit IRC | 20:30 | |
| *** kopecmartin|off has quit IRC | 20:33 | |
| *** kopecmartin has joined #openstack-ansible | 20:36 | |
| *** bgmccollum has joined #openstack-ansible | 20:38 | |
| broken_one | VP approval for contributing to OSA..now we get to talk to the CEO, if that goes through we can develop with everyone on OSA. :D | 20:38 |
| *** dcdamien has joined #openstack-ansible | 20:40 | |
| mnaser | broken_one: that's awesome and super citingi | 20:41 |
| mnaser | broken_one: the symlink fix is one that we ran into, i think guilhermesp was working on pushing a fix there | 20:41 |
| broken_one | so we are now thinking this was our issue with all OVS deploy | 20:42 |
| broken_one | but we did take advice from the channel and seperated control plane into linuxbridges | 20:43 |
| broken_one | the team is excited about contributing | 20:43 |
| broken_one | mnaser: we will try jrosser's commit to see if that settles the issue | 20:44 |
| broken_one | now its time to work on our ext-net | 20:44 |
| jrosser | credit must go to jenglisch_ for that really - i see similar in my lab so will also try it out next week | 20:44 |
| broken_one | we think we found a bug there as well but we are going to do a lot more testing and "fixes" before we actually report anything here | 20:44 |
| broken_one | ah yes, sorry over looked jenglisch being the commiter | 20:46 |
| broken_one | i hope you all have a good long weekend. see ya on Tuesday | 20:46 |
| *** Darcidride_ has quit IRC | 21:03 | |
| mnaser | it's a long weekend? | 21:04 |
| mnaser | oh it is | 21:04 |
| *** strattao has joined #openstack-ansible | 21:10 | |
| *** radeks__ has quit IRC | 21:18 | |
| *** noonedeadpunk[h] has quit IRC | 21:19 | |
| *** aedc has quit IRC | 21:26 | |
| openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-ops master: Add openstack integration to skydive https://review.openstack.org/631858 | 21:30 |
| *** spatel has quit IRC | 21:47 | |
| *** priteau has joined #openstack-ansible | 21:56 | |
| *** priteau has quit IRC | 22:10 | |
| *** strattao has quit IRC | 22:48 | |
| *** strobelight has quit IRC | 22:57 | |
| *** strobelight has joined #openstack-ansible | 23:00 | |
| *** dcdamien has quit IRC | 23:37 | |
| *** tosky has quit IRC | 23:55 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!
