| *** logan2 has quit IRC | 00:01 | |
| *** logan2 has joined #openstack-ansible | 00:08 | |
| *** markvoelker has quit IRC | 00:32 | |
| *** openstackgerrit has quit IRC | 00:51 | |
| *** openstackgerrit has joined #openstack-ansible | 00:51 | |
| *** davidself has joined #openstack-ansible | 01:04 | |
| *** britthouser has quit IRC | 01:30 | |
| *** galstrom_zzz is now known as galstrom | 01:46 | |
| *** galstrom is now known as galstrom_zzz | 02:38 | |
| *** daneyon has joined #openstack-ansible | 02:46 | |
| *** britthouser has joined #openstack-ansible | 03:15 | |
| *** britthouser has quit IRC | 03:15 | |
| *** sdake_ has joined #openstack-ansible | 03:17 | |
| *** sdake_ has quit IRC | 03:22 | |
| *** daneyon has quit IRC | 03:25 | |
| *** sdake has quit IRC | 04:36 | |
| *** sdake has joined #openstack-ansible | 04:37 | |
| *** sdake has quit IRC | 05:08 | |
| *** sdake has joined #openstack-ansible | 05:13 | |
| *** sdake_ has joined #openstack-ansible | 05:21 | |
| *** shausy has joined #openstack-ansible | 05:33 | |
| *** sdake_ has quit IRC | 05:35 | |
| *** sdake_ has joined #openstack-ansible | 05:35 | |
| *** sdake_ has quit IRC | 05:48 | |
| *** sdake has quit IRC | 05:56 | |
| *** subha_ has joined #openstack-ansible | 06:30 | |
| subha_ | we have deployed suse openstack 5.0 . we have one controller , two compute and one admin server node. the instances hosted by the compute nodes are not getting any IP but in the dashboard ip is shown as 192.168.123.5 and 192.168.123.6. what could be the reason? please help. it is little urgent | 06:30 |
|---|---|---|
| subha_ | please let us know | 06:31 |
| subha_ | we have deployed suse openstack 5.0 . we have one controller , two compute and one admin server node. the instances hosted by the compute nodes are not getting any IP but in the dashboard ip is shown as 192.168.123.5 and 192.168.123.6. what could be the reason? please help. it is little urgent | 06:33 |
| *** cbaesema_ has joined #openstack-ansible | 06:51 | |
| *** cbaesema has quit IRC | 06:52 | |
| *** cbaesema_ is now known as cbaesema | 06:52 | |
| subha_ | we have deployed suse openstack 5.0 . we have one controller , two compute and one admin server node. the instances hosted by the compute nodes are not getting any IP but in the dashboard ip is shown as 192.168.123.5 and 192.168.123.6. what could be the reason? please help. it is little urgent | 06:53 |
| *** odyssey4me has joined #openstack-ansible | 08:22 | |
| *** odyssey4me has quit IRC | 08:42 | |
| *** persia has joined #openstack-ansible | 08:46 | |
| *** odyssey4me has joined #openstack-ansible | 08:50 | |
| hughsaunders | subha_: did you deploy using os-ansible-deployment? | 08:50 |
| hughsaunders | subha_: if not, you're probably better off asking in #openstack | 08:50 |
| *** odyssey4me_ has joined #openstack-ansible | 08:59 | |
| *** odyssey4me has quit IRC | 09:00 | |
| *** odyssey4me_ is now known as odyssey4me | 09:01 | |
| openstackgerrit | git-harry proposed stackforge/os-ansible-deployment: Restart mysql when config changed https://review.openstack.org/200054 | 09:04 |
| subha_ | i actually dont know what ansible deployment means..we are trying with suse cloud 5 | 09:12 |
| subha_ | and services we deployed through crowbar | 09:12 |
| odyssey4me | subha_ this channel is for those deploying openstack specifically with the Ansible deployment tool | 09:12 |
| odyssey4me | I would suggest that you ask your questions relating to your situation with SuSE/Chef/Crowbar in #openstack, or perhaps #openstack-chef | 09:13 |
| *** subha_ has left #openstack-ansible | 09:16 | |
| *** odyssey4me has quit IRC | 09:39 | |
| *** odyssey4me has joined #openstack-ansible | 09:40 | |
| *** fawadkhaliq has joined #openstack-ansible | 09:43 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: SSL support for haproxy https://review.openstack.org/198957 | 10:21 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Enable all services to use Keystone 'insecurely' https://review.openstack.org/201070 | 10:23 |
| *** fawadkhaliq has quit IRC | 10:38 | |
| *** fawadkhaliq has joined #openstack-ansible | 10:40 | |
| *** fangfenghua has joined #openstack-ansible | 10:44 | |
| odyssey4me | git-harry hughsaunders https://docs.ansible.com/playbooks_variables.html#variable-precedence-where-should-i-put-a-variable | 10:48 |
| *** lkoranda has quit IRC | 11:10 | |
| *** jaypipes has joined #openstack-ansible | 11:29 | |
| *** lkoranda has joined #openstack-ansible | 11:44 | |
| *** tlian has joined #openstack-ansible | 11:45 | |
| cloudnull | Morning. | 12:12 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: SSL support for haproxy https://review.openstack.org/198957 | 12:15 |
| odyssey4me | morning cloudnull | 12:15 |
