| palendae | javeriak: Oops | 00:01 |
|---|---|---|
| palendae | javeriak: I'm honestly not sure - can always move it | 00:01 |
| javeriak | so I should go ahead and just push to kilo ? | 00:02 |
| javeriak | is that appropiate? | 00:02 |
| palendae | javeriak: Is it the plumgrid stuff? | 00:03 |
| javeriak | yes its is | 00:03 |
| *** mahito has quit IRC | 00:03 | |
| palendae | I would say put it in kilo for now, and we can discuss on the Thursday meeting at worst | 00:03 |
| javeriak | I decided it was better to do this in two steps, introduce the modularizing stuff in one spec/commit and the add the plumgrid plugin in the next | 00:03 |
| palendae | I don't know that it would make it in for our first kilo release (which will be 11.0.0), but better to have it in the specs repo than not | 00:04 |
| palendae | I'd agree with that | 00:04 |
| javeriak | what's the code freeze date for 11.0.0? | 00:05 |
| palendae | It'll be after kilo is officially released upstream; we're working on the kilo package support now | 00:06 |
| javeriak | ok, right | 00:09 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 00:09 | |
| *** mahito has joined #openstack-ansible | 00:11 | |
| palendae | javeriak: This patch will be the starting point https://review.openstack.org/#/c/166986/ | 00:11 |
| palendae | javeriak: Hopefully that'll be accepted by Wednesday at the latest | 00:11 |
| javeriak | and if we were interested in 10.1.1, backporting would be possible after the original commits merged on master correct? | 00:13 |
| palendae | The 10.x version would likely be 10.2, but yes - if you're interested, backporting could be possible. It's worth noting that 10.x is considered Rackspace technical debt; 11.0.0 and on are to be the community driven ones | 00:18 |
| palendae | What that means in technical terms is the playbooks and variables are in different locations | 00:19 |
| *** vmtrooper has joined #openstack-ansible | 00:19 | |
| javeriak | So 11.0.0 seems to be bringing in major changes. We would generallly be interesed in 10.x at the moment. | 00:22 |
| javeriak | Are 11.0.0 and 10.2 going to be tagged at the same time? | 00:23 |
| palendae | That I don't know | 00:24 |
| palendae | 11.0.0 will be cut from master | 00:24 |
| palendae | Master has the 'galaxified' roles | 00:24 |
| *** vmtrooper has quit IRC | 00:24 | |
| palendae | Which were pulled out of the rpc_deployment directoy | 00:24 |
| palendae | We're not abandoning 10.x though | 00:24 |
| javeriak | 10.1.3 doesnt seem to have come out yet either? | 00:25 |
| javeriak | Since I've been working off of the 10.x branch, I take it I should commit from the same rather than master? | 00:26 |
| palendae | That'll probably be easiest. the numbered ones are tags, the icehouse, juno, and master ones are the branches | 00:30 |
| javeriak | yep right thats what I meant, ok cool thanks, for now I will just push in my specs to the kilo directory | 00:36 |
| palendae | Cool | 00:40 |
| stevelle | palendae: if you have any clues about my gating joy, I'd be happy for deets | 00:42 |
| palendae | stevelle: the command vs shell? | 00:43 |
| stevelle | palendae: no, the lxc patch | 00:43 |
| stevelle | just lots of tempest failures | 00:44 |
| stevelle | root cause escaping me | 00:44 |
| palendae | Looking | 00:45 |
| palendae | 500s >.< | 00:45 |
| palendae | stevelle: I haven't seen these in my testing so far | 00:48 |
| stevelle | yeah, it has to be a problem in the code | 00:48 |
| stevelle | guessing it's "ip link add name ${LXC_BRIDGE} type bridge" | 00:49 |
| palendae | It's failing on neutron too... | 00:50 |
| palendae | I don't see anything in the console with 'LXC container', which is what the script is outputting | 00:53 |
| stevelle | palendae: https://review.openstack.org/#/c/169069/5/playbooks/roles/lxc_hosts/templates/lxc-system-manage.j2,unified line 206 I think the condition is inverted. needs a ! | 00:53 |
