| *** gtt116_ has joined #openstack-ansible | 00:05 | |
| *** gtt116__ has quit IRC | 00:08 | |
| bgmccollum | i see in Kilo the lxc package isn't getting pinned, but it is in Juno. Should this pin be removed as part of the upgrade? | 00:24 |
|---|---|---|
| *** mss_ has quit IRC | 00:24 | |
| *** markvoelker has quit IRC | 00:45 | |
| *** sdake has quit IRC | 00:53 | |
| *** ctina_ has joined #openstack-ansible | 00:55 | |
| *** eil397 has quit IRC | 00:59 | |
| *** ctina_ has quit IRC | 01:00 | |
| cloudnull | bgmccollum: yes/no I believe the pin is in group_vars for Juno which will be removed when you change the playbooks | 01:00 |
| cloudnull | https://github.com/openstack/openstack-ansible/blob/juno/rpc_deployment/inventory/group_vars/all.yml#L76-L81 | 01:00 |
| cloudnull | if the pin is in user_vars then its up to the deployer if it should be removed | 01:01 |
| cloudnull | bgmccollum: RE: https://review.openstack.org/#/c/249209/ yes to adding openrc_ in front of the various variables. | 01:06 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Ability to define the mtu globally (for all the container networks) https://review.openstack.org/248064 | 01:09 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Use Ansible AIO playbook for run-aio-build.sh https://review.openstack.org/244708 | 01:10 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Added to ability to set mtu for lxcbr0 https://review.openstack.org/248061 | 01:11 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Added in keystone reserved port https://review.openstack.org/196702 | 01:16 |
| jamielennox | cloudnull: so is there any intention to publish some of these modules individually? | 01:16 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Update Master SHA's - 01.12.2015 https://review.openstack.org/245966 | 01:17 |
| jamielennox | i was trying to use os_keystone for some development stuff and isolate it from the whole openstack-ansible deployment | 01:17 |
| cloudnull | jamielennox: absolutly. | 01:17 |
| cloudnull | we're in the process of doing that right now | 01:17 |
| cloudnull | so far I've done the host only roles. | 01:17 |
| jamielennox | but things are very intertwined, turned out to be impossible without also including rabbit, memcache and something failed in there that i gave up on | 01:17 |
| cloudnull | https://github.com/openstack/openstack-ansible-specs/blob/master/specs/mitaka/independent-role-repositories.rst | 01:18 |
| cloudnull | thats the spec for it | 01:18 |
| cloudnull | so we're breaking it all out as independent roles and then creating gate tests for the role as a stand alone role | 01:18 |
| cloudnull | Example: https://github.com/openstack/openstack-ansible-lxc_container_create | 01:19 |
| jamielennox | cloudnull: are you expecting to keep the role dependencies? | 01:19 |
| jamielennox | (hoping for no) | 01:19 |
| cloudnull | which ones? in the meta? | 01:20 |
| cloudnull | or all of the galera, rabbit bits and such ? | 01:20 |
| *** mss has joined #openstack-ansible | 01:20 | |
| jamielennox | cloudnull: let me find some examples | 01:21 |
| cloudnull | RE: these deps https://github.com/openstack/openstack-ansible/blob/master/playbooks/roles/os_keystone/meta/main.yml#L32-L40 ? | 01:21 |
| jamielennox | cloudnull: some of those | 01:23 |
| cloudnull | yes and no. things like pip_lock_down, no -- things like memcahed when the token provider is a memcached driver, probably. | 01:24 |
| jamielennox | there's a few things like groups['rabbitmq_all'] which just fail if you don't set things up specifically | 01:24 |
| cloudnull | but if there's things that you'd like to see in that role as we break it out im game to work that in | 01:24 |
| cloudnull | ah yes those will be out. | 01:25 |
| *** alkari has quit IRC | 01:25 | |
| cloudnull | the plan so far is to break out the "crud" ops from the os_* roles | 01:25 |
| cloudnull | likely in its own role or in an optional running section. | 01:26 |
| jamielennox | ideally even for the memcache drivers i'd kind of like to just provide keystone_memcache as a va | 01:26 |
| jamielennox | r | 01:26 |
| jamielennox | i'm not a good enough ansibler but i would expect you could integrate them via defaults if available | 01:26 |
| cloudnull | so theres just a connection string and assume memcached is simply installed at that local ? | 01:27 |
| cloudnull | as an example | 01:27 |
| jamielennox | well some way that i can orchestate memcache and just feed the appropriate vars to keystone | 01:28 |
| jamielennox | same with galera_client i guess | 01:28 |
| jamielennox | i want to deal with setting up mysql/galera/postgres whatever as a seperate role | 01:29 |
| cloudnull | for sure. | 01:29 |
| cloudnull | that will be done | 01:29 |
| bgmccollum | cloudnull: re upgrade pins...10.1.17 has lxc pinned to 1.0.7. trying to upgrade to 11.0.4 barfs because lxc 1.0.8 is required, but nothing cleared out the lxc pin (its appears unbound in kilo)...juno head has updated lxc pin to 1.0.8. | 01:29 |
| bgmccollum | actually, i think the problem is the upgrade scripts runs lxc-hosts-setup.yml before re-templating the pins... | 01:30 |
| kysse | hmh. I should install ssd disks to those infra nodes, I hate slow disks )-; | 01:30 |
| jamielennox | cloudnull: ok, sweet - just looks like i was trying to use individually roles a bit prematurely | 01:30 |
| cloudnull | yea, sorry about that | 01:30 |
| cloudnull | itll be there soon-ish | 01:30 |
| * cloudnull goes to find more hours in the day :) | 01:31 | |
| bgmccollum | haha | 01:31 |
| jamielennox | nah, i had to do a bunch of hacking and forking just to get the setup to run | 01:31 |
| cloudnull | jamielennox: i hope to have it done for most of the OS_* roles by the end of the month. my next batch will be all of the galera, rabbit, memcached, etc . | 01:32 |
| jamielennox | will simplify some of my test deployments a lot to have it | 01:32 |
| cloudnull | but those should go quic,k | 01:32 |
| cloudnull | bgmccollum: is that the new or the old upgrade script ? | 01:33 |
| jamielennox | can't believe there aren't standard modules already for things like rabbit, memcache, galera | 01:33 |
| cloudnull | there are "some" but from what we found they're a bit suspect | 01:33 |
| bgmccollum | cloudnull: old :/ | 01:34 |
| cloudnull | debopts has some nice roles but they wont work cross platfor which is something that we're going toward soonish also | 01:34 |
| cloudnull | bgmccollum: that too | 01:34 |
| bgmccollum | cloudnull: i think i need to just set apt_pinned_packages: [] in the user vars... | 01:34 |
| bgmccollum | so its "defined", but empty...should clean out the pins | 01:34 |
| cloudnull | bgmccollum: is the apt pinning should run. is the pin in the user_vars ? | 01:35 |
| cloudnull | kysse: great success yet ? | 01:35 |
| kysse | not yet, I have slow disks. | 01:35 |
| cloudnull | ah. now i get it | 01:35 |
| cloudnull | load in another hampster :p | 01:36 |
| cloudnull | jamielennox: when i get to the keystone role ill ping you, if you dont mind, so that we get it working for your usecase | 01:36 |
| kysse | not everyone has 10 node cluster | 01:36 |
| bgmccollum | cloudnull: the pin was laid down in 10.1.17...i dont see any pins defined in 11.0.4, so when it gets to here to actually templating the pinning, its skipped -- https://github.com/openstack/openstack-ansible/blob/kilo/playbooks/roles/apt_package_pinning/tasks/main.yml#L20 | 01:36 |
| cloudnull | kysse: this is true | 01:36 |
| cloudnull | ah, yea that would do it bgmccollum | 01:37 |
| cloudnull | its not removing the pins because nothing is defined | 01:37 |
| bgmccollum | so i just need to ensure apt_pinned_packages: [] is in my user vars | 01:37 |
| jamielennox | cloudnull: sounds good, be glad to try it out, there's a lot of stuff in os_keystone that i'm keen to have an automated way to do | 01:37 |
| cloudnull | makes sense now | 01:37 |
| bgmccollum | yeah... :/ | 01:37 |
| cloudnull | jamielennox: sweet, then will be now soon. ill let you know. | 01:38 |
| cloudnull | :) | 01:38 |
| cloudnull | bgmccollum: i guess you could re-pin it to a new version ? | 01:39 |
| cloudnull | i dont think it would be a good idea for us to remove pins by default because there may be other things in that list of pinned packages which some deployer may need . | 01:40 |
| cloudnull | but then again i dont think we should be pinning packages ever. but thats a conversation for the ages. | 01:41 |
| kysse | I ran into weird problem with haproxy, 'AnsibleError: host not found:. hngh. | 01:41 |
| cloudnull | kinda like maintianing a private fork, never updating, and then complaining the deployment lacks functionality or is insecure. :\ | 01:42 |
| bgmccollum | haha | 01:42 |
| kysse | oh. I'm running it too early.. | 01:42 |
| cloudnull | ok i've got ot go , have a good one all. ttyl | 01:49 |
| bgmccollum | later | 01:49 |
| *** networkninja has quit IRC | 01:53 | |
| *** alkari has joined #openstack-ansible | 02:48 | |
| *** iceyao has quit IRC | 03:21 | |
| *** cemmason has joined #openstack-ansible | 03:36 | |
| *** iceyao has joined #openstack-ansible | 03:38 | |
| *** sacharya has joined #openstack-ansible | 03:47 | |
| *** alkari has quit IRC | 03:58 | |
| *** tlian has quit IRC | 04:47 | |
| *** fawadkhaliq has joined #openstack-ansible | 04:48 | |
| *** sacharya has quit IRC | 04:50 | |
| *** sacharya has joined #openstack-ansible | 04:55 | |
| *** darrenc is now known as darrenc_afk | 05:10 | |
| *** alkari has joined #openstack-ansible | 05:22 | |
| *** markvoelker has joined #openstack-ansible | 05:24 | |
| *** darrenc_afk is now known as darrenc | 05:32 | |
| *** markvoelker has quit IRC | 05:41 | |
| *** sacharya has quit IRC | 05:48 | |
| *** sacharya has joined #openstack-ansible | 05:49 | |
| kysse | cloudnull: it worked | 05:55 |
| *** alkari has quit IRC | 06:08 | |
| *** targon has joined #openstack-ansible | 06:13 | |
| *** harlowja has quit IRC | 06:16 | |
| *** javeriak has joined #openstack-ansible | 06:22 | |
| *** mss has quit IRC | 06:25 | |
| *** sacharya has quit IRC | 06:32 | |
| *** phiche has joined #openstack-ansible | 06:45 | |
| *** phiche has quit IRC | 06:50 | |
| *** phiche has joined #openstack-ansible | 07:00 | |
| *** mss has joined #openstack-ansible | 07:21 | |
| *** sacharya has joined #openstack-ansible | 07:33 | |
| *** hassaan has joined #openstack-ansible | 07:35 | |