| cloudnull | How you doing brother ? | 12:16 |
| odyssey4me | for the above-mentioned patch, is there a different/better way of including a list item based on a condition? | 12:16 |
| * cloudnull looking | 12:17 | |
| cloudnull | Odyssey4me where are you trying to do that ? | 12:19 |
| odyssey4me | cloudnull well, the only way I could make it work was to include a task from inside the role in the playbook and to add the haproxy_service_configs as vars for that poast_task | 12:20 |
| *** britthouser has joined #openstack-ansible | 12:22 | |
| odyssey4me | initially I had hoped to conditionally just pop the service list item onto the list in the vars/configs/haproxy_config.yml file | 12:22 |
| cloudnull | What you've got so far looks good. | 12:23 |
| odyssey4me | it works, I'm just wondering if there's a better way to achieve the same goal | 12:24 |
| cloudnull | One thing that we could consider is to move the service cnfigs out of the massive service list and in to a meta dependency that runs a "haproxy_client" role , or something similar. That would be a very clean , and ansible centric way of doing this. | 12:25 |
| cloudnull | I've fought haproxy as not being something that we want to use in prod but its unstoppable it seems. | 12:26 |
| odyssey4me | I think that os-ansible-deployment will have to develop more improved support over time, but for now we're still treating it as a dev/test tool only - with patches welcome. | 12:27 |
| odyssey4me | alright, considering that works and it nicely cleans up after itself if the vars change, I think it's good to go | 12:28 |
| cloudnull | Indeed, but soon I think we should consider spec / design something better than what we have. | 12:28 |
| cloudnull | But yes. What you have looks great. | 12:28 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: SSL support for haproxy https://review.openstack.org/198957 | 12:34 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: SSL support for haproxy https://review.openstack.org/198957 | 12:34 |
| *** fangfenghua has quit IRC | 12:48 | |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Add v3 calls for federation to keystone module https://review.openstack.org/197677 | 12:49 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: [WIP] Keystone IdP configuration https://review.openstack.org/194259 | 12:55 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: [WIP] Keystone SP configuration https://review.openstack.org/194395 | 12:55 |
| *** TheIntern has joined #openstack-ansible | 13:14 | |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: [WIP] Adjust the network rings are sync'd on https://review.openstack.org/201123 | 13:25 |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: [WIP] Keystone SP configuration https://review.openstack.org/194395 | 13:36 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Parameterize galera slow/unindexed logging options https://review.openstack.org/200728 | 13:37 |
| *** fawadkhaliq has quit IRC | 13:40 | |
| *** shausy has quit IRC | 13:46 | |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Add ability to target remote hosts for swift https://review.openstack.org/201136 | 13:46 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: [WIP] Add ability to target remote hosts for swift https://review.openstack.org/201136 | 13:47 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust swift_rings.py to work on specified regions https://review.openstack.org/200114 | 13:50 |
| *** alextricity has joined #openstack-ansible | 13:59 | |
| cloudnull | https://review.openstack.org/194259 https://review.openstack.org/197677 are still failing lint tests | 14:05 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Add configurable option [cinder]/cross_az_attach https://review.openstack.org/194213 | 14:06 |
| cloudnull | hughsaunders: looks mostly like flake8 failures | 14:06 |
| cloudnull | cc miguelgrinberg ^ | 14:07 |
| hughsaunders | cloudnull: yeah | 14:07 |
| hughsaunders | lint on the dependant patch is passing, so I should fix the v3 functions patch | 14:08 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Add role system-crontab-coordination https://review.openstack.org/201142 | 14:09 |
| *** Mudpuppy has joined #openstack-ansible | 14:09 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Fix scripts/run-upgrade.sh https://review.openstack.org/200597 | 14:12 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:14 | |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Add v3 calls for federation to keystone module https://review.openstack.org/197677 | 14:17 |
| hughsaunders | cloudnull: my editor whinges about flake8 errors on save for *.py but those libs don't have an extension :/ | 14:19 |