| stevelle | not confident of that | 00:54 |
| palendae | That reads to me like the conditional is attempting to actually create it | 00:55 |
| palendae | So if it successfully works, configure it | 00:55 |
| palendae | Interesting that we use ip link in one part and brctl add in another | 00:56 |
| stevelle | that was done specifically so we could avoid screen scraping existence of the bridge with "brctl show" | 00:59 |
| stevelle | alternative would be like if [ $(brctl show | grep ${LXC_BRIDGE} | wc -l) ] | 01:00 |
| *** javeriak has quit IRC | 01:03 | |
| *** javeriak has joined #openstack-ansible | 01:15 | |
| palendae | Ah, right | 01:20 |
| *** javeria has joined #openstack-ansible | 01:26 | |
| *** javeria has quit IRC | 01:28 | |
| *** javeriak_ has joined #openstack-ansible | 01:30 | |
| *** javeriak has quit IRC | 01:31 | |
| *** javeriak_ has quit IRC | 01:31 | |
| *** javeriak has joined #openstack-ansible | 01:32 | |
| *** stevemar has joined #openstack-ansible | 01:39 | |
| cloudnull | stevelle the latest set of failures (500 errors) loot to be neutron specific I've gone ahead and fired off a retry. we'll see if it still dies. | 02:04 |
| palendae | cloudnull: Did you see javeriak's questions? Do we have a way to communicate estimated release dates on launchpad targets? | 02:05 |
| palendae | I assume we could edit at https://launchpad.net/openstack-ansible/+milestones ? | 02:05 |
| cloudnull | within launchpad we can set release dates. | 02:06 |
| cloudnull | 11.x, per launchpad, has no expected release date. | 02:06 |
| palendae | Would be good to relay some of those timelines to the community | 02:06 |
| palendae | Right | 02:06 |
| palendae | Nor does 9.0.5 :) | 02:06 |
| cloudnull | 10.1.3 is past due, due to a few issues being worked through. | 02:07 |
| cloudnull | but 10/9 are rax technical debt at this point. | 02:08 |
| palendae | Yeah, though aren't we supporting the past 2 series? | 02:08 |
| *** vmtrooper has joined #openstack-ansible | 02:08 | |
| palendae | Well, current and current-1 | 02:08 |
| cloudnull | we are. | 02:08 |
| palendae | (also, I dunno if that's written down for os-a-d) | 02:09 |
| *** vmtrooper has quit IRC | 02:13 | |
| openstackgerrit | Steve Lewis proposed stackforge/os-ansible-deployment: Updated repository for minimum viable kilo install https://review.openstack.org/166986 | 02:40 |
| *** javeriak has quit IRC | 02:43 | |
| *** javeriak has joined #openstack-ansible | 02:45 | |
| *** javeriak has quit IRC | 03:03 | |
| *** alextricity has quit IRC | 03:16 | |
| *** mahito has quit IRC | 03:41 | |
| *** vmtrooper has joined #openstack-ansible | 03:57 | |
| *** vmtrooper has quit IRC | 04:02 | |
| *** raminoid has joined #openstack-ansible | 05:32 | |
| *** javeriak has joined #openstack-ansible | 05:33 | |
| raminoid | Hi, I am new to OpenStack and I am tasked with bringing up a complete infrastructure on a sandbox account. This includes VMs, Floating IPs, Load Balancers, Security Groups, Puppet Master and Package server. I was told Ansible can help me here. I could use any guidance you have | 05:33 |
| palendae | raminoid: It's late for me, but this channel's largely about this project: http://git.openstack.org/cgit/stackforge/os-ansible-deployment | 05:37 |
| palendae | It will set up an openstack environment for you, and can do what we call an 'all-in-one' to test it out on a single machine | 05:38 |
| palendae | However, for multi-nodes, it does not try to do network configuration, nor does it try to do load balancing for production. There's an ansible role for HAProxy, but that's really only for development and testing | 05:39 |
| raminoid | thanks for the clarification | 05:44 |
| palendae | raminoid: Sure thing. If you do check out the code and run it, the scripts/gate-check-commit.sh script will set up an AIO and try the tests | 05:45 |