| hassaan | I have a clean juno Openstack setup. I am trying to upgrade juno to kilo using the run-upgrade scripts. I ran into an error on the task: lxc_hosts | Install apt packages. The error is as follows: The following packages have unmet dependencies: lxc-dev : Depends: lxc (= 1.0.8-0ubuntu0.3) but 1.0.7-0ubuntu0.7 is to be installed | 07:36 |
| *** sacharya has quit IRC | 07:38 | |
| *** mpavone has joined #openstack-ansible | 07:45 | |
| mattt | hassaan: i heard chatter of this one yesterday | 08:17 |
| mattt | hassaan: mancdaz can probably fill you in w/ more info | 08:17 |
| hassaan | mattt: Thank you for the lead | 08:19 |
| mattt | hassaan: from what i gather openstack-ansible pins LXC version, and that causes some breakage w/ a recent Ubuntu 14.04 lxc package update | 08:19 |
| mattt | hassaan: i'm sure we'll have a fix for this one out shortly | 08:20 |
| hassaan | mattt: I understand. Thanks! | 08:22 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Make instance info output to stdout on failure https://review.openstack.org/252247 | 08:25 |
| *** egonzalez has joined #openstack-ansible | 08:29 | |
| *** fawadkhaliq has quit IRC | 08:30 | |
| odyssey4me | hassan mattt we just merged a patch to update the pin yesterday | 08:31 |
| odyssey4me | we're waiting for a few other patches to land before we tag the next juno release | 08:31 |
| odyssey4me | hassaan ^ | 08:31 |
| *** javeriak has quit IRC | 08:32 | |
| hassaan | odyssey4me: Can I get a link to the patch so I can apply it on my local setup | 08:33 |
| *** sdake has joined #openstack-ansible | 08:34 | |
| odyssey4me | hassaan you can either apply https://review.openstack.org/251945, or simply add the updated apt_pinned_packages dict to your user_variables | 08:34 |
| *** sdake has quit IRC | 08:37 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: [WIP] Make instance info output to stdout on failure https://review.openstack.org/252247 | 08:38 |
| *** sdake has joined #openstack-ansible | 08:40 | |
| *** fawadkhaliq has joined #openstack-ansible | 08:49 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Convert AIO bootstrap from bash to Ansible https://review.openstack.org/239525 | 08:55 |
| *** gus has quit IRC | 09:07 | |
| *** gus has joined #openstack-ansible | 09:09 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Gate: Make instance info output to stdout on failure https://review.openstack.org/252247 | 09:13 |
| *** egonzalez has quit IRC | 09:14 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Gate: Make instance info output to stdout on failure https://review.openstack.org/252247 | 09:15 |
| *** raginbajin has joined #openstack-ansible | 09:17 | |
| hassaan | odyssey4me: Thankyou | 09:22 |
| odyssey4me | hassaan if you're setting up a new environment, I recommend that you rather opt to setup a Kilo or Liberty environment... Juno is EOL | 09:25 |
| hassaan | odyssey4me: Noted. Thanks alot | 09:25 |
| *** javeriak has joined #openstack-ansible | 09:45 | |
| evrardjp | odyssey4me: was there a decision about the changes done on the roles that will are already moved in the IRR? cf. conversation here: https://review.openstack.org/#/c/248064/ | 09:45 |
| *** javeriak has quit IRC | 09:59 | |
| *** hassaan has quit IRC | 10:03 | |
| *** hassaan_ has joined #openstack-ansible | 10:03 | |
| *** hassaan_ has quit IRC | 10:05 | |
| *** gparaskevas has joined #openstack-ansible | 10:12 | |
| *** javeriak has joined #openstack-ansible | 10:14 | |
| *** alkari has joined #openstack-ansible | 11:04 | |
| *** javeriak has quit IRC | 11:40 | |
| *** fawadkhaliq has quit IRC | 11:40 | |
| *** alkari has quit IRC | 11:47 | |
| *** hassaan has joined #openstack-ansible | 11:47 | |
| *** javeriak has joined #openstack-ansible | 11:48 | |
| *** javeriak has quit IRC | 11:53 | |
| hassaan | Please note, I was able to continue with the upgrade from juno to kilo only after I updated the liblxc1 package even with the fix applied. Without that the upgrade keeps failing since lxc package cannot be updated for the lib-dev to be installed. Hence the script fails. | 11:54 |
| *** javeriak has joined #openstack-ansible | 11:56 | |
| *** hassaan has quit IRC | 11:59 | |
| *** mpavone has quit IRC | 12:00 | |
| *** javeriak has quit IRC | 12:01 | |
| *** egonzalez has joined #openstack-ansible | 12:13 | |
| *** hassaan has joined #openstack-ansible | 12:42 | |
| odyssey4me | evrardjp sorry, was out - what decision are you looking for? I'm a little confused. :) | 12:55 |
| odyssey4me | hassaan was this with the patch applied (and the playbooks re-run), or was this with the application of updated pins in user_variables? | 12:56 |
| *** sdake has quit IRC | 13:00 | |
| mgariepy | good morning everyone | 13:02 |
| *** hassaan has quit IRC | 13:04 | |
| odyssey4me | o/ mgariepy | 13:06 |
| mgariepy | how are you doing ? | 13:06 |
| mhayden | morning | 13:10 |
| odyssey4me | morning mhayden - so we're beyond the flavor differences, and now into services not behaving well for https://review.openstack.org/239525 | 13:15 |
| odyssey4me | so I guess that something isn't being done quite right in the config somewhere... I'd appreciate your help trying to figure out where | 13:15 |
| * mhayden is looking | 13:16 | |
| *** javeriak has joined #openstack-ansible | 13:16 | |
| mhayden | u'message': u'No valid host was found. There are not enough hosts available.' | 13:17 |
| mhayden | i wonder if we're hitting a ram limit | 13:17 |
| odyssey4me | mhayden unlikely, that's be from one of the schedulers | 13:19 |
| odyssey4me | so we need to figure out which one, then why it thinks the host is not available | 13:19 |
| mhayden | aren't we only running one scheduler? | 13:20 |
| mhayden | i'm going to build this on a rax 8gb right quick so i can mess with it | 13:21 |
| kmARC | Quick question guys: what's the preferred method to restart a service (let's say, neutron)? | 13:23 |
| kmARC | Currently I'm doing a setup-everything. Maybe there is --limit neutron_all or something? | 13:24 |
| kmARC | I see the os_neutron/handlers/main.yml which has a hint (Restart neutron services). Can I invoke this via openstack-ansible or do I need some pure ansible machinery? | 13:28 |
| gparaskevas | do you need to restart all neutron serviceS? | 13:28 |
| kmARC | yes, for example | 13:29 |
| kmARC | I guess it should be with some delays | 13:29 |
| gparaskevas | just a restart no config change? | 13:29 |
| kmARC | I'm interested in both | 13:30 |
| mhayden | odyssey4me: that build came up with three rabbitmq containers | 13:32 |
| mhayden | but it should have had one | 13:33 |
| * mhayden is looking for where that logic went | 13:33 | |
| odyssey4me | mhayden don't worry about that, I reverted that logic back to the original intent | 13:33 |
| mhayden | oh, so we want to have 3? | 13:33 |
| odyssey4me | it was cut down because we thought that there was a resource issue, but it turned out to be a disk space issue | 13:33 |
| odyssey4me | but that's easy enough to cut them down as a tester | 13:34 |
| gparaskevas | kmARC: if you need to deploy a new change then as you mentioned you do the change to the desired ansible template and then openstack-ansible --limit neutron_all --tags "the tag(s) name you want to call" | 13:35 |
| mhayden | there seems to be enough RAM... the scheduler claims > 5GB RAM available | 13:35 |
| kmARC | gparaskevas: thanks. what limit options are there I can choose from, and also what entities are denoted by `tags`? | 13:36 |
| gparaskevas | kmARC: every task in openstack ansible is taged for that specific reason so you can call explicit set of tasks. | 13:36 |
| mhayden | but the scheduler claimed that something else fialed | 13:36 |
| kmARC | ah, ok | 13:36 |
| *** markvoelker has joined #openstack-ansible | 13:36 | |
| mhayden | odyssey4me: libvirtError: internal error: no supported architecture for os type 'hvm' | 13:38 |
| mhayden | i can't find where we're setting libvirt_type anywhere | 13:40 |
| mhayden | it's been a while since i looked at nova's defaults | 13:40 |
| *** markvoelker_ has joined #openstack-ansible | 13:41 | |
| odyssey4me | aha, so this is the issue that was being discussed yesterday | 13:41 |
| odyssey4me | I deliberately left out the hard setting of the virt_type as there is code that's support to detect it. Apparently that code doesn't work. :/ | 13:42 |
| odyssey4me | let me track it down. | 13:42 |
| *** markvoelker has quit IRC | 13:42 | |
| odyssey4me | thanks for finding that! | 13:42 |
| gparaskevas | limit option are ragard to the inventory so you can check /etc/openstack_deploy/env.d/neutron.yml | 13:43 |
| gparaskevas | kmARC: limit option are ragard to the inventory so you can check /etc/openstack_deploy/env.d/neutron.yml | 13:43 |
| mhayden | odyssey4me: no problem! :) | 13:48 |
| kmARC | gparaskevas: ah, I see, it makes sense now | 13:49 |
| gparaskevas | kmARC: yeah , always pleasure! | 13:50 |
| kmARC | gparaskevas: so and what about tags? | 13:50 |
| kmARC | I saw some in the playbooks | 13:50 |
| *** adac has joined #openstack-ansible | 13:51 | |
| adac | Guys, i did a fresh installation of openstack-ansible today but still the galera cluster, executing the command ansible galera_container -m shell -a "cat /var/lib/mysql/grastate.dat" right after the installation just shows just the following: https://gist.github.com/anonymous/440236da2852e12f6279 | 13:53 |
| odyssey4me | mhayden https://github.com/openstack/openstack-ansible/commit/42bca0e390eb1cccbda6acabd231827ef798123f | 13:59 |
| mhayden | oof | 14:00 |
| *** javeriak has quit IRC | 14:00 | |
| mhayden | so those flags can be passed through without nested virt being enabled | 14:00 |
| mhayden | perhaps we should add a nested virt check? | 14:00 |
| odyssey4me | well, I just tested on a vm i cloud | 14:02 |
| odyssey4me | I get a zero value, so my virt_type should be qemu - which should work as before | 14:02 |
| odyssey4me | ah, but the value is being set incorrectly in nova.conf | 14:02 |
| kmARC | OK, I have a networking related question now. gparaskevas maybe you can help me with this too :) | 14:06 |
| odyssey4me | mhayden and I see why - it's because nova_virt_type is defined as a default... I'll patch this up | 14:07 |
| kmARC | So on the config file the examples enumerate some networks. What I need, is flat networking, however, in the example it maps to eth12, which is good in the containers, not correct for the compute node though (it doesn't have containers) | 14:07 |