| palendae | hughsaunders: Nope :( have to use go based off shebangs | 14:19 |
| cloudnull | ^ | 14:19 |
| palendae | i can't sentence good | 14:19 |
| palendae | I've been thinking I should probably make a spec for splitting up the scripts-library.sh into "really needs root" and "doesn't need root" components | 14:20 |
| palendae | So we could run linting and docs builds locally without sudo | 14:20 |
| hughsaunders | I did think its strange that all those libs have shebang when they aren't going to be executed, but then there is a thing somewhere that greps for the shebang | 14:20 |
| cloudnull | there is, the modules we have are modeled after upstream . | 14:21 |
| palendae | hughsaunders: the gate-check-lint.sh does the shebang grep | 14:22 |
| hughsaunders | thats the one | 14:22 |
| cloudnull | IE https://github.com/ansible/ansible/blob/release1.6.9/library/files/file | 14:23 |
| palendae | Though since it sources the script-library.sh, it'll ask you for root locally | 14:23 |
| cloudnull | but since the great upstream module split they've added .py to the modules in core/extra but we've not done that. | 14:24 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Added in keystone reserved port https://review.openstack.org/196702 | 14:30 |
| *** markvoelker has joined #openstack-ansible | 14:33 | |
| odyssey4me | sigmavirus24 hughsaunders I'm getting http://paste.openstack.org/show/371462/ when applying https://review.openstack.org/196943 to an existing master build | 14:36 |
| sigmavirus24 | odyssey4me: yeah I'm running a build myself to test pit a solution | 14:39 |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: Upgrade the Keystone library to use v3 https://review.openstack.org/196943 | 14:40 |
| sigmavirus24 | odyssey4me: ^ should work but I haven't finished testing it | 14:40 |
| *** markvoelker_ has joined #openstack-ansible | 14:42 | |
| odyssey4me | sigmavirus24 yep, that does it | 14:43 |
| sigmavirus24 | oh good, I don't even need to test it =P | 14:44 |
| *** markvoelker has quit IRC | 14:44 | |
| *** galstrom_zzz is now known as galstrom | 15:04 | |
| *** daneyon has joined #openstack-ansible | 15:05 | |
| *** jwagner_away is now known as jwagner | 15:16 | |
| *** markvoelker has joined #openstack-ansible | 15:20 | |
| *** markvoelker_ has quit IRC | 15:22 | |
| *** markvoelker has quit IRC | 15:23 | |
| *** weezS has joined #openstack-ansible | 15:36 | |
| *** toddnni has quit IRC | 15:36 | |
| *** miguelgrinberg has quit IRC | 15:36 | |
| *** mordred has quit IRC | 15:36 | |
| *** mattt has quit IRC | 15:36 | |
| *** charz has quit IRC | 15:36 | |
| *** vincent_vdk has quit IRC | 15:36 | |
| *** hrvojem has quit IRC | 15:36 | |
| *** jcannava has quit IRC | 15:36 | |
| *** gregdek has quit IRC | 15:36 | |
| *** hrvojem has joined #openstack-ansible | 15:37 | |
| *** gregdek has joined #openstack-ansible | 15:37 | |
| *** vincent_vdk has joined #openstack-ansible | 15:37 | |
| *** mattt has joined #openstack-ansible | 15:37 | |
| *** jcannava has joined #openstack-ansible | 15:37 | |
| *** miguelgrinberg_ has joined #openstack-ansible | 15:37 | |
| *** toddnni has joined #openstack-ansible | 15:37 | |
| *** mordred has joined #openstack-ansible | 15:37 | |
| *** charz has joined #openstack-ansible | 15:37 | |
| *** daneyon_ has joined #openstack-ansible | 15:41 | |
| *** daneyon has quit IRC | 15:44 | |
| *** CheKoLyN has joined #openstack-ansible | 15:45 | |
| odyssey4me | cloudnull please review https://review.openstack.org/194255 today if you have a gap | 15:46 |
| * cloudnull reading now | 15:46 | |
| *** jwagner is now known as jwagner_away | 15:46 | |
| andymccr | cloudnull: you need to start trading those for beers :D | 15:47 |
| cloudnull | +1 | 15:48 |
| cloudnull | will review for beer | 15:48 |
| *** fawadkhaliq has joined #openstack-ansible | 15:50 | |
| palendae | Doing upgrade from 10.1.10 to current master, saw this: http://paste.openstack.org/show/371639/ anyone seen an i18n import error with that before? | 15:56 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Wait until mongo responds after restart https://review.openstack.org/200252 | 15:58 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Adding galera_root_password token during upgrade Closes-Bug: #1473548 https://review.openstack.org/200721 | 16:00 |
| openstack | bug 1473548 in openstack-ansible kilo "Missing variables from Juno to Kilo upgrade" [Undecided,New] https://launchpad.net/bugs/1473548 | 16:00 |
| *** tlian has quit IRC | 16:02 | |
| palendae | Hm, yeah - i18n is not installed on Keystone containers after upgrade | 16:03 |
| palendae | Weird | 16:03 |
| palendae | oslo.i18n that is | 16:03 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Updated master for new dev work - 07.07.2015 https://review.openstack.org/199126 | 16:04 |