| *** vmtrooper has joined #openstack-ansible | 05:46 | |
| palendae | Or you can manually run scripts/bootstrap-aio.sh, then openstack-ansible (a wrapper script that source ansible variable files for you) setup-hosts, setup-infrastructre, and setup-openstack | 05:46 |
| *** markvoelker has quit IRC | 05:48 | |
| *** vmtrooper has quit IRC | 05:51 | |
| *** markvoelker has joined #openstack-ansible | 06:19 | |
| *** markvoelker has quit IRC | 06:26 | |
| openstackgerrit | Javeria Khan proposed stackforge/os-ansible-deployment-specs: Add modularize-neutron-plays specification https://review.openstack.org/169189 | 06:44 |
| openstackgerrit | Javeria Khan proposed stackforge/os-ansible-deployment-specs: Add modularize-neutron-plays specification https://review.openstack.org/169189 | 06:50 |
| *** bilal has joined #openstack-ansible | 06:52 | |
| *** sdake has quit IRC | 06:56 | |
| *** stevemar has quit IRC | 07:16 | |
| *** markvoelker has joined #openstack-ansible | 07:22 | |
| *** markvoelker has quit IRC | 07:27 | |
| *** mahito has joined #openstack-ansible | 07:27 | |
| *** vmtrooper has joined #openstack-ansible | 07:35 | |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: [WIP] Heat config changes for kilo https://review.openstack.org/167947 | 07:38 |
| odyssey4me | morning | 07:38 |
| *** vmtrooper has quit IRC | 07:40 | |
| odyssey4me | any thoughts on why we're effectively setting up the container bridge twice in https://review.openstack.org/169069 ? | 07:44 |
| *** raminoid has left #openstack-ansible | 07:45 | |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: [WIP] Heat config changes for kilo https://review.openstack.org/167947 | 07:56 |
| *** javeriak has quit IRC | 08:05 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Updated nova logic to correct issues with addon devices https://review.openstack.org/166578 | 08:21 |
| *** markvoelker has joined #openstack-ansible | 08:23 | |
| *** markvoelker has quit IRC | 08:27 | |
| odyssey4me | the python wheels aren't building today, thanks to https://bugs.launchpad.net/horizon/+bug/1438543 | 09:01 |
| openstack | Launchpad bug 1438543 in OpenStack Dashboard (Horizon) "wrong package name 'XStatic-Angular-Irdragndrop' in horizon/requirements.txt" [Undecided,New] | 09:01 |
| mattt | openstack: bah | 09:11 |
| mattt | odyssey4me: rather! | 09:12 |
| mattt | i was wondering what the heck that failure was | 09:12 |
| odyssey4me | the repo happens to be in stackforge and no-one's done a review yet, so I think I'll do it quickly | 09:12 |
| odyssey4me | https://github.com/stackforge/xstatic-angular-lrdragndrop | 09:12 |
| odyssey4me | once that's done, then horizon has to be fixed :/ | 09:13 |
| odyssey4me | mattt seeing as horizon needs a patch to fix it now anyway, perhaps you can get on that? | 09:14 |
| git-harry | odyssey4me: that project is a bit confusing | 09:22 |
| git-harry | they use i and l interchangably | 09:23 |
| git-harry | *interchangeably | 09:23 |
| odyssey4me | git-harry yeah, it seems that someone did the pypi packaging configuration in stackforge... but the real stuff is elsewhere | 09:23 |
| odyssey4me | the packaging setup was broken | 09:23 |
| *** markvoelker has joined #openstack-ansible | 09:23 | |
| odyssey4me | somehow it's been fixed, but not in stackforge | 09:23 |
| *** vmtrooper has joined #openstack-ansible | 09:24 | |
| odyssey4me | so pypi is serving the right stuff now, but horizon's trying to use the old naming | 09:24 |
| git-harry | odyssey4me: the readme on the project uses i which would suggest the pypi package is wrong | 09:26 |
| odyssey4me | the readme on the source project? https://github.com/lorenzofox3/lrDragNDrop | 09:27 |
| odyssey4me | it uses lowercase l | 09:27 |
| mattt | odyssey4me: you want me to do what? | 09:28 |
| mattt | sorry, i was looking at something else | 09:28 |