| kmARC | what's kind of suspicous is the host_bind_override: "eth12" line. However I have no idea what this host_bind_override does | 14:07 |
| mhayden | odyssey4me: woot | 14:08 |
| mhayden | let me know when the review is up and i'll gander | 14:08 |
| mhayden | i will apply a +1 with gusto | 14:08 |
| gparaskevas | mhayden: ok, dont bother if its not on the compute nodes because thereis an option that overides that for the compute nodes "host_bind_override:" and there you define the specific interface of the compute node | 14:10 |
| gparaskevas | mhayden: whrong usename :P | 14:11 |
| mhayden | gparaskevas: i still like ya | 14:11 |
| kmARC | (Just for the record, I might have found it: https://github.com/openstack/openstack-ansible/blob/master/playbooks/library/provider_networks#L197 - it seems this configuration sets the corresponding mapped interface) | 14:11 |
| gparaskevas | kmARC: ok, dont bother if its not on the compute nodes because thereis an option that overides that for the compute nodes "host_bind_override:" and there you define the specific interface of the compute node | 14:11 |
| kmARC | Yepp, found it | 14:11 |
| gparaskevas | kmARC: the option can be found in the openstack_user_config.yml | 14:11 |
| gparaskevas | kmARC: great | 14:12 |
| *** hassaan has joined #openstack-ansible | 14:12 | |
| kmARC | ok now it seems the config changed, let me check | 14:13 |
| hassaan | odyssey4me, sorry for late reply, was away. Yes this was with the application of updated pins in user_variables and playbook rerun | 14:13 |
| *** fawadkhaliq has joined #openstack-ansible | 14:15 | |
| *** hassaan has quit IRC | 14:18 | |
| openstackgerrit | Darren Birkett proposed openstack/openstack-ansible: turn neutron notifications off by default https://review.openstack.org/252416 | 14:20 |
| gparaskevas | kmARC: what brancj are you on? | 14:24 |
| kmARC | gparakevas: so I'm trying to debug something now while looking at rsyslog messages, however I'm still getting OLD messages but don't know why. Have a look here: http://paste.openstack.org/show/480641/ | 14:25 |
| kmARC | gparaskevas: I'm on Liberty | 14:25 |
| gparaskevas | kmARC: this is from compute node? | 14:28 |
| *** tlian has joined #openstack-ansible | 14:29 | |
| *** hassaan has joined #openstack-ansible | 14:31 | |
| kmARC | no, it's from the rsyslog container | 14:36 |
| kmARC | gparaskevas: anyways, I think I have multiple issues here :) It'd be great if I'd have the logs in cronological order, and then I would able to fix my 'no compute host found' issue. | 14:39 |
| gparaskevas | kmARC: good luck :) | 14:42 |
| gparaskevas | kmARC: you are going to need it! | 14:42 |
| gparaskevas | kmARC: make small changes so you dont get lost always | 14:43 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Fix nova_virt_type auto-detection https://review.openstack.org/252426 | 14:43 |
| odyssey4me | mhayden ^ | 14:43 |
| mhayden | wheee | 14:43 |
| * mhayden ganders | 14:44 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Convert AIO bootstrap from bash to Ansible https://review.openstack.org/239525 | 14:44 |
| *** fawadkhaliq has quit IRC | 14:44 | |
| mhayden | odyssey4me: you get my strongest +1 | 14:45 |
| mhayden | :P | 14:45 |
| mhayden | that's a clean fix | 14:45 |
| mhayden | i like it | 14:45 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Fix nova_virt_type auto-detection https://review.openstack.org/252426 | 14:46 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Convert AIO bootstrap from bash to Ansible https://review.openstack.org/239525 | 14:46 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Gate: Make instance info output to stdout on failure https://review.openstack.org/252247 | 14:47 |
| odyssey4me | mhayden spelling error in the commit msg, fixed :) | 14:48 |
| *** KLevenstein has joined #openstack-ansible | 14:53 | |
| *** spotz_zzz is now known as spotz | 14:56 | |
| *** mss has quit IRC | 14:58 | |
| *** leifmadsen has quit IRC | 15:01 | |
| mattt | this kvm auto detect thing is on interest | 15:01 |
| *** leifmadsen has joined #openstack-ansible | 15:03 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible: Use Ansible AIO playbook for run-aio-build.sh https://review.openstack.org/244708 | 15:04 |
| *** baker has joined #openstack-ansible | 15:06 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 15:09 | |
| adac | also additonally run this command does not solve it "openstack-ansible -e galera_ignore_cluster_state=true galera-install.yml" Not sure what I can try else. If it is not even working wehn istalling everything from scratch with a clean ubuntu installation :/ | 15:16 |
| *** targon has quit IRC | 15:16 | |
| *** raginbajin has quit IRC | 15:16 | |
| *** raginbajin has joined #openstack-ansible | 15:17 | |
| *** linggao has joined #openstack-ansible | 15:21 | |
| *** hassaan has quit IRC | 15:26 | |
| *** fawadkhaliq has joined #openstack-ansible | 15:28 | |
| *** markvoelker_ has quit IRC | 15:28 | |
| adac | There is also a log node, can I check the galera/mariadb log probably there? | 15:29 |
| *** ChanServ has quit IRC | 15:34 | |
| *** raginbajin has quit IRC | 15:34 | |
| *** gparaskevas has quit IRC | 15:34 | |
| *** marekd has quit IRC | 15:34 | |
| *** grumpycatt has quit IRC | 15:34 | |
| *** pabelanger has quit IRC | 15:34 | |
| *** egonzalez has quit IRC | 15:34 | |
| *** gtt116_ has quit IRC | 15:34 | |
| *** mcarden has quit IRC | 15:34 | |
| *** jbweber has quit IRC | 15:34 | |
| *** erikmwilson has quit IRC | 15:34 | |
| *** jwagner has quit IRC | 15:34 | |
| *** misc has quit IRC | 15:34 | |
| *** xek has quit IRC | 15:34 | |
| *** metral has quit IRC | 15:34 | |
| *** stevelle has quit IRC | 15:34 | |
| *** dalees has quit IRC | 15:34 | |
| *** finchd has quit IRC | 15:34 | |
| *** mgagne has quit IRC | 15:34 | |
| *** b3rnard0 has quit IRC | 15:34 | |
| *** fawadkhaliq has quit IRC | 15:34 | |
| *** adac has quit IRC | 15:34 | |
| *** cemmason has quit IRC | 15:34 | |
| *** alextricity has quit IRC | 15:35 | |
| *** jlvillal has quit IRC | 15:35 | |
| *** thrawn01 has quit IRC | 15:35 | |
| *** e-vad has quit IRC | 15:35 | |
| *** mattoliverau has quit IRC | 15:35 | |
| *** SamYaple has quit IRC | 15:35 | |
| *** mrda has quit IRC | 15:35 | |
| *** phschwartz has quit IRC | 15:35 | |
| *** charz has quit IRC | 15:35 | |
| *** mgariepy has quit IRC | 15:35 | |
| *** rev_dr has quit IRC | 15:35 | |
| *** Sam-I-Am has quit IRC | 15:35 | |
| *** rcarrillocruz has quit IRC | 15:35 | |
| *** jhesketh has quit IRC | 15:35 | |
| *** dweaver has quit IRC | 15:35 | |
| *** dmsimard has quit IRC | 15:35 | |
| *** sirushti has quit IRC | 15:35 | |
| *** wabu has quit IRC | 15:35 | |
| *** arbrandes has quit IRC | 15:35 | |
| *** evrardjp has quit IRC | 15:35 | |
| *** ggillies has quit IRC | 15:35 | |
| *** pellaeon_ has quit IRC | 15:35 | |
| *** sshen_ has quit IRC | 15:35 | |
| *** dstanek has quit IRC | 15:35 | |
| *** persia has quit IRC | 15:35 | |
| *** lkoranda has quit IRC | 15:35 | |
| *** andymccr has quit IRC | 15:35 | |
| *** coolj has quit IRC | 15:35 | |
| *** dolpher has quit IRC | 15:35 | |
| *** Ti-mo has quit IRC | 15:35 | |
| *** bapalm has quit IRC | 15:35 | |
| *** richoid has quit IRC | 15:35 | |
| *** galstrom_zzz has quit IRC | 15:35 | |
| *** d9k has quit IRC | 15:35 | |
| *** toddnni has quit IRC | 15:35 | |
| *** _hanhart has quit IRC | 15:35 | |
| *** rackertom has quit IRC | 15:35 | |
| *** etoews has quit IRC | 15:35 | |
| *** maximov_ has quit IRC | 15:35 | |
| *** meteorfox has quit IRC | 15:35 | |
| *** mhayden has quit IRC | 15:35 | |
| *** errr has quit IRC | 15:35 | |
| *** baker has quit IRC | 15:35 | |
| *** leifmadsen has quit IRC | 15:35 | |
| *** KLevenstein has quit IRC | 15:35 | |
| *** mancdaz has quit IRC | 15:35 | |
| *** jaypipes has quit IRC | 15:35 | |
| *** tiagogomes has quit IRC | 15:35 | |
| *** mrodden has quit IRC | 15:35 | |
| *** kore has quit IRC | 15:35 | |
| *** elo has quit IRC | 15:35 | |
| *** Tridde has quit IRC | 15:35 | |
| *** darrenc has quit IRC | 15:35 | |
| *** niknakpaddywak has quit IRC | 15:35 | |
| *** toan has quit IRC | 15:35 | |
| *** McMurlock has quit IRC | 15:35 | |
| *** mattt has quit IRC | 15:35 | |
| *** Guest4784 has quit IRC | 15:35 | |
| *** kysse has quit IRC | 15:35 | |
| *** tlbr has quit IRC | 15:35 | |
| *** antonym has quit IRC | 15:35 | |
| *** logan- has quit IRC | 15:35 | |
| *** admiralboom has quit IRC | 15:35 | |
| *** fignew- has quit IRC | 15:35 | |
| *** miguelgrinberg has quit IRC | 15:35 | |
| *** rromans has quit IRC | 15:35 | |
| *** jmccrory has quit IRC | 15:35 | |
| *** prometheanfire has quit IRC | 15:35 | |
| *** jcannava has quit IRC | 15:35 | |
| *** larsks has quit IRC | 15:35 | |
| *** andyhky has quit IRC | 15:35 | |
| *** git-harry has quit IRC | 15:35 | |
| *** bryan_att has quit IRC | 15:35 | |
| *** spotz has quit IRC | 15:35 | |
| *** serverascode has quit IRC | 15:35 | |
| *** lbragstad has quit IRC | 15:35 | |
| *** jroll has quit IRC | 15:35 | |
| *** Apsu has quit IRC | 15:35 | |
| *** jamielennox has quit IRC | 15:35 | |
| *** rETROpunK has quit IRC | 15:35 | |
| *** mattronix has quit IRC | 15:35 | |
| *** cloudnull has quit IRC | 15:35 | |
| *** bogeyon18 has quit IRC | 15:35 | |
| *** d34dh0r53 has quit IRC | 15:35 | |
| *** sigmavirus24 has quit IRC | 15:35 | |
| *** dolphm has quit IRC | 15:35 | |
| *** eglute has quit IRC | 15:35 | |
| linggao | Why did everyone quit? What did I do? | 15:36 |
| *** palendae has quit IRC | 15:37 | |
| *** sacharya has joined #openstack-ansible | 15:37 | |
| *** dweaver has joined #openstack-ansible | 15:37 | |
| *** dmsimard has joined #openstack-ansible | 15:37 | |
| *** sirushti has joined #openstack-ansible | 15:37 | |
| *** wabu has joined #openstack-ansible | 15:37 | |
| *** arbrandes has joined #openstack-ansible | 15:37 | |
| *** evrardjp has joined #openstack-ansible | 15:37 | |
| *** ggillies has joined #openstack-ansible | 15:37 | |
| *** pellaeon_ has joined #openstack-ansible | 15:37 | |
| *** sshen_ has joined #openstack-ansible | 15:37 | |
| *** palendae has joined #openstack-ansible | 15:38 | |
| *** mhayden has joined #openstack-ansible | 15:38 | |
| evrardjp | odyssey4me: sorry for the late answer, but you answered yourself with a -1 workflow on the commit :) | 15:38 |
| mhayden | hooray for netsplits | 15:38 |
| linggao | Morning ansible. | 15:39 |