| cloudnull | palendae: odd, i wonder if its not being set as a dep ? | 16:04 |
| cloudnull | the wheel is there http://rpc-repo.rackspace.com/os-releases/11.1.0/oslo.i18n-1.5.0-py2.py3-none-any.whl | 16:04 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: SSL support for haproxy https://review.openstack.org/198957 | 16:05 |
| palendae | cloudnull: Yeah, I'm looking | 16:05 |
| cloudnull | if we need to we can add it to the play as a pip package to install. | 16:06 |
| palendae | Evidently it's only upgrades; I'm assuming Fernet tokens are being made successfully in the gate | 16:06 |
| galstrom | cloudnull: http://lists.openstack.org/pipermail/openstack-dev/2015-July/068859.html LinuxBridge CI Status | 16:07 |
| cloudnull | nice! | 16:08 |
| palendae | cloudnull: I added it to keystone_pip_packages and I'm re-running, will see if that fixes | 16:08 |
| *** alop has joined #openstack-ansible | 16:08 | |
| palendae | Apsu: ^^^ | 16:08 |
| palendae | “I’d like to report that with the merge of a | 16:08 |
| palendae | couple patches to fix race conditions[1][2] “ | 16:08 |
| palendae | galstrom: Hell yeah! | 16:09 |
| palendae | Apsu, specifically https://review.openstack.org/#/c/197162/ sounds kinda sorta relevant | 16:10 |
| sigmavirus24 | mancdaz: cloudnull http://lists.openstack.org/pipermail/openstack-dev/2015-July/069314.html | 16:11 |
| *** sdake has joined #openstack-ansible | 16:14 | |
| *** miguelgrinberg_ is now known as miguelgrinberg | 16:16 | |
| Apsu | palendae: Huh, neat | 16:16 |
| palendae | Turns out things get fixed when people look at them | 16:17 |
| Apsu | How about that | 16:17 |
| galstrom | palendae: that is yall pushing the community.. well done! | 16:22 |
| palendae | galstrom: Sure, yeah, I'll take credit for that! | 16:22 |
| cloudnull | lol +1 palendae for prez! | 16:23 |
| palendae | That is the key 2016 issue | 16:23 |
| palendae | LinuxBridge in Neutron | 16:23 |
| cloudnull | Sam-I-Am: cc | 16:23 |
| * palendae convenes special session of Senate | 16:23 | |
| *** sdake has quit IRC | 16:23 | |
| *** sdake has joined #openstack-ansible | 16:24 | |
| Sam-I-Am | cloudnull: wat | 16:25 |
| palendae | Sam-I-Am: http://lists.openstack.org/pipermail/openstack-dev/2015-July/068859.html | 16:25 |
| Sam-I-Am | palendae: yeah | 16:26 |
| Sam-I-Am | palendae: so i've been working with sean collins on fixing linuxbridge gating | 16:26 |
| palendae | Yeah, ^ is the real winner :) | 16:27 |
| Sam-I-Am | how to fix neutron without contributing code | 16:27 |
| Sam-I-Am | except docs, that is | 16:28 |
| palendae | Bug the shit out of a core? | 16:28 |
| Sam-I-Am | more like showing the neutron devs how their stuff works | 16:28 |
| Sam-I-Am | linuxbridge was sort of an unknown | 16:28 |
| *** jwagner_away is now known as jwagner | 16:37 | |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Implement Ceilometer https://review.openstack.org/201244 | 16:39 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Wait until mongo responds after restart https://review.openstack.org/201245 | 16:39 |
| *** TheIntern has quit IRC | 16:40 | |
| *** mattt has quit IRC | 16:49 | |
| *** mattt has joined #openstack-ansible | 16:49 | |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: Add v3 calls for federation to keystone module https://review.openstack.org/197677 | 16:59 |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: [WIP] Keystone IdP configuration https://review.openstack.org/194259 | 16:59 |
| openstackgerrit | Ian Cordasco proposed stackforge/os-ansible-deployment: [WIP] Keystone SP configuration https://review.openstack.org/194395 | 17:00 |
| *** annashen has joined #openstack-ansible | 17:01 | |
| *** annashen has quit IRC | 17:12 | |
| *** TheIntern has joined #openstack-ansible | 17:13 | |
| *** annashen has joined #openstack-ansible | 17:20 | |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Adjust swift_rings.py to work on specified regions https://review.openstack.org/200114 | 17:36 |
| *** fawadkhaliq has quit IRC | 17:52 | |
| *** abitha has joined #openstack-ansible | 18:01 | |
| Verilium | Is anyone familiar with Openstack's Neutron/DVR setup? Specifically, if I have compute nodes running in different datacenters, how do I go about for the tunnel network? Do all the nodes need to talk to each other, still? VPN to connect the tunnel network? Or do they only need to talk with a local network node? | 18:27 |
| Sam-I-Am | Verilium: osad does not support dvr | 18:27 |
| *** miguelgrinberg has quit IRC | 18:28 | |
| Verilium | Oh my, sorry, wrong channel. | 18:28 |
| Verilium | Sam-I-Am: Thx. Good to know. :) | 18:28 |
| stevelle | :) | 18:28 |
| Verilium | Sam-I-Am: I guess/hope it doesn't support dvr "yet". :P | 18:28 |