| *** markvoelker has quit IRC | 09:28 | |
| git-harry | Oh I'm I looking at completely the wrong thing - https://github.com/stackforge/xstatic-angular-lrdragndrop | 09:28 |
| *** vmtrooper has quit IRC | 09:28 | |
| odyssey4me | mattt my suggestion was to fix this patch to use lowercase (e)l and not uppercase (eye)I | 09:29 |
| odyssey4me | https://github.com/openstack/horizon/commit/55df9369b6461275248bd1440570f3ad6eabe49e#diff-b4ef698db8ca845e5845c4618278f29a | 09:29 |
| mattt | why is this breaking now tho, no recent commit has gone into requirements.txt | 09:29 |
| odyssey4me | hmm, someone's already on it https://review.openstack.org/#/c/169132/ | 09:30 |
| git-harry | mattt: look at the date of the package on pypi | 09:30 |
| git-harry | I suspect it's been updated and the old one removed | 09:30 |
| odyssey4me | yup, looks like it to me | 09:30 |
| git-harry | and they've uses the different spelling | 09:30 |
| mattt | who does that | 09:30 |
| odyssey4me | well, the package was probably just renamed - if that's possible | 09:30 |
| odyssey4me | Thai Tran does that, apparently | 09:31 |
| mattt | so like | 09:33 |
| mattt | great if that merges into master, but shame we're not using master :P | 09:34 |
| odyssey4me | horizon patch in flight too: https://review.openstack.org/#/c/169133 | 09:34 |
| odyssey4me | we will have to do something to work around this, not sure what exactly | 09:34 |
| mattt | he has to rename it back | 09:36 |
| mattt | otherwise splosion everywhere | 09:36 |
| mattt | (or she) | 09:36 |
| odyssey4me | or we need to move horizon to a sha in our test builds, and need to sticky-tape the requirements.txt somehow? | 09:37 |
| mattt | probably best to sit on it for a few mins and see what happens upstream | 09:38 |
| mattt | before we get out the duct tape | 09:38 |
| odyssey4me | ah, easy enough - we already pin global requirements to a sha... so we just update that, and update horizon's pin to a sha too | 09:41 |
| odyssey4me | this is what we get for working with pre-release software :) | 09:41 |
| mattt | this affects everything tho | 09:42 |
| mattt | actually no it doesn't | 09:43 |
| mattt | guess this was introduced in kilo | 09:43 |
| git-harry | odyssey4me: I don't think it's because we're using pre-release. It's because the package was removed from pypi | 09:44 |
| svg | Good 'morning' fellows of the stack | 09:44 |
| svg | The reason that haproxy is not meant for production, is that because that setup is not HA by itself, or are there other reason? | 09:44 |
| odyssey4me | svg we've put no effort into making the configuration 'production ready' or highly available as we recommend that physical load balancers are used instead and have only tooled it up for development purposes | 09:46 |
| odyssey4me | if anyone else wishes to submit patches and use it in their reference architecture, that's fine and dandy | 09:46 |
| svg | ok, so that's a yes, thx | 09:47 |
| odyssey4me | hmm, why are projects putting requirements into both their own requirements.txt and the global requirements.txt ? | 09:55 |
| *** mahito has quit IRC | 10:04 | |
| openstackgerrit | Hugh Saunders proposed stackforge/os-ansible-deployment: Update cinder services to kilo https://review.openstack.org/168066 | 10:05 |
| *** markvoelker has joined #openstack-ansible | 10:24 | |
| *** markvoelker has quit IRC | 10:29 | |
| *** bilal has quit IRC | 10:41 | |
| *** vmtrooper has joined #openstack-ansible | 10:59 | |
| *** markvoelker has joined #openstack-ansible | 11:25 | |
| *** markvoelker has quit IRC | 11:29 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Updated nova logic to correct issues with addon devices https://review.openstack.org/166578 | 11:34 |
| *** galstrom_zzz is now known as galstrom | 11:46 | |
| *** britthouser has quit IRC | 11:54 | |
| *** vmtrooper has quit IRC | 11:55 | |