| *** egonzalez has joined #openstack-ansible | 15:39 | |
| *** gtt116_ has joined #openstack-ansible | 15:39 | |
| *** mcarden has joined #openstack-ansible | 15:39 | |
| *** jbweber has joined #openstack-ansible | 15:39 | |
| *** erikmwilson has joined #openstack-ansible | 15:39 | |
| *** jwagner has joined #openstack-ansible | 15:39 | |
| *** misc has joined #openstack-ansible | 15:39 | |
| *** xek has joined #openstack-ansible | 15:39 | |
| *** metral has joined #openstack-ansible | 15:39 | |
| *** stevelle has joined #openstack-ansible | 15:39 | |
| *** dalees has joined #openstack-ansible | 15:39 | |
| *** finchd has joined #openstack-ansible | 15:39 | |
| *** mgagne has joined #openstack-ansible | 15:39 | |
| *** b3rnard0 has joined #openstack-ansible | 15:39 | |
| *** d9k has joined #openstack-ansible | 15:39 | |
| *** galstrom_zzz has joined #openstack-ansible | 15:39 | |
| *** toddnni has joined #openstack-ansible | 15:39 | |
| *** _hanhart has joined #openstack-ansible | 15:39 | |
| *** rackertom has joined #openstack-ansible | 15:39 | |
| *** etoews has joined #openstack-ansible | 15:39 | |
| *** maximov_ has joined #openstack-ansible | 15:39 | |
| evrardjp | morning everyone | 15:39 |
| *** marekd has joined #openstack-ansible | 15:39 | |
| *** grumpycatt has joined #openstack-ansible | 15:39 | |
| *** palendae has quit IRC | 15:39 | |
| *** palendae has joined #openstack-ansible | 15:39 | |
| *** mgariepy has joined #openstack-ansible | 15:40 | |
| evrardjp | mhayden: so I'm not the only one to have these issues? It's been like 1 week of issues, maybe it's time to move to a more recent protocol, like xmpp? | 15:40 |
| evrardjp | sorry mhayden, the complete sentence was not for you. | 15:40 |
| *** raginbajin has joined #openstack-ansible | 15:40 | |
| mhayden | lots of DDoS going on this week :( | 15:40 |
| *** antonym has joined #openstack-ansible | 15:40 | |
| *** logan- has joined #openstack-ansible | 15:40 | |
| *** admiralboom has joined #openstack-ansible | 15:40 | |
| *** fignew- has joined #openstack-ansible | 15:40 | |
| *** miguelgrinberg has joined #openstack-ansible | 15:40 | |
| *** rromans has joined #openstack-ansible | 15:40 | |
| *** jmccrory has joined #openstack-ansible | 15:40 | |
| *** prometheanfire has joined #openstack-ansible | 15:40 | |
| *** jcannava has joined #openstack-ansible | 15:40 | |
| *** larsks has joined #openstack-ansible | 15:40 | |
| *** andyhky has joined #openstack-ansible | 15:40 | |
| mhayden | looks like people are coming back online now | 15:40 |
| *** jhesketh has joined #openstack-ansible | 15:41 | |
| mhayden | i was on wilhelm.freenode.net which seemed to avoid some of the shenanigans until today | 15:41 |
| evrardjp | mhayden: DDoS are the origin of the disconnections? | 15:41 |
| kmARC | anyone see anything like this: http://paste.openstack.org/show/480653/ I can list, show the image, my compute node cannot | 15:41 |
| mgariepy | disconnect/reconnect.. | 15:41 |
| mhayden | evrardjp: yeah, freenode is getting DDoS'ed in a few places | 15:41 |
| mhayden | not sure if this message came through earlier, but some eyes on https://review.openstack.org/#/c/245813/ would be appreciated | 15:41 |
| evrardjp | pff. ain't nobody got time for that? | 15:42 |
| mhayden | lololol | 15:42 |
| *** sacharya has quit IRC | 15:42 | |
| *** gparaskevas has joined #openstack-ansible | 15:42 | |
| *** dstanek has joined #openstack-ansible | 15:42 | |
| *** persia has joined #openstack-ansible | 15:42 | |
| *** coolj has joined #openstack-ansible | 15:42 | |
| *** lkoranda has joined #openstack-ansible | 15:42 | |
| *** andymccr has joined #openstack-ansible | 15:42 | |
| *** dolpher has joined #openstack-ansible | 15:42 | |
| *** Ti-mo has joined #openstack-ansible | 15:42 | |
| *** bapalm has joined #openstack-ansible | 15:42 | |
| *** richoid has joined #openstack-ansible | 15:42 | |
| *** meteorfox has joined #openstack-ansible | 15:42 | |
| *** errr has joined #openstack-ansible | 15:42 | |
| *** pabelanger has joined #openstack-ansible | 15:42 | |
| linggao | Hi all. What is the status for ironic support in ansible? Can I get draft to try? | 15:43 |
| *** git-harry has joined #openstack-ansible | 15:44 | |
| *** bryan_att has joined #openstack-ansible | 15:44 | |
| *** spotz has joined #openstack-ansible | 15:44 | |
| *** serverascode has joined #openstack-ansible | 15:44 | |
| *** lbragstad has joined #openstack-ansible | 15:44 | |
| *** jroll has joined #openstack-ansible | 15:44 | |
| *** Apsu has joined #openstack-ansible | 15:44 | |
| *** jamielennox has joined #openstack-ansible | 15:44 | |
| *** rETROpunK has joined #openstack-ansible | 15:44 | |
| *** mattronix has joined #openstack-ansible | 15:44 | |
| *** cloudnull has joined #openstack-ansible | 15:44 | |
| *** bogeyon18 has joined #openstack-ansible | 15:44 | |
| *** d34dh0r53 has joined #openstack-ansible | 15:44 | |
| *** sigmavirus24 has joined #openstack-ansible | 15:44 | |
| *** dolphm has joined #openstack-ansible | 15:44 | |
| *** eglute has joined #openstack-ansible | 15:44 | |
| *** phschwartz_ has joined #openstack-ansible | 15:44 | |
| *** phschwartz_ has quit IRC | 15:45 | |
| *** phschwartz_ has joined #openstack-ansible | 15:48 | |
| evrardjp | openstack-ansible doesn't currently have any ironic support, or at least none that I'm aware of | 15:48 |
| evrardjp | but I'd happy to see some of it | 15:48 |
| evrardjp | :D | 15:48 |
| cloudnull | morning | 15:49 |
| linggao | evrardjp, I heard that someone was working on it. I forgot the name of the person. | 15:49 |
| evrardjp | there is a work done on ironic standalone and ansible, it's called bifrost | 15:49 |
| evrardjp | https://github.com/openstack/bifrost | 15:49 |
| evrardjp | morning cloudnull | 15:50 |
| evrardjp | cloud: I'll adapt my mtu thingies inside your new roles (IRR) if you want | 15:51 |
| linggao | evrardjp, yes, I tried that. I am using openstack-ansible to install nodes. Just hope it support ironic. | 15:51 |
| *** ctina has joined #openstack-ansible | 15:51 | |
| evrardjp | linggao: at the moment, the deployer chooses its way to deploy the bare metal nodes, and openstack-ansible takes the relay further | 15:51 |
| evrardjp | however ironic as nova backend (I don't know how you call that), seems very interesting IMO | 15:52 |
| evrardjp | I'm not sure I understood your question to be honest :p | 15:54 |
| stevelle | I thought I heard that someone may have been building a role for ironic but I don't know status of it and don't recall a name of the author either linggao | 15:54 |
| *** markvoelker has joined #openstack-ansible | 15:55 | |
| linggao | evrardjp, like I am using openstack-ansible to install nova, neutron, glance etc. I hope it can do ironic. | 15:55 |
| linggao | serverascode, I guess 'a role for ironic' is the stuff I should ask. thanks :-) | 15:56 |
| linggao | I saw https://github.com/jriguera/ansible-ironic-standalone, is it related to this project? | 15:57 |
| stevelle | I don't believe that is | 15:58 |
| odyssey4me | linggao https://github.com/mrda/openstack-ansible-ironic is the current effort which is ready for some eyes, but not yet necessarily ready for integration | 15:58 |
| *** adac has joined #openstack-ansible | 15:59 | |
| *** fawadkhaliq has joined #openstack-ansible | 15:59 | |
| *** alextricity has joined #openstack-ansible | 15:59 | |
| *** Sam-I-Am has joined #openstack-ansible | 15:59 | |
| *** jlvillal has joined #openstack-ansible | 15:59 | |
| *** thrawn01 has joined #openstack-ansible | 15:59 | |
| *** e-vad has joined #openstack-ansible | 15:59 | |
| *** mattoliverau has joined #openstack-ansible | 15:59 | |
| *** SamYaple has joined #openstack-ansible | 15:59 | |
| *** mrda has joined #openstack-ansible | 15:59 | |
| *** charz has joined #openstack-ansible | 15:59 | |
| *** rev_dr has joined #openstack-ansible | 15:59 | |
| *** rcarrillocruz has joined #openstack-ansible | 15:59 | |
| odyssey4me | stevelle evrardjp ^ | 15:59 |
| linggao | odyssey4me, thanks, will it eventually be checked in to openstack-ansible tree? | 16:00 |
| odyssey4me | evrardjp those roles aren't cloudnull's - they are now the community's - they're being used for the gate tests :) | 16:00 |
| stevelle | linggao: file:///Users/hairaldo/dev/github.com/openstack/openstack-ansible/doc/build/html/developer-docs/extending.html#adding-galaxy-roles | 16:01 |
| odyssey4me | linggao they will become part of the umbrella, yes - I'll be discussing the status with mrda later this evening | 16:01 |
| *** k_stev has joined #openstack-ansible | 16:01 | |
| stevelle | err sorry linggao but in the docs we have an entry for adding galaxy roles | 16:01 |
| odyssey4me | haha stevelle I think you meant: http://docs.openstack.org/developer/openstack-ansible/developer-docs/extending.html#adding-galaxy-roles | 16:01 |
| evrardjp | odyssey4me: indeed, but cloudnull started the work on it, that's why I say that, sorry for my wrong language :p | 16:01 |
| *** adaccada has joined #openstack-ansible | 16:04 | |
| stevelle | I did, odyssey4me. forgot the link I had open was from a local build | 16:04 |
| cloudnull | mattt hughsaunders if you have a moment can you review https://review.openstack.org/#/c/241483/ again | 16:06 |
| cloudnull | im blocked on any more of the IRR work until that goes in | 16:06 |
| *** ctina has quit IRC | 16:07 | |
| adaccada | This is the output of mysld.log: https://gist.github.com/anonymous/54fc25147686b298d78c and here is the output of mysql_safe.log: https://gist.github.com/anonymous/6584c73456a02f1d2ed4 all looks fine and the recovery seems to have taken place and the mysqld server is running, however a cat on /var/lib/mysql/grastate.dat still only shows https://gist.github.com/anonymous/671f3492b6311c968eb6 | 16:07 |
| *** ctina has joined #openstack-ansible | 16:07 | |
| adaccada | not sure why this is happening | 16:08 |
| adaccada | It is a fresh installation, I did not modify anything yet | 16:08 |
| *** adac has quit IRC | 16:09 | |
| *** fawadkhaliq has quit IRC | 16:09 | |
| *** alextricity has quit IRC | 16:09 | |
| *** jlvillal has quit IRC | 16:09 | |
| *** thrawn01 has quit IRC | 16:09 | |
| *** e-vad has quit IRC | 16:09 | |
| *** mattoliverau has quit IRC | 16:09 | |
| *** SamYaple has quit IRC | 16:09 | |
| *** mrda has quit IRC | 16:09 | |
| *** charz has quit IRC | 16:09 | |
| *** rev_dr has quit IRC | 16:09 | |
| *** Sam-I-Am has quit IRC | 16:09 | |