| Sam-I-Am | Verilium: for tunnel networks in general, its a big mesh among all of the compute and network nodes | 18:28 |
| *** miguelgrinberg has joined #openstack-ansible | 18:29 | |
| Sam-I-Am | Verilium: only thing required is IP endpoints, so layer 3 between nodes | 18:29 |
| *** miguelgrinberg has quit IRC | 18:29 | |
| Verilium | Sam-I-Am: Hmm... Suppose I was in normal neutron/non-dvr setup, I'd need all the compute/network nodes to be able to talk to each other, I guess. | 18:29 |
| Verilium | Yeah, ok, hmm. | 18:30 |
| Sam-I-Am | Verilium: dvr requires ovs, and osad currently only supports linux bridge | 18:30 |
| Verilium | So as long as the routing and such works out... | 18:30 |
| * Verilium nods. | 18:30 | |
| *** miguelgrinberg has joined #openstack-ansible | 18:30 | |
| *** sdake has quit IRC | 18:41 | |
| *** sdake has joined #openstack-ansible | 18:44 | |
| *** metral is now known as metral_zzz | 18:48 | |
| palendae | cloudnull: Huh, even adding oslo.i18n to the deps, keystone plays are failing. No client now | 18:53 |
| palendae | Also tried master greenfield install - it's fine | 18:53 |
| palendae | Will try 10.1.10 -> head of kilo | 18:53 |
| cloudnull | so this is an upgrade from 10.1.x to 11.0.4 ? | 18:57 |
| palendae | cloudnull: 10.1.10 to head of master | 18:58 |
| cloudnull | is the package absent ? | 18:58 |
| cloudnull | or just the wrong version ? | 18:58 |
| palendae | Checking | 18:59 |
| palendae | oslo.i18n it was absent | 18:59 |
| cloudnull | and with it specifically defined it barfed elsewhere ? | 19:01 |
| palendae | Yeah, on pip install keystone. Just hopped into the container and it's on WebOb specifically. In the repo, I have 1.4.1 for master, keystone wants >=1.2.3, <=1.3.1 | 19:03 |
| palendae | Might be a branch issue | 19:03 |
| palendae | Container's looking at os-releases/master | 19:03 |
| palendae | 11.0.4 on my repo container has WebOb 1.4.4 too | 19:04 |
| palendae | Same with os-releases/kilo | 19:04 |
| palendae | keystone==2014.2.4.dev5 is on my keystone container, but is keystone-8.0.0.0b2.dev35-py2-none-any.whl in os-releases/master | 19:05 |
| palendae | Oh right, cause 2014.2.4.dev5 would be right for 10.1.10 | 19:06 |
| palendae | Gonna try telling the keystone role's pip stuff to do 'latest' for the packages instead of 'present' | 19:08 |
| palendae | Nope, still failed | 19:09 |
| palendae | But if i pip uninstall keystone then pip install it, it's fine | 19:09 |
| palendae | wth | 19:09 |
| palendae | I'll try 10.1.10 to head of kilo before I file a bug. afaik, we don't officially support upgrades to master directly | 19:12 |
| *** annashen has quit IRC | 19:20 | |
| *** persia has quit IRC | 19:21 | |
| *** annashen has joined #openstack-ansible | 19:21 | |
| *** persia has joined #openstack-ansible | 19:23 | |
| *** annashen has quit IRC | 19:23 | |
| *** annashen has joined #openstack-ansible | 19:23 | |
| sigmavirus24 | palendae: so pip install -U keystone fails? | 19:23 |
| palendae | Yes | 19:24 |
| palendae | On the webob issue | 19:24 |
| palendae | Looks like I might be running into similar on glance now | 19:26 |
| *** metral_zzz is now known as metral | 19:27 | |
| *** jwagner is now known as jwagner_away | 19:29 | |
| palendae | sigmavirus24: Yeah... http://paste.openstack.org/show/372090/ | 19:30 |
| sigmavirus24 | palendae: what's `pip freeze` give you? | 19:31 |
| palendae | All of it or glance? | 19:32 |
| palendae | http://paste.openstack.org/show/372091/ glance stuff | 19:32 |
| sigmavirus24 | sorry wanted all of it | 19:35 |
| palendae | One moment | 19:35 |
| palendae | sigmavirus24: http://paste.openstack.org/show/372092/ | 19:38 |
| sigmavirus24 | palendae: that's really bizarre | 19:43 |
| sigmavirus24 | oh | 19:44 |
| sigmavirus24 | wait | 19:44 |
| sigmavirus24 | wait | 19:44 |
| sigmavirus24 | palendae: you're targetting master not kilo right? | 19:44 |
| sigmavirus24 | master has versions without epochs | 19:44 |
| sigmavirus24 | palendae: e.g., http://lists.openstack.org/pipermail/openstack-dev/2015-July/069314.html | 19:44 |
| sigmavirus24 | that's why it won't work going to master | 19:44 |
| sigmavirus24 | If the kilo branch isn't properly targetting kilo SHAs then that could be the same reason | 19:44 |
| sigmavirus24 | Kilo should be 2015.1.{something}.dev{something} | 19:45 |
| sigmavirus24 | and that upgrade will work | 19:45 |
| palendae | sigmavirus24: Yeah, I'm using maser | 19:45 |
| palendae | master* | 19:45 |
| sigmavirus24 | palendae: but to liberty it won't work | 19:45 |
| *** persia has quit IRC | 19:45 | |
| palendae | Ok, I'll give kilo a shot | 19:45 |
| sigmavirus24 | palendae: because upstream is being a bit obstinate about versioning | 19:45 |