| *** britthouser has joined #openstack-ansible | 11:55 | |
| svg | I'm at the 'name: Ensure Keystone Service' playing on 'hosts: glance_api[0]'; this task keeps running, seems the command hangs on connecting to the lb internal ip on 35357 | 12:03 |
| svg | I can ping the lb internal ip, I can't make a tcp connection to it | 12:04 |
| mattt | svg: did you run the haproxy playbook? | 12:06 |
| svg | yes | 12:06 |
| svg | I can curl that endpoint fine from other hosts, also from the infra hosts where that container is located | 12:06 |
| svg | I planned to use f5, but when wiating for the network colleague to set that up, I tried haproxy for now; I did use the ip from that infra hosts where haproxy is running, not e separate dedicated one | 12:07 |
| svg | ouch, another pebkac, bad network mask :( | 12:12 |
| mattt | svg: good | 12:12 |
| mattt | because i didn't have a clue what it could be :) | 12:13 |
| Sam-I-Am | i was suspecting a network issue | 12:13 |
| mattt | Sam-I-Am: speak up next time! | 12:13 |
| cloudnull | Morning. | 12:13 |
| Sam-I-Am | evening | 12:13 |
| mattt | morning cloudnull | 12:14 |
| cloudnull | How's it mattt | 12:14 |
| mattt | cloudnull: purdy good thank you | 12:14 |
| *** markvoelker has joined #openstack-ansible | 12:17 | |
| cloudnull | What's new and exciting in openstack land? | 12:18 |
| Sam-I-Am | kilo | 12:20 |
| Sam-I-Am | of course | 12:20 |
| cloudnull | Odyssey4me did we resolve the python requirements issue already ? | 12:25 |
| mattt | cloudnull: it didn't fail for me the last time i tried, so perhaps it was resolved upstream | 12:31 |
| mattt | yeah looks like XStatic-Angular-IrDragNDrop is available again | 12:32 |
| openstackgerrit | git-harry proposed stackforge/os-ansible-deployment: Update cinder services to kilo https://review.openstack.org/168066 | 12:35 |
| cloudnull | Mattt so we've not rebuilt our own mirror in some time. Once the min kilo stuff merges we can rebuild the tax mirror with some of the new packages which will mitigate this for some this. | 12:43 |
| cloudnull | Which would've helped because we would have a built wheel in our mirror. | 12:45 |
| mattt | cloudnull: yeah true! | 12:48 |
| svg | Anyone up for a swordfight? | 12:51 |
| cloudnull | How's it svg | 12:52 |
| svg | I'm slacking off, it's deploying the openstack stuff now | 12:52 |
| svg | working on nova atm | 12:52 |
| cloudnull | Nice. | 12:53 |
| mattt | we should build on top of gentoo | 12:53 |
| cloudnull | Its "compiling"... | 12:53 |
| mattt | 5 hours later .......... | 12:53 |
| cloudnull | Crap the kernel is missing some module .... | 12:54 |
| mattt | hehe | 12:54 |
| Sam-I-Am | cloudnull: compile it all again | 12:55 |
| Sam-I-Am | grab a beer | 12:55 |
| mattt | a beer? you're a slow drinker. | 12:55 |
| svg | cloudnull: that would become 'crap systemd is missing python' | 12:55 |
| cloudnull | Or just crap systemd... | 12:55 |
| svg | ack | 12:55 |
| svg | ah, next uop, neutron | 12:55 |
| * svg goes for a walk | 12:56 | |
| mattt | svg: it shouldn't take that long :) | 12:56 |
| cloudnull | Sam-i-am loves neutron ;) | 12:56 |
| Sam-I-Am | mattt: i call it 'inexperienced' | 12:56 |
| mattt | Sam-I-Am: i may actually need your help w/ neutron today | 12:57 |
| mattt | i have no idea what i'm doing wrong, but it aint working :) | 12:57 |
| Sam-I-Am | first guess - you're using it. | 12:57 |
| mattt | hehe | 12:58 |
| *** galstrom is now known as galstrom_zzz | 13:02 | |
| Sam-I-Am | lulz @ nova gate still using sql_connection | 13:03 |
| *** subscope has joined #openstack-ansible | 13:07 | |
| svg | woohoo, can login to horizon :) | 13:12 |
| svg | colleagues that are testing mirantis/fuel and started at the same time as me, are still stuck with setting up pxe stuff for their hardware | 13:13 |
| hughsaunders | svg: woot :) | 13:15 |
| *** logan2 has quit IRC | 13:16 | |