| *** rcarrillocruz has quit IRC | 16:09 | |
| cloudnull | adaccada: looks like its running | 16:11 |
| cloudnull | are you not able to connect to the galera cluster ? | 16:11 |
| adaccada | cloudnull, not sure how to test that? | 16:12 |
| *** javeriak has joined #openstack-ansible | 16:12 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-lxc_hosts: Added to ability to set mtu for lxcbr0 https://review.openstack.org/252489 | 16:14 |
| *** jaypipes has joined #openstack-ansible | 16:14 | |
| *** palendae has quit IRC | 16:14 | |
| adaccada | cloudnull, I thought that this command would give me the overview over the cluster: ansible galera_container -m shell -a "cat /var/lib/mysql/grastate.dat" but when I start it (before I did all the steps from here: http://docs.openstack.org/developer/openstack-ansible/install-guide/ops-galera-start.html), then I only get this output: https://gist.github.com/anonymous/1fc8ecf022ec56e066e7 | 16:16 |
| *** palendae has joined #openstack-ansible | 16:16 | |
| adaccada | So as odyssey4me told me this does mean the cluster is not really ready | 16:16 |
| prometheanfire | odyssey4me: do those ldap review need testing with the memberof plugin? or just basic ldap? | 16:17 |
| *** javeriak has quit IRC | 16:19 | |
| *** javeriak has joined #openstack-ansible | 16:22 | |
| openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-lxc_container_create: Ability to define the mtu globally (for all the container networks) https://review.openstack.org/252494 | 16:22 |
| cloudnull | adaccada: its showing the cluster state, however to know if the cluster is "ready" you should query: show status like '%wsrep%'; | 16:22 |
| cloudnull | that will look at the wsrep state | 16:23 |
| adaccada | "wsrep_local_state_comment : Synced", wsrep_cluster_size 2, wsrep_connected ON | 16:23 |
| adaccada | cloudnull, yeah just did that, seems to be online | 16:23 |
| adaccada | :) | 16:23 |
| cloudnull | lgtm | 16:23 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Add documentation for HA ceilometer https://review.openstack.org/248905 | 16:23 |
| cloudnull | :) | 16:23 |
| adaccada | jay!! :) Not sure why I have such a brutal output then with ansible galera_container -m shell -a "cat /var/lib/mysql/grastate.dat" | 16:24 |
| *** javeriak_ has joined #openstack-ansible | 16:25 | |
| gparaskevas | cloudnull: Hello cloudnull, just an irelevant question regarding openstack(i hoped you maybe know) does it support boot from iso? I have seen an post about it but i havent doen it with success | 16:26 |
| *** javeriak has quit IRC | 16:26 | |
| gparaskevas | cloudnull: just in case you know if it is suported or not :) | 16:27 |
| evrardjp | gparaskevas: not sure but I think you should use cinder for that | 16:27 |
| gparaskevas | evrardjp: i have read in post that when you specify an iso image then nova mounts a cinder volume as well | 16:27 |
| gparaskevas | evrardjp: https://wiki.openstack.org/wiki/BootFromISO | 16:28 |
| evrardjp | did you check if glance has something to do with iso? | 16:28 |
| cloudnull | gparaskevas: boot from ISO in creating a VM ? | 16:28 |
| * cloudnull looking | 16:28 | |
| gparaskevas | evrardjp: yes | 16:28 |
| gparaskevas | cloudnull: yes, launch instance with image type ISO | 16:28 |
| cloudnull | that should work | 16:28 |
| gparaskevas | cloudnull: let me try once more | 16:29 |
| gparaskevas | thanks both | 16:29 |
| cloudnull | you might need to change the console type for your install, if i remember correct. | 16:30 |
| cloudnull | i think it worked well with novnc | 16:30 |
| *** alkari has joined #openstack-ansible | 16:30 | |
| cloudnull | but ive not tried with spice | 16:30 |
| gparaskevas | mmm let me check | 16:30 |
| cloudnull | sorry , recalling memories w/out enough coffee | 16:31 |
| evrardjp | yeah that's really far for me too | 16:31 |
| evrardjp | like 2 years ago | 16:31 |
| evrardjp | :p | 16:31 |
| *** javeriak has joined #openstack-ansible | 16:31 | |
| gparaskevas | cloudnull: lol I am on redbull today... | 16:32 |
| evrardjp | maybe less I can't remember, which doesn't help. Anyway, cloudnull you I've see there was a swapiness in your change here https://review.openstack.org/#/c/248061/ | 16:32 |
| evrardjp | I didn't set it in the follow-up commit (because it's not related to mtu), so if you need it, feel free to create another commit :) | 16:33 |
| *** tlbr has joined #openstack-ansible | 16:34 | |
| *** Guest4784 has joined #openstack-ansible | 16:34 | |
| *** mattt has joined #openstack-ansible | 16:34 | |
| *** McMurlock has joined #openstack-ansible | 16:34 | |
| *** toan has joined #openstack-ansible | 16:34 | |
| *** niknakpaddywak has joined #openstack-ansible | 16:34 | |
| *** darrenc has joined #openstack-ansible | 16:34 | |
| *** Tridde has joined #openstack-ansible | 16:34 | |
| *** elo has joined #openstack-ansible | 16:34 | |
| *** kore has joined #openstack-ansible | 16:34 | |
| *** kysse has joined #openstack-ansible | 16:34 | |
| *** mrodden has joined #openstack-ansible | 16:34 | |
| *** tiagogomes has joined #openstack-ansible | 16:34 | |
| *** mancdaz has joined #openstack-ansible | 16:34 | |
| *** Klev-meeting has joined #openstack-ansible | 16:34 | |
| *** leifmadsen has joined #openstack-ansible | 16:34 | |
| *** baker has joined #openstack-ansible | 16:34 | |
| *** mss has joined #openstack-ansible | 16:34 | |
| *** fawadkhaliq has joined #openstack-ansible | 16:34 | |
| *** alextricity has joined #openstack-ansible | 16:34 | |
| *** Sam-I-Am has joined #openstack-ansible | 16:34 | |
| *** jlvillal has joined #openstack-ansible | 16:34 | |
| *** thrawn01 has joined #openstack-ansible | 16:34 | |
| *** e-vad has joined #openstack-ansible | 16:34 | |
| *** mattoliverau has joined #openstack-ansible | 16:34 | |
| *** SamYaple has joined #openstack-ansible | 16:34 | |
| *** mrda has joined #openstack-ansible | 16:34 | |
| *** charz has joined #openstack-ansible | 16:34 | |
| *** rev_dr has joined #openstack-ansible | 16:34 | |
| *** rcarrillocruz has joined #openstack-ansible | 16:34 | |
| *** ysm_ has joined #openstack-ansible | 16:34 | |
| *** javeriak_ has quit IRC | 16:35 | |
| *** eil397 has joined #openstack-ansible | 16:36 | |
| cloudnull | evrardjp: no need from me, good change though. :) | 16:36 |
| evrardjp | swappiness is damn important, but I'm not sure in which role it would be best to put it... a common one? ;) | 16:37 |
| evrardjp | or just let the deployer choose. | 16:37 |
| gparaskevas | swappiness regarding every host? | 16:38 |
| *** _hanhart has quit IRC | 16:39 | |
| linggao | Hi mrda, I saw you are working on ironc https://github.com/mrda/openstack-ansible-ironic now and also read your design. How far is it from being useful? | 16:42 |
| evrardjp | gparaskevas: yes | 16:44 |
| evrardjp | define just something that defines the swapiness | 16:44 |
| evrardjp | maybe with a variable depending on the group | 16:44 |
| evrardjp | defaulting to something really low | 16:44 |
| evrardjp | but that's not really a priority | 16:44 |
| evrardjp | :p | 16:44 |
| *** mss has quit IRC | 16:45 | |
| *** fawadkhaliq has quit IRC | 16:45 | |
| *** alextricity has quit IRC | 16:45 | |
| *** jlvillal has quit IRC | 16:45 | |
| *** thrawn01 has quit IRC | 16:45 | |
| *** e-vad has quit IRC | 16:45 | |
| *** mattoliverau has quit IRC | 16:45 | |
| *** SamYaple has quit IRC | 16:45 | |
| *** mrda has quit IRC | 16:45 | |
| *** charz has quit IRC | 16:45 | |
| *** rev_dr has quit IRC | 16:45 | |
| *** Sam-I-Am has quit IRC | 16:45 | |
| *** rcarrillocruz has quit IRC | 16:45 | |
| *** ysm_ has quit IRC | 16:45 | |
| *** baker has quit IRC | 16:45 | |
| *** leifmadsen has quit IRC | 16:45 | |
| *** Klev-meeting has quit IRC | 16:45 | |
| *** mancdaz has quit IRC | 16:45 | |
| *** tiagogomes has quit IRC | 16:45 | |
| *** mrodden has quit IRC | 16:45 | |
| *** kore has quit IRC | 16:45 | |
| *** elo has quit IRC | 16:45 | |
| *** Tridde has quit IRC | 16:45 | |
| *** darrenc has quit IRC | 16:45 | |
| *** niknakpaddywak has quit IRC | 16:45 | |
| *** toan has quit IRC | 16:45 | |
| *** McMurlock has quit IRC | 16:45 | |
| *** mattt has quit IRC | 16:45 | |
| *** Guest4784 has quit IRC | 16:45 | |
| *** kysse has quit IRC | 16:45 | |
| *** tlbr has quit IRC | 16:45 | |
| evrardjp | another round of netsplits! | 16:45 |
| *** gparaskevas has quit IRC | 16:45 | |
| *** ysm_ has joined #openstack-ansible | 16:46 | |
| *** rcarrillocruz has joined #openstack-ansible | 16:46 | |
| *** rev_dr has joined #openstack-ansible | 16:46 | |
| *** charz has joined #openstack-ansible | 16:46 | |
| *** mrda has joined #openstack-ansible | 16:46 | |
| *** SamYaple has joined #openstack-ansible | 16:46 | |
| *** mattoliverau has joined #openstack-ansible | 16:46 | |
| *** e-vad has joined #openstack-ansible | 16:46 | |
| *** thrawn01 has joined #openstack-ansible | 16:46 | |
| *** jlvillal has joined #openstack-ansible | 16:46 | |
| *** Sam-I-Am has joined #openstack-ansible | 16:46 | |
| *** alextricity has joined #openstack-ansible | 16:46 | |
| *** fawadkhaliq has joined #openstack-ansible | 16:46 | |
| *** mss has joined #openstack-ansible | 16:46 | |
| *** baker has joined #openstack-ansible | 16:46 | |
| *** leifmadsen has joined #openstack-ansible | 16:46 | |
| *** Klev-meeting has joined #openstack-ansible | 16:46 | |
| *** mancdaz has joined #openstack-ansible | 16:46 | |
| *** tiagogomes has joined #openstack-ansible | 16:46 | |
| *** mrodden has joined #openstack-ansible | 16:46 | |
| *** kysse has joined #openstack-ansible | 16:46 | |
| *** kore has joined #openstack-ansible | 16:46 | |
| *** elo has joined #openstack-ansible | 16:46 | |
| *** Tridde has joined #openstack-ansible | 16:46 | |
| *** darrenc has joined #openstack-ansible | 16:46 | |
| *** niknakpaddywak has joined #openstack-ansible | 16:46 | |
| *** toan has joined #openstack-ansible | 16:46 | |
| *** McMurlock has joined #openstack-ansible | 16:46 | |
| *** mattt has joined #openstack-ansible | 16:46 | |
| *** Guest4784 has joined #openstack-ansible | 16:46 | |
| *** tlbr has joined #openstack-ansible | 16:46 | |
| -cloudnull- Freenode is having issues today: https://twitter.com/freenodestaff?ref_src=twsrc%5Egoogle%7Ctwcamp%5Eserp%7Ctwgr%5Eauthor -- if you have a message missed or otherwise ignored please repost it, its likely the message was never seen | 16:48 | |
| *** sdake has joined #openstack-ansible | 16:51 | |