| palendae | I'm not sure if master was supported at all or not | 19:45 |
| sigmavirus24 | And I'm about to finish an email about that | 19:45 |
| sigmavirus24 | palendae: it may have been but can't be any more | 19:45 |
| palendae | Though my kilo versions are looking the same on the repo | 19:45 |
| sigmavirus24 | Not since they changed the versioning scheme | 19:45 |
| sigmavirus24 | palendae: if kilo is building not as 2015.1.x we're using the wrong SHAs on kilo and that's a sever bug | 19:46 |
| sigmavirus24 | *severe | 19:46 |
| palendae | Yeah, 1 moment | 19:46 |
| palendae | Let me confirm | 19:46 |
| palendae | keystone-2015.1.1.dev13-py2-none-any.whl | 19:47 |
| palendae | glance-2015.1.1.dev13-py2-none-any.whl | 19:47 |
| palendae | So kilo's looking right | 19:47 |
| palendae | sigmavirus24: For the sake of time, I'm going to just wipe that install and try 10.1.10 to kilo's head | 19:50 |
| sigmavirus24 | palendae: good | 19:52 |
| sigmavirus24 | sorry | 19:52 |
| sigmavirus24 | I should have realized that faster | 19:52 |
| sigmavirus24 | Since I was literally just writing an email about that | 19:52 |
| palendae | sigmavirus24: Not your fault. Thanks for keeping up with the ML | 19:52 |
| palendae | My main concern is making sure we have a kilo release that upgrades smoothly for downstream (rpc-openstack) right now | 19:52 |
| sigmavirus24 | palendae: yeah I get that | 19:53 |
| alextricity | where is our frozen repo? | 19:53 |
| sigmavirus24 | palendae: just use the kilo branch for a better time | 19:53 |
| sigmavirus24 | alextricity: in antartica | 19:53 |
| sigmavirus24 | sorry, that wasn't helpful | 19:53 |
| palendae | sigmavirus24: Yeah | 19:53 |
| alextricity | :( | 19:53 |
| sigmavirus24 | alextricity: rpc-repo.cloudnull.io or something like that | 19:53 |
| sigmavirus24 | cloudnull: knows | 19:54 |
| *** jwagner_away is now known as jwagner | 19:55 | |
| cloudnull | what ? | 19:56 |
| sigmavirus24 | cloudnull: where are the frozen repos? | 19:59 |
| palendae | s/frozen/canonical/ | 19:59 |
| cloudnull | https://rpc-repo.rackspace.com/ | 19:59 |
| sigmavirus24 | arg, so close but yet so far =P | 20:01 |
| cloudnull | lol | 20:01 |
| *** auld has quit IRC | 20:02 | |
| *** persia has joined #openstack-ansible | 20:07 | |
| *** persia has quit IRC | 20:07 | |
| *** persia has joined #openstack-ansible | 20:07 | |
| *** metral is now known as metral_zzz | 20:22 | |
| alextricity | Hey, what was the reason behind https://github.com/stackforge/os-ansible-deployment/blob/7d2a0231d8ae29f43150a083ea6aeb19f8ac814c/playbooks/roles/openstack_hosts/defaults/main.yml#L94 | 20:25 |
| alextricity | Does anybody know why vm.swappiness was set to 5? | 20:25 |
| cloudnull | so it has access to swap but does not use a lot of it. | 20:25 |
| alextricity | :) | 20:26 |
| cloudnull | for why ? | 20:27 |
| *** auld has joined #openstack-ansible | 20:28 | |
| alextricity | customer was wondering | 20:28 |
| *** metral_zzz is now known as metral | 20:40 | |
| *** metral is now known as metral_zzz | 20:41 | |
| *** annashen has quit IRC | 20:41 | |
| *** abitha has quit IRC | 20:44 | |
| *** galstrom is now known as galstrom_zzz | 20:51 | |
| *** annashen has joined #openstack-ansible | 21:01 | |
| *** jwagner is now known as jwagner[away] | 21:06 | |
| *** jwagner[away] is now known as jwagner | 21:08 | |
| *** metral_zzz is now known as metral | 21:13 | |
| *** jwagner is now known as jwagner_away | 21:15 | |
| palendae | sigmavirus24: 10.1.10 -> kilo head has installed keystone and glance successfully | 21:25 |
| sigmavirus24 | palendae: yeah | 21:25 |
| sigmavirus24 | here's hoping downstream doesn't try to shit over all proper use of version numbers for python packages | 21:25 |
| sigmavirus24 | Looks like I have Josh Harlow (Yahoo) on my side at least + Donald | 21:26 |
| *** metral is now known as metral_zzz | 21:41 | |
| *** Mudpuppy has quit IRC | 22:08 | |
| klindgren_ | sigmavirus24, it already has - imho | 22:15 |
| sigmavirus24 | klindgren_: sorry? | 22:16 |
| klindgren_ | it doesn't support greater than 3 versions x.y.z tryign to do x.y.z.w where w == a local branch where an operator has made change and would like ot track those == broken in newer versions of pbr | 22:16 |
| klindgren_ | here's hoping downstream doesn't try to shit over all proper use of version numbers for python packages | 22:16 |
| *** britthouser has quit IRC | 22:18 | |
| sigmavirus24 | klindgren_: pbr enforces use of semantic versioning | 22:21 |
| sigmavirus24 | epochs don't figure into that | 22:21 |
| sigmavirus24 | klindgren_: pbr != downstream. Downstream is debian, ubuntu, red hat (fedora, centos, etc.), and the rest of the people who turn the python packages into distro packages | 22:22 |
| sigmavirus24 | they have their own versioning schemes that upstream is trying to fit into inspite of its harm to people using the raw python packages | 22:23 |
| *** auld has quit IRC | 22:27 | |
| *** markvoelker has joined #openstack-ansible | 22:27 | |
| *** markvoelker has quit IRC | 22:27 | |
| *** markvoelker has joined #openstack-ansible | 22:27 | |
| klindgren_ | I assume this was the discussion about setting epoch in setup.cfg | 22:27 |
| sigmavirus24 | klindgren_: yep | 22:29 |
| klindgren_ | if thats the case I htink I mgiht have misstated I work with josh on the anvil project and we are going through some "pbr" hell right now. Some of it is specifically dealing with the the down streaming naming and some of it is specifically dealing how both yahoo & GoDaddy name their packages that we build which are basedupon a stable release | 22:29 |
| klindgren_ | and how PBR totally f's that all to hell | 22:29 |
| klindgren_ | and the pbr is some weird mix between semver and and pep440 but is only partially pep440 complaint. | 22:30 |
| palendae | Python Build Ridiculousness | 22:30 |
| *** metral_zzz is now known as metral | 22:30 | |
| klindgren_ | +1million | 22:31 |
| klindgren_ | we asked for a specific feature to get put into to pbr to bascially set the version to whatever the hell I tell you the version name actually is. | 22:31 |
| palendae | Isn't that like half of pbr's code? | 22:32 |
| palendae | Calculating verions, taht is | 22:32 |
| klindgren_ | i thought half of pbr code was trying to autiomatically figure out what version name based upon hte git tag + some number of commits since + the git hash or some stupid crazyness | 22:32 |
| palendae | Right | 22:32 |
| palendae | I'm all for deleting that half | 22:33 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: [WIP] Adjust the network rings are sync'd on https://review.openstack.org/201123 | 22:33 |
| palendae | And probably the rest | 22:33 |
| klindgren_ | but in the ned from an end user perspective using pbr to try to get a valid name of a built version - sometimes I just want the name to be the freaking tag I gave you. IE 2015.1.0.3 should not come out 2015.1.0.dev3 | 22:33 |
| *** metral is now known as metral_zzz | 22:34 | |
| *** harlowja_ has joined #openstack-ansible | 22:40 | |
| klindgren_ | anyway - jsut read the last part of that convo - I mssed the bit about Ubuntu and !. I totally agree about producing python packages and "downstream" needs to modifiy their stuff to deal with that | 22:42 |
| sigmavirus24 | klindgren_: lol that's not even the argument I'm being given in -infra. It's because you have to quote things on the shell that have a ! in it | 22:44 |
| sigmavirus24 | So anyone dealing with our tarballs have to escape/quote it | 22:44 |
| harlowja_ | sigmavirus24 :-/ | 22:44 |
| harlowja_ | life is tough | 22:44 |
| harlowja_ | lol | 22:44 |
| sigmavirus24 | And I'm just like "Whelp, guess you've never used tab completion on a shell before" | 22:45 |
| sigmavirus24 | Not even goign to go there | 22:45 |
| harlowja_ | :) | 22:45 |
| harlowja_ | i once had to drink water from a stream, after i filtered it | 22:45 |
| harlowja_ | first-world-problem | 22:45 |
| * harlowja_ i could have died!!! | 22:46 | |
| harlowja_ | i mean http://images.mentalfloss.com/sites/default/files/styles/article_640x430/public/oregon-trail-dysentery_5.jpg could have happend man | 22:46 |
| sigmavirus24 | harlowja_: seriously | 22:46 |
| sigmavirus24 | I drink water from a well | 22:46 |
| harlowja_ | omg! | 22:46 |
| harlowja_ | lol | 22:46 |
| sigmavirus24 | what if i bring the bucket up and there's no water | 22:46 |
| sigmavirus24 | like | 22:46 |
| sigmavirus24 | omgwtfbbq | 22:47 |
| harlowja_ | lol | 22:47 |
| *** sdake has quit IRC | 22:48 | |
| harlowja_ | sigmavirus24 good luck, tapping out of that, lol | 22:49 |
| sigmavirus24 | harlowja_: I'll have to tap out shortly | 22:50 |
| sigmavirus24 | This is mostly why I was hoping to have this on the ML | 22:50 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Add ability to target remote hosts for swift https://review.openstack.org/201136 | 22:50 |
| harlowja_ | sigmavirus24 if u need me to help u boil some water, let me know, lol | 22:50 |
| *** metral_zzz is now known as metral | 22:50 | |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: [WIP] Add ability to target remote hosts for swift https://review.openstack.org/201136 | 22:50 |
| sigmavirus24 | harlowja_: that'd be greeeeat | 22:51 |
| klindgren_ | sigmavirus24, omg - maybe they should have through about that before they decided to invalidate 4+ years of version numbers. | 22:51 |
| sigmavirus24 | andymccr: go to bed | 22:51 |
| sigmavirus24 | klindgren_: hah | 22:51 |