| cloudnull | evrardjp: sorry was in another meeting . the swappieness is in openstack_hosts | 16:52 |
| cloudnull | https://github.com/openstack/openstack-ansible-openstack_hosts/blob/master/defaults/main.yml#L93 | 16:52 |
| odyssey4me | prometheanfire the ldap plugins - honestly I have no idea right now - I have yet to look at any of it | 16:53 |
| evrardjp | ok it could make sense to add it on lxc hosts too then | 16:53 |
| cloudnull | i guess we could | 16:54 |
| prometheanfire | odyssey4me: just not sure why that was one of the instructions | 16:54 |
| cloudnull | i'd keep the value between the openstack_hosts and lxc_hosts roles the same so it doesnt flip flop | 16:55 |
| evrardjp | cloudnull: roger that. Will try do it tomorrow. | 16:55 |
| cloudnull | idk if we should have the value be 5 or 10 | 16:55 |
| cloudnull | but ill leave that up to you | 16:55 |
| evrardjp | yeah like this is gonna make a huge difference | 16:55 |
| evrardjp | :p | 16:55 |
| odyssey4me | linggao mrda is in Australia, and therefore likely asleep at this time. :) | 16:56 |
| *** baker has quit IRC | 17:01 | |
| *** baker has joined #openstack-ansible | 17:02 | |
| *** ChanServ has joined #openstack-ansible | 17:05 | |
| *** wolfe.freenode.net sets mode: +o ChanServ | 17:05 | |
| *** Mudpuppy has joined #openstack-ansible | 17:05 | |
| *** mgariepy has quit IRC | 17:06 | |
| *** mgariepy has joined #openstack-ansible | 17:06 | |
| *** Mudpuppy is now known as Mudpuppy_awa | 17:06 | |
| *** baker has quit IRC | 17:06 | |
| *** adaccada has quit IRC | 17:08 | |
| *** Mudpuppy_awa is now known as Mudpuppy | 17:11 | |
| linggao | odyssey4me, thanks let me know. :-). | 17:14 |
| *** sacharya has joined #openstack-ansible | 17:18 | |
| *** phiche has quit IRC | 17:19 | |
| *** eil397 has quit IRC | 17:23 | |
| *** woodard has joined #openstack-ansible | 17:24 | |
| *** hassaan has joined #openstack-ansible | 17:24 | |
| *** phschwartz_ is now known as phschwartz | 17:25 | |
| *** sacharya_ has joined #openstack-ansible | 17:26 | |
| *** woodard has quit IRC | 17:28 | |
| *** k_stev has quit IRC | 17:28 | |
| *** k_stev1 has joined #openstack-ansible | 17:28 | |
| *** k_stev1 has quit IRC | 17:28 | |
| *** sacharya has quit IRC | 17:29 | |
| *** egonzalez has quit IRC | 17:29 | |
| *** k_stev has joined #openstack-ansible | 17:29 | |
| *** eil397 has joined #openstack-ansible | 17:30 | |
| *** baker has joined #openstack-ansible | 17:31 | |
| *** k_stev has quit IRC | 17:32 | |
| *** alkari has quit IRC | 17:32 | |
| *** k_stev has joined #openstack-ansible | 17:32 | |
| *** hassaan has quit IRC | 17:33 | |
| *** targon has joined #openstack-ansible | 17:33 | |
| *** targon has quit IRC | 17:36 | |
| *** Klev-meeting is now known as KLevenstein | 17:39 | |
| *** shausy has joined #openstack-ansible | 17:40 | |
| *** hassaan has joined #openstack-ansible | 17:45 | |
| *** b3rnard0 is now known as b3rnard0_away | 17:45 | |
| *** hassaan has quit IRC | 17:51 | |
| *** javeriak_ has joined #openstack-ansible | 17:52 | |
| *** Bjoern_ has joined #openstack-ansible | 17:52 | |
| *** hassaan has joined #openstack-ansible | 17:54 | |
| *** shausy has quit IRC | 17:54 | |
| *** javeriak has quit IRC | 17:56 | |
| *** shausy has joined #openstack-ansible | 17:56 | |
| *** baker has quit IRC | 17:57 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 17:57 | |
| *** tiagogomes has quit IRC | 17:57 | |
| *** alkari has joined #openstack-ansible | 18:00 | |
| *** shausy has quit IRC | 18:00 | |
| *** shausy has joined #openstack-ansible | 18:00 | |
| *** hassaan has quit IRC | 18:01 | |
| *** galstrom_zzz is now known as galstrom | 18:04 | |
| *** sdake has quit IRC | 18:06 | |
| *** sacharya_ has quit IRC | 18:10 | |
| *** harlowja has joined #openstack-ansible | 18:12 | |
| *** sacharya has joined #openstack-ansible | 18:13 | |
| *** hassaan has joined #openstack-ansible | 18:13 | |
| *** Bjoern_ is now known as BjoernT | 18:14 | |
| *** rebase_ has joined #openstack-ansible | 18:20 | |
| *** b3rnard0_away is now known as b3rnard0 | 18:23 | |
| -cloudnull- Freenode is having issues today: https://twitter.com/freenodestaff?ref_src=twsrc%5Egoogle%7Ctwcamp%5Eserp%7Ctwgr%5Eauthor -- if you have a message missed or otherwise ignored please repost it, its likely the message was never seen | 18:24 | |
| *** alkari has quit IRC | 18:34 | |
| *** hassaan_ has joined #openstack-ansible | 18:36 | |
| cloudnull | kysse: so liberty is all happy now ? | 18:38 |
| cloudnull | was going to ask earlier however IRC was not happy | 18:38 |
| cloudnull | any core arond to do a review on https://review.openstack.org/#/c/241483/ ? | 18:40 |
| *** hassaan has quit IRC | 18:40 | |
| cloudnull | odyssey4me mattt hughsaunders andymccr d34dh0r53 ? ^ | 18:40 |
| cloudnull | stevelle: ty for the review already | 18:40 |
| *** bryan_att has quit IRC | 18:41 | |
| cloudnull | odyssey4me: https://review.openstack.org/#/c/252100/ | 18:44 |
| cloudnull | cores we need that to go in too ^ | 18:44 |
| cloudnull | large deployments of liberty are effectively broken unless neutron_l2_population is enabled. | 18:45 |
| *** ysm_ has quit IRC | 18:45 | |
| *** ysm_ has joined #openstack-ansible | 18:46 | |
| *** adaccada has joined #openstack-ansible | 18:48 | |
| *** fawadkhaliq has quit IRC | 18:49 | |
| *** fawadkhaliq has joined #openstack-ansible | 18:49 | |
| *** ysm_ has quit IRC | 18:50 | |
| *** k_stev has quit IRC | 18:55 | |
| odyssey4me | cloudnull let me understand this properly | 18:56 |
| odyssey4me | if l2pop is enabled, l3ha is unusable | 18:56 |
| cloudnull | correct | 18:56 |
| odyssey4me | however if l3ha is enabled, l2pop must be disabled - rendering neutron unusable | 18:56 |
| *** javeriak_ has quit IRC | 18:56 | |
| *** hassaan_ has quit IRC | 18:57 | |
| Sam-I-Am | i thought l3ha and l2pop got along in liberty | 18:57 |
| Sam-I-Am | its l2pop and dvr that do not | 18:57 |
| odyssey4me | is this disaster a special case for when we only have one agents container? | 18:57 |
| cloudnull | Sam-I-Am: nope | 18:57 |
| cloudnull | the issue comes up when there are multiple agents , a single l3 agent everything is ok | 18:57 |
| Sam-I-Am | well, l3ha with only one agent is not ha | 18:58 |
| odyssey4me | so if l2pop is disabled, and l3ha enabled, and there are multiple agents - then everything is fine? | 18:58 |
| odyssey4me | and conversely if l2pop is enabled, l3ha disabled and there is one agent - then everything is fine? | 18:58 |
| cloudnull | odyssey4me: and there is something eles on the backend that is doing l2 population for the deployment yes | 18:58 |
| cloudnull | to the first one | 18:59 |
| cloudnull | l2 pop enabled, l3ha disabled works fine for multiple network nodes | 18:59 |
| Sam-I-Am | i'm looking for the bug about this | 18:59 |
| odyssey4me | oh, so l3ha is thus pretty much broken then.... unless we implement something to replace l2pop | 18:59 |
| *** bryan_att has joined #openstack-ansible | 18:59 | |
| cloudnull | ++ | 19:00 |
| *** BjoernT has quit IRC | 19:00 | |
| cloudnull | this would be the thrid release where l3ha is functionally unusable. | 19:00 |
| cloudnull | which is why we still have the att neutron tool | 19:00 |
| odyssey4me | cloudnull in that case we should disable the automagic stuff and let the deployer decide - let's leave l3ha disabled by default and l2pop enabled by default | 19:01 |
| cloudnull | Sam-I-Am: if it works, then theres something else going on, because kysse beat on this for several days and its busted. | 19:01 |
| Sam-I-Am | we enabled support for l3ha in liberty, right? | 19:02 |
| Sam-I-Am | i seem to recall this patch | 19:02 |
| odyssey4me | before we do that, though, I'd like us to engage with some neutron devs to figure this one out | 19:02 |
| odyssey4me | Sam-I-Am yes | 19:02 |
| *** baker has joined #openstack-ansible | 19:02 | |
| Sam-I-Am | odyssey4me cloudnull - https://bugs.launchpad.net/neutron/+bug/1365476 | 19:03 |
| openstack | Launchpad bug 1365476 in neutron "HA routers interact badly with l2pop" [High,Fix released] - Assigned to Mike Kolesnik (mkolesni) | 19:03 |
| *** baker has quit IRC | 19:03 | |
| *** Bjoern_ has joined #openstack-ansible | 19:03 | |
| odyssey4me | so this seems to imply that l2pop and l3ha can work togetheR? | 19:04 |
| *** phiche has joined #openstack-ansible | 19:04 | |
| Sam-I-Am | that is the idea | 19:05 |
| odyssey4me | specifically part of the commit message: This patch also changes the L3 HA failover test to use l2pop. Note that the test does not pass when using l2pop without this patch. | 19:05 |
| Sam-I-Am | but my l3ha guy is afk right now | 19:05 |
| *** baker has joined #openstack-ansible | 19:05 | |
| odyssey4me | mhayden ping | 19:06 |
| *** Bjoern_ is now known as BjoernT | 19:06 | |
| mhayden | ohai | 19:06 |
| Sam-I-Am | odyssey4me: that commit message makes sense | 19:06 |
| Sam-I-Am | at least to me | 19:06 |
| Sam-I-Am | although does it mean that l3ha does not work without l2pop | 19:07 |
| * Sam-I-Am sticks head in sand | 19:07 | |
| Sam-I-Am | aios only have 1 agents container, right? | 19:08 |
| cloudnull | Sam-I-Am: correct | 19:08 |
| cloudnull | which is why we didnt see the issue in the gate | 19:08 |
| cloudnull | kysse: had 2 network nodes which caused the issue | 19:08 |
| mhayden | odyssey4me: echo reply | 19:09 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 19:09 | |
| Sam-I-Am | there's a lot of patches that the gate doesnt check | 19:09 |
| cloudnull | we can do an affinity group for agent containers and test to make sure l2 pop is in fact working when l3ha is enabled | 19:09 |
| Sam-I-Am | not specifically ours, but neutrons | 19:09 |
| Sam-I-Am | that was my next thought | 19:09 |
| cloudnull | and that l3ha is working/the stack can pass tempest | 19:09 |
| Sam-I-Am | although i also have a non-OA lab that might work | 19:10 |
| mhayden | odyssey4me: last AIO ansible failure is peculiar -- it seems we have a new nodepool in bluebox's cloud?! | 19:11 |
| mhayden | http://logs.openstack.org/25/239525/70/check/gate-openstack-ansible-dsvm-commit/0db4c51/console.html#_2015-12-02_17_54_45_469 | 19:12 |
| odyssey4me | mhayden yep, another new provider is ovh | 19:12 |
| Sam-I-Am | hopefully they suck less | 19:13 |
| mhayden | seems like bluebox might not have a lot of base disk space nor extra disks from what i see | 19:13 |
| mhayden | (initial look) | 19:13 |
| *** javeriak has joined #openstack-ansible | 19:14 | |
| *** BjoernT is now known as Bjoern_zZzZzZzZ | 19:15 | |
| mhayden | it looks like bare-trusty will ensure we get on hp/rax | 19:17 |
| mhayden | https://git.openstack.org/cgit/openstack-infra/project-config/tree/nodepool/nodepool.yaml | 19:17 |
| *** Bjoern_zZzZzZzZ is now known as BjoernT | 19:18 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Gate: Make instance info output to stdout on failure https://review.openstack.org/252247 | 19:18 |
| odyssey4me | mhayden ^ will help us find that out | 19:19 |
| *** ysm has joined #openstack-ansible | 19:20 | |
| *** mancdaz has quit IRC | 19:25 | |
| odyssey4me | cloudnull it seems to me that either we need to enable both l3ha and l2pop if the number of agents>1, and we need to disable l3ha if # agents == 1 | 19:25 |
| Sam-I-Am | why is adding a distro package to a bug in launchpad such a bitch | 19:25 |
| Sam-I-Am | odyssey4me: or test this before we do things? | 19:25 |
| odyssey4me | Sam-I-Am they see you ridin', they hatin' | 19:25 |
| Sam-I-Am | like disabling arp spoof protection? | 19:26 |
| odyssey4me | Sam-I-Am eh? | 19:26 |
| *** mancdaz has joined #openstack-ansible | 19:26 | |
| *** adaccada has quit IRC | 19:27 | |
| Sam-I-Am | i thought we disable arp spoof protection by default in liberty because reasons | 19:27 |
| Sam-I-Am | something i think i was supposed to look into, but time and unicorns | 19:27 |
| *** baker has quit IRC | 19:28 | |
| *** fawadkhaliq has quit IRC | 19:28 | |
| odyssey4me | Sam-I-Am it got disabled by default because it resulted in severe brokenness | 19:28 |
| *** fawadkhaliq has joined #openstack-ansible | 19:28 | |
| odyssey4me | while we could build, we could never pass a full tempest test | 19:28 |
| Sam-I-Am | which is odd, because i havent seen that here | 19:28 |
| odyssey4me | Sam-I-Am you perhaps forget that I worked with you to cover every single avenue that time permitted | 19:29 |
| odyssey4me | I am happy to revisit that, but I can confidently tell you that it broke networking. | 19:29 |
| odyssey4me | Note also that there is no linuxbridge voted testing in the gate upstream. | 19:30 |
| *** BjoernT is now known as Bjoern_zZzZzZzZ | 19:30 | |
| *** Bjoern_zZzZzZzZ is now known as BjoernT | 19:31 | |
| odyssey4me | Also it seemed to me at the time that I looked into it that upstream does not do floating address testing in the gate for integrated tests... which we do. | 19:31 |
| odyssey4me | The blocker for upstream linuxbridge gate testing merged yesterday, finally. The next step will be to revisit the patch to make it voting. | 19:32 |
| Sam-I-Am | do you recall which tests failed? i have fips with arp spoof in all of my labs and nothing obvious breaks | 19:32 |
| *** shausy has quit IRC | 19:32 | |
| *** sc68cal has joined #openstack-ansible | 19:32 | |
| odyssey4me | the tempest test for floating ip access failed | 19:32 |
| *** niknakpaddywak has quit IRC | 19:32 | |
| Sam-I-Am | odyssey4me: meet sc68cal | 19:32 |
| Sam-I-Am | who is working on the lb gate job | 19:33 |
| sc68cal | bear with me - let me read the eavesdrop log for this channel to catch up | 19:33 |
| *** maximov_ has quit IRC | 19:33 | |
| *** etoews has quit IRC | 19:33 | |
| *** serverascode has quit IRC | 19:33 | |
| *** meteorfox has quit IRC | 19:33 | |
| odyssey4me | haha, sc68cal is the owner of https://review.openstack.org/205674 :) | 19:33 |
| sc68cal | yep | 19:34 |
| odyssey4me | it's time to revise the patch, as the dependent patch has merged - finally | 19:34 |
| odyssey4me | much bike shedding was had | 19:34 |
| Sam-I-Am | sc68cal: tl;dr - we had to disable arp spoof protection (even with port_security ext) in liberty due to issues of some sort | 19:34 |
| Sam-I-Am | this sort of came out of a 'does l3ha work with l2pop yet' discussion | 19:34 |
| sc68cal | ok I've caught up on today's scrollback | 19:35 |
| sc68cal | odyssey4me: oh did it???? | 19:35 |
| sc68cal | PRAISE THE SUN | 19:35 |
| sc68cal | PRRAAAIIISSSEEEE THE SUUUUUUNN | 19:35 |
| odyssey4me | hahaha | 19:36 |
| odyssey4me | yep, it is done - time for some rebase magic to prep for the next bike shed | 19:36 |
| sc68cal | odyssey4me: ack. Let's let it cook for a bit and see what our pass rate gets to with that fix in | 19:37 |
| sc68cal | since reason I set it as a depends-on in the voting patch was so I don't get hit with a wall of hate | 19:37 |
| sc68cal | when LB job starts failing at the gate | 19:37 |
| odyssey4me | sc68cal sure, happy to let it sit for a bit - that patch set should probably also be backported to liberty at least | 19:38 |
| sc68cal | odyssey4me: good point | 19:38 |
| *** baker has joined #openstack-ansible | 19:39 | |
| odyssey4me | sc68cal so we found that arp spoofing broke our integrated gate check consistently | 19:40 |
| odyssey4me | our build uses linuxbridge | 19:40 |
| sc68cal | odyssey4me: what's the job name? | 19:40 |
| odyssey4me | sc68cal which job - one that has arp spoofing enabled? | 19:41 |
| *** niknakpaddywak has joined #openstack-ansible | 19:41 | |
| sc68cal | odyssey4me:right | 19:41 |
| odyssey4me | I can whip a review up quickly to re-test all that. I haven't tried it since just before Liberty release. | 19:42 |
| sc68cal | odyssey4me: ok - but you said "our integrated gate check" - who is our | 19:42 |
| Sam-I-Am | openstack-ansible | 19:42 |
| odyssey4me | our = OpenStack-Ansible | 19:42 |
| odyssey4me | sc68cal ie https://github.com/openstack/openstack-ansible | 19:43 |
| *** adaccada has joined #openstack-ansible | 19:43 | |
| * sc68cal looks in project-config | 19:43 | |
| odyssey4me | sc68cal lemme whip up a test review | 19:44 |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: [WIP] Enable Neutron prevent_arp_spoofing https://review.openstack.org/252574 | 19:46 |
| *** sdake has joined #openstack-ansible | 19:46 | |
| * sc68cal looks | 19:46 | |
| *** luckyinva has joined #openstack-ansible | 19:48 | |
| sc68cal | odyssey4me: ok I think I'm starting to follow now :) | 19:48 |
| sc68cal | odyssey4me: is there a bug on the neutron side already for this issue? if not, and your patch comes back and explodes let's go ahead and make a bug on the neutron side to help track | 19:49 |
| odyssey4me | sc68cal I would be absolutely happy to orientate you as necessary. Ping me any time. | 19:49 |
| sc68cal | Sam-I-Am: thanks for the introduction | 19:49 |
| *** sacharya has quit IRC | 19:49 | |
| odyssey4me | sc68cal sure, can do that - and I'm happy to facilitate discussions between whichever parties to resolve some of the issues we pick up | 19:50 |
| sc68cal | odyssey4me: cool. Once CI reports back on that patch we can use that as an artifact if we open a bug on the neutron side | 19:51 |
| Sam-I-Am | sc68cal: sure | 19:51 |
| odyssey4me | sc68cal grab a coffee, it'll take around an hour once it kicks off | 19:51 |
| Sam-I-Am | (or beer) | 19:52 |
| Sam-I-Am | odyssey4me: heard from the l3ha guy. l2pop and l3ha should work together in kilo and liberty, but use of l2pop and l3ha is not recommended until further performance refinements | 19:53 |
| Sam-I-Am | sc68cal: o-a is probably one of the larger deployment tools using lb | 19:55 |
| Sam-I-Am | multinode-ish testing is already possible on a single node with affinity groups | 19:55 |
| Sam-I-Am | now if only i can get my ovs testing done on it :) | 19:55 |
| *** markvoelker has quit IRC | 19:55 | |
| odyssey4me | Sam-I-Am (cc cloudnull) so what I'm hearing is that we should enable l2pop and disable l3ha by default, and add some documentation around the known issue - is there a bug/blueprint for the improvements to fix performance? | 19:56 |
| odyssey4me | it'd be good to know at what scale this is an issue | 19:56 |
| odyssey4me | Sam-I-Am if you can find a sponsor who cares about OVS, the patch is quite simple :) we just need someone who cares enough and understands OVS enough... because we don't | 19:57 |
| Sam-I-Am | i think its deployment-specific | 19:57 |
| Sam-I-Am | vlan-type networks can't use l2pop, so there's much less of an issue with l3ha there | 19:57 |
| Sam-I-Am | also only applies to installs with project networks in the first place | 19:58 |
| *** rackertom has quit IRC | 19:59 | |
| *** markvoelker has joined #openstack-ansible | 19:59 | |
| alextricity | I literally just got done turning on l2_population on my mutli-neutron-server deployment | 20:01 |
| alextricity | I should of grabbed the error.. | 20:01 |
| odyssey4me | alextricity your finding? | 20:02 |
| alextricity | When l2_pop was disabled..I was getting a port binding error | 20:02 |
| alextricity | I didn't really look into though | 20:02 |
| * alextricity regrets it | 20:02 | |
| alextricity | I just turned on l2_pop and it seemed to fix it | 20:02 |
| odyssey4me | yep, that's the same experience that was being discussed over the last few days | 20:02 |
| *** sacharya has joined #openstack-ansible | 20:02 | |
| odyssey4me | alextricity we need to definitively determine whether l2pop must be enabled when l3ha is enabled - it all relates to https://review.openstack.org/252100 (see the review comments) | 20:05 |
| *** BjoernT has quit IRC | 20:05 | |
| *** luckyinva has quit IRC | 20:05 | |
| alextricity | I was looking at that earlier today. I was about to +1 one it until I saw your comments | 20:05 |
| *** sdake has quit IRC | 20:05 | |
| alextricity | The changed seemed to fix my port binding issues | 20:06 |
| alextricity | I can look back to see what error I was getting exactly | 20:06 |
| *** etoews has joined #openstack-ansible | 20:07 | |
| odyssey4me | perhaps l3ha should only be enabled if #agents>1, but l2pop should always be enabled | 20:08 |
| *** spotz is now known as spotz_zzz | 20:08 | |
| odyssey4me | but perhaps the decision should lie with the deployer - and those are just our defaults | 20:08 |
| *** sdake has joined #openstack-ansible | 20:08 | |
| alextricity | That'll definitely do it. However, l3ha still wouldn't be working in it's currently state, right? At least that's what i'm getting at after reading the patch set | 20:10 |
| alextricity | modprobe: ERROR: could not insert 'ip_vs': Operation not permitted | 20:11 |
| alextricity | neutron-ha-tool: ERROR <html><body><h1>503 Service Unavailable</h1>#012No server is available to handle this request.#012</body></html> | 20:11 |