| harlowja_ | klindgren_ impassible! | 22:51 |
| sigmavirus24 | klindgren_: btw, this channel /is/ logged fwiw | 22:51 |
| harlowja_ | why'd u guys do that, lol | 22:52 |
| *** auld has joined #openstack-ansible | 22:52 | |
| klindgren_ | sigmavirus24, thats fine - I am not saying anything that I haven't said before. I may have the "unpopular" opinion but its justifiable in this case | 22:52 |
| andymccr | sigmavirus24: http://s-ak.buzzfed.com/static/imagebuzz/web02/2010/6/23/21/i-do-what-i-want-22875-1277341571-14.jpg | 22:54 |
| sigmavirus24 | andymccr: nsfw nsfw | 22:55 |
| sigmavirus24 | =P | 22:55 |
| harlowja_ | that may or may not be klindgren_ | 22:55 |
| harlowja_ | lol | 22:56 |
| sigmavirus24 | harlowja_: why'd we turn logging on? because we're awesome | 22:56 |
| andymccr | depends where you work :D | 22:56 |
| harlowja_ | :) | 22:56 |
| * sigmavirus24 is out for real this time | 22:56 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:57 | |
| *** sdake has joined #openstack-ansible | 22:57 | |
| *** britthouser has joined #openstack-ansible | 22:57 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: SSL support for haproxy https://review.openstack.org/198957 | 22:58 |
| *** sdake_ has joined #openstack-ansible | 23:00 | |
| *** TheIntern has quit IRC | 23:00 | |
| *** weezS has quit IRC | 23:04 | |
| *** sdake has quit IRC | 23:04 | |
| *** CheKoLyN has quit IRC | 23:11 | |
| *** metral is now known as metral_zzz | 23:12 | |
| *** darrenc is now known as darrenc_afk | 23:19 | |
| *** sdake has joined #openstack-ansible | 23:37 | |
| palendae | cloudnull: You around? | 23:39 |
| cloudnull | yup | 23:39 |
| cloudnull | palendae: whats up ? | 23:40 |
| palendae | Question: the temp plays in run-upgrade.sh don't have access to the old rpc_user_variables.yml values, do they? | 23:40 |
| palendae | Trying to remove the old http galera/mariadb repo and replace it with the new https version | 23:41 |
| cloudnull | the old vars are available. | 23:41 |
| *** sdake_ has quit IRC | 23:41 | |
| cloudnull | the backup dir is rpc_deploy.OLD i think | 23:41 |
| palendae | So you'd point at that with -e? | 23:41 |
| cloudnull | you could, but you'd likely created all kinds of name collisions. | 23:42 |
| palendae | I don't see that used anywhere else in the upgrade script, so I'm thinking it's a bad idea | 23:42 |
| palendae | Makes sense | 23:42 |
| palendae | I'd rather not reach in and delete the file | 23:42 |
| palendae | Actually let me try something | 23:42 |
| cloudnull | we use the old vars but within python directly | 23:42 |
| *** darrenc_afk is now known as darrenc | 23:42 | |
| palendae | Yeah, I see that | 23:43 |
| palendae | Oh, nope | 23:43 |
| palendae | Damn | 23:43 |
| palendae | Was hoping with http://docs.ansible.com/apt_repository_module.html I could force overwrite of that file | 23:43 |
| palendae | Hm | 23:44 |
| palendae | Any option here feels dirty | 23:45 |
| palendae | Copy the repo's URL directly into run-upgrade.sh, use a dirty python script to read the value out and return it, hack together a way to get it into a temp play | 23:45 |
| palendae | Basically trying to address https://bugs.launchpad.net/openstack-ansible/+bug/1473525 | 23:46 |
| openstack | Launchpad bug 1473525 in openstack-ansible trunk " Duplicate sources.list entry for mariadb during kilo upgrade" [Low,In progress] - Assigned to Bjoern Teipel (bjoern-teipel) | 23:46 |
| cloudnull | you could do `find /etc/apt/sources.list.d/ -name '*[Mm][Aa][Rr][Ii][Aa]*' -exec rm {} \; ` and that would simply remove whatever is there. | 23:47 |
| cloudnull | and then you would let the playbooks put back whatever is supposed to be there. | 23:47 |
| palendae | lol | 23:47 |
| palendae | Yeah, that would work | 23:47 |
| palendae | k, I'll try that and run it | 23:48 |
| cloudnull | its dirty but should force whatever mariadb repo is there to be gone. | 23:48 |
| palendae | Right | 23:48 |
| palendae | Like I said, I think any solution here is dirty | 23:48 |
| palendae | Without clean access to the old vars | 23:48 |
| cloudnull | ++ | 23:48 |
| palendae | Or a way to tell apt-repository to remove the file | 23:49 |
| palendae | But I see why it can't, it works directly with a string | 23:49 |
| palendae | apt-add-repository, even | 23:49 |
| cloudnull | yea | 23:49 |
| *** metral_zzz is now known as metral | 23:53 | |
| *** annashen has quit IRC | 23:54 | |
| *** metral is now known as metral_zzz | 23:55 | |
| *** metral_zzz is now known as metral | 23:55 | |
| palendae | cloudnull: https://github.com/nrb/os-ansible-deployment/commit/f72a2ac0dc69e3dd165db85f3a12b8ce6f0abdf4, I'll run it tonight and see what happens before I push it to gerritt | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