| alextricity | ERROR neutron.plugins.ml2.managers [req-32a1ea1e-aedb-4528-a9c5-3dc17f751a65 - - - - -] Failed to bind port 20cc04e0-fdc3-40a8-a8e3-8023866427ff on host osa-infra3_neutron_agents_container-40a619db | 20:11 |
| alextricity | Unfortunately, the only thing that doesn't go over my head is the neutron-ha-tool 503 error | 20:14 |
| *** sdake has quit IRC | 20:14 | |
| odyssey4me | alextricity the ha tool only handles the non-ha routers, as I understand it | 20:15 |
| *** serverascode has joined #openstack-ansible | 20:16 | |
| *** revdr has joined #openstack-ansible | 20:16 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 20:20 | |
| *** sdake has joined #openstack-ansible | 20:21 | |
| *** njohnston has joined #openstack-ansible | 20:21 | |
| *** meteorfox has joined #openstack-ansible | 20:29 | |
| *** javeriak has quit IRC | 20:32 | |
| *** k_stev has joined #openstack-ansible | 20:33 | |
| *** rackertom has joined #openstack-ansible | 20:34 | |
| Sam-I-Am | odyssey4me: does it break if it sees an ha router? | 20:38 |
| odyssey4me | Sam-I-Am it's been patched ot ignore ha routers | 20:38 |
| odyssey4me | *to | 20:38 |
| Sam-I-Am | mmkay | 20:40 |
| *** hybridpollo has joined #openstack-ansible | 20:42 | |
| *** spotz_zzz is now known as spotz | 20:43 | |
| *** maximov_ has joined #openstack-ansible | 20:44 | |
| *** k_stev has quit IRC | 20:44 | |
| *** spotz is now known as spotz_zzz | 20:47 | |
| *** git-harry has quit IRC | 20:49 | |
| *** fawadkhaliq has quit IRC | 20:49 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Convert AIO bootstrap from bash to Ansible https://review.openstack.org/239525 | 20:50 |
| *** git-harry has joined #openstack-ansible | 20:50 | |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Check mode compatibility for security role https://review.openstack.org/245813 | 20:50 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Updating tests for openstack-ansible-security https://review.openstack.org/251430 | 20:50 |
| openstackgerrit | Major Hayden proposed openstack/openstack-ansible-security: Adjusting commonly failing tasks https://review.openstack.org/252009 | 20:51 |
| odyssey4me | mhayden I'm dropping the absolute AIO minimum space on / to 50GB, as that's what seems to be required for gate checks | 20:57 |
| mhayden | good stuff | 20:57 |
| *** baker has quit IRC | 21:00 | |
| *** phiche has quit IRC | 21:02 | |
| *** phiche has joined #openstack-ansible | 21:02 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 21:03 | |
| *** Bjoern_ has joined #openstack-ansible | 21:06 | |
| Sam-I-Am | bbiab | 21:06 |
| *** adaccada has quit IRC | 21:19 | |
| *** spotz_zzz is now known as spotz | 21:30 | |
| *** Bjoern_ is now known as Bjoern_zZzZzZzZ | 21:30 | |
| *** Bjoern_zZzZzZzZ is now known as Bjoern_ | 21:33 | |
| openstackgerrit | Merged openstack/openstack-ansible: Add documentation for HA ceilometer https://review.openstack.org/248905 | 21:33 |
| *** galstrom is now known as galstrom_zzz | 21:36 | |
| *** dolphm is now known as dolphm_418 | 21:38 | |
| *** ctina_ has joined #openstack-ansible | 21:45 | |
| *** Bjoern_ is now known as Bjoern_zZzZzZzZ | 21:49 | |
| *** ctina has quit IRC | 21:49 | |
| *** Bjoern_zZzZzZzZ is now known as Bjoern_ | 21:49 | |
| *** ctina_ has quit IRC | 21:50 | |
| *** loquacities has joined #openstack-ansible | 21:52 | |
| *** Bjoern_ has quit IRC | 21:54 | |
| *** KLevenstein has quit IRC | 21:54 | |
| *** Bjoern_ has joined #openstack-ansible | 21:56 | |
| odyssey4me | alextricity please backport https://review.openstack.org/248905 to liberty | 21:58 |
| odyssey4me | Sam-I-Am sc68cal and would you look at that, https://review.openstack.org/252574 passed the gate check :) | 22:01 |
| alextricity | odyssey4me: Sure | 22:01 |
| odyssey4me | alextricity good patch :) | 22:02 |
| openstackgerrit | Miguel Alex Cantu proposed openstack/openstack-ansible: Add documentation for HA ceilometer https://review.openstack.org/252651 | 22:02 |
| alextricity | Thanks! | 22:03 |
| odyssey4me | stevelle FYI, not sure if you're aware of this but alextricity has contributed most of the Ceilometer related stuff into OSA so far. If you haven't met before, consider this an intro. :) | 22:03 |
| odyssey4me | thanks alextricity for yet another good patch :) | 22:03 |
| alextricity | (wave) | 22:03 |
| odyssey4me | mcarden welcome :) | 22:05 |
| odyssey4me | iceyao ping? | 22:05 |
| *** Mudpuppy has quit IRC | 22:06 | |
| *** phiche has quit IRC | 22:07 | |
| mcarden | odyssey4me: Thanks | 22:11 |
| cloudnull | odyssey4me: sorry was in meetings for the most of the day | 22:14 |
| *** baker has joined #openstack-ansible | 22:24 | |
| openstackgerrit | Jesse Pretorius proposed openstack/openstack-ansible: Use PyPi packages for ceph python bindings https://review.openstack.org/245154 | 22:24 |
| *** mancdaz has quit IRC | 22:30 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: FIX: provider_networks module for multiple vlans https://review.openstack.org/252658 | 22:30 |
| *** mancdaz has joined #openstack-ansible | 22:31 | |
| *** eil397 has quit IRC | 22:38 | |
| *** Bjoern_ is now known as BjoernT | 22:40 | |
| stevelle | we've talked a couple of times, I believe alextricity | 22:43 |
| *** mss_ has joined #openstack-ansible | 22:45 | |
| *** mss has quit IRC | 22:48 | |
| *** sacharya_ has joined #openstack-ansible | 22:49 | |
| *** eil397 has joined #openstack-ansible | 22:49 | |
| *** rromans has quit IRC | 22:52 | |
| *** sacharya has quit IRC | 22:52 | |
| *** njohnston has quit IRC | 22:53 | |
| *** hybridpollo has quit IRC | 22:55 | |
| odyssey4me | mhayden so it appears that we have some flavours in openstack-ci with 75G disks now: http://logs.openstack.org/25/239525/71/check/gate-openstack-ansible-dsvm-commit/f7679a8/logs/instance-info/host_info_1449092126.log | 22:56 |
| *** rromans has joined #openstack-ansible | 22:56 | |
| odyssey4me | so that's one more issue down - now to try and figure out the next issue | 22:57 |
| linggao | Hi mrda | 22:57 |
| odyssey4me | mhayden the last check failed with a build timeout - waiting for 15mins for the swift proxy container to be alive :/ | 22:58 |
| *** hybridpollo has joined #openstack-ansible | 22:59 | |
| linggao | odyssey4me, do you when is best time to talk to mrda? | 23:00 |
| odyssey4me | linggao I expect that he's alive - not sure whether he's stuck in meetings or something | 23:00 |
| linggao | ok. I will try later. thanks. | 23:01 |
| odyssey4me | linggao do you have a running openstack-ansible setup? | 23:01 |
| linggao | odyssey4me, yes a all-in-one | 23:01 |
| odyssey4me | linggao from what I understand, ironic needs ipmi access to host bmc devices... this is usually the blocker for testing | 23:02 |
| odyssey4me | and this is an issue we're trying to work out | 23:02 |
| odyssey4me | we want to be able to gate test it, and this is not easy | 23:03 |
| odyssey4me | ideally we need a sponsor for some sort of external CI I think | 23:03 |
| kysse | Our galera gluster went borken, cluster size is now 1. We booted galera nodes, first node 1 and after it came up we booted node 2. | 23:04 |
| kysse | what to do? Only node 2 is up and running now, when I launch node 1 it says like 10 secs that cluster size is 2, and then it fails. | 23:05 |
| linggao | odyssey4me, I can help with the initial test if needed. I have been working on ironic. But I do not think I have time to CI, setting up a CI takes time. We did it for ironic before. | 23:06 |
| odyssey4me | linggao ah, so you know ironic then - that helps from a review standpoint | 23:09 |
| linggao | odyssey4me, yes. | 23:11 |
| sc68cal | odyssey4me: it's a christmas miracle | 23:11 |
| sc68cal | odyssey4me: I know that there was some work done recently on arp spoofing - I just don't remember if it was OVS only or LB only | 23:12 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 23:12 | |
| odyssey4me | sc68cal this is why we can't have nice things :p | 23:12 |
| odyssey4me | sc68cal with a voting lb gate, we could have nice things :) | 23:12 |
| sc68cal | odyssey4me: haha - ok ok I'll rebase the patch tomorrow. promise | 23:13 |
| odyssey4me | sc68cal I'll buy you a C.H.I.P. if you get it voted through by the end of the week. :) | 23:15 |
| * Sam-I-Am appear | 23:16 | |
| sc68cal | odyssey4me: no idea what a CHIP is - is it beer? | 23:17 |
| Sam-I-Am | sc68cal: its a thing that goes well with beer | 23:17 |
| *** baker has quit IRC | 23:18 | |
| odyssey4me | sc68cal http://getchip.com/ - with beer, it runs openstack | 23:18 |
| * sc68cal googles | 23:18 | |
| sc68cal | oh. I already have 2 RasPis | 23:18 |
| sc68cal | I'll take a $3 beer at the summit - that saves you 5 DOLLARS! | 23:18 |
| odyssey4me | sc68cal and now you could have diversity :) | 23:19 |
| Sam-I-Am | with beer anything runs openstack | 23:19 |
| odyssey4me | sc68cal I'll buy you a scotch, or a beer as a fall back. | 23:19 |
| sc68cal | odyssey4me: deal. :) | 23:19 |
| *** ysm has quit IRC | 23:20 | |
| sc68cal | Sam-I-Am: we need to make that a success thingy that the IRC bot picks up | 23:20 |
| sc68cal | Sam-I-Am: https://wiki.openstack.org/wiki/Successes | 23:20 |
| *** baker has joined #openstack-ansible | 23:21 | |
| Sam-I-Am | sc68cal: ha, didnt know that existed | 23:21 |
| Sam-I-Am | i know cloudnull wrote a successerator function | 23:21 |
| *** baker has quit IRC | 23:21 | |
| stevelle | just saw it used over in #openstack-glance | 23:22 |
| stevelle | pretty neat | 23:22 |
| odyssey4me | how does it work? #magic ? | 23:22 |
| odyssey4me | heh: http://docs.openstack.org/infra/system-config/irc.html#statusbot | 23:24 |
| odyssey4me | #success I just worked out how to make successbot work! :) | 23:24 |
| openstackstatus | odyssey4me: Added success to Success page | 23:24 |
| Sam-I-Am | haha | 23:24 |
| Sam-I-Am | success - open source didnt kill me today | 23:25 |
| Sam-I-Am | then again no one could agree on a patch to accomplish that task :) | 23:25 |
| odyssey4me | alright, on that note - I'm out for the night | 23:25 |
| Sam-I-Am | odyssey4me: scotch | 23:26 |
| odyssey4me | Sam-I-Am that'd be PS4-ish :) | 23:27 |
| Sam-I-Am | heh | 23:28 |
| *** jaypipes has quit IRC | 23:38 | |
| *** eil397 has quit IRC | 23:40 | |
| *** eil397 has joined #openstack-ansible | 23:45 | |
| *** sacharya_ has quit IRC | 23:49 | |
| *** sacharya has joined #openstack-ansible | 23:50 | |
| *** spotz is now known as spotz_zzz | 23:51 | |
| *** dolphm_418 is now known as dolphm | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!