| *** logan2 has joined #openstack-ansible | 13:19 | |
| cloudnull | Winning all around ;) | 13:20 |
| mattt | svg: YOU WIN | 13:27 |
| *** jaypipes has quit IRC | 13:29 | |
| *** Bjoern__ has joined #openstack-ansible | 13:32 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Add support for passing api version to glance module https://review.openstack.org/167959 | 13:41 |
| *** jaypipes has joined #openstack-ansible | 13:42 | |
| mattt | Sam-I-Am: ok i tracked down my problem (as in, what is happening), still have no idea what's causing it | 13:44 |
| mattt | Sam-I-Am: when i boot an instance w/ a private IP, the instance doesn't get an IP via dhcp | 13:45 |
| mattt | Sam-I-Am: looks like entries aren't getting added to bridge fdb | 13:45 |
| mattt | Sam-I-Am: when i manually add and reboot the instance it comes up w/ an IP | 13:45 |
| Sam-I-Am | manually add what, an ip? | 13:46 |
| Sam-I-Am | or manually add the entries to the fdb | 13:46 |
| mattt | Sam-I-Am: yes manually add entries to fdb | 13:46 |
| mattt | Sam-I-Am: on both the neutron agents controller and hypervisor | 13:47 |
| Sam-I-Am | hmm | 13:47 |
| Sam-I-Am | juno? | 13:47 |
| mattt | Sam-I-Am: this is kilo | 13:48 |
| Sam-I-Am | does this have anything to do with what Apsu is working on with things saying they happened but not really happening? | 13:48 |
| mattt | Sam-I-Am: the same AIO build on juno "just works" | 13:48 |
| mattt | Sam-I-Am: it's possible, but i've turned on debugging and i don't see neutron adding these entries | 13:48 |
| Sam-I-Am | thats odd | 13:49 |
| mattt | (i do see it trying to remove them if you tear down the network tho) | 13:49 |
| Sam-I-Am | does this happen consistently, or only in tempest? | 13:49 |
| mattt | Sam-I-Am: consistently | 13:51 |
| mattt | (for me at least) | 13:51 |
| mattt | i'm spinning up instances manually here | 13:51 |
| *** logan2 has quit IRC | 13:53 | |
| *** logan2 has joined #openstack-ansible | 13:56 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 13:59 | |
| Sam-I-Am | mattt: i dont have a working kilo install to play with yet | 14:00 |
| mattt | Sam-I-Am: ok think i found it | 14:04 |
| Sam-I-Am | mattt: what? | 14:05 |
| mattt | neutron | 14:05 |
| mattt | flipped to nova-network and solved. | 14:05 |
| Sam-I-Am | oh, right. that works. | 14:06 |
| mattt | nah, i think we're dropping a bad config in our kilo minor patch | 14:06 |
| mattt | just validating now | 14:06 |
| Sam-I-Am | mattt: if its always the case, then probably. | 14:06 |
| Sam-I-Am | did much change for kilo? | 14:07 |
| *** alextricity has joined #openstack-ansible | 14:09 | |
| *** Mudpuppy has joined #openstack-ansible | 14:20 | |
| *** Bjoern__ is now known as BjoernT | 14:21 | |
| *** sdake has joined #openstack-ansible | 14:29 | |
| *** galstrom_zzz is now known as galstrom | 14:40 | |
| sigmavirus24 | odyssey4me: I'm confused, do you want me to remove the reference to the keystone-kilofication blueprint? | 14:42 |
| odyssey4me | sigmavirus24 I would guess that the appropriate action is to mark the blueprint in launchpad as superceded, and to reference the blueprint in the commit message... although the approach you took with referencing both was the least offensive. :p | 14:44 |
| *** hj-hp has joined #openstack-ansible | 14:51 | |
| *** jaypipes has quit IRC | 14:53 | |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Updated for 9.0.7 tag https://review.openstack.org/169355 | 14:59 |
| rackertom | o/ | 15:01 |
| sigmavirus24 | \o | 15:02 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Update to readme to reference doc source more generally https://review.openstack.org/169358 | 15:03 |
| d34dh0r53 | meetings? | 15:05 |
| palendae | d34dh0r53: Off by an hour | 15:06 |
| palendae | d34dh0r53: It's not 1600 UTC | 15:06 |
| sigmavirus24 | hm | 15:06 |
| sigmavirus24 | oh damnit | 15:06 |
| Sam-I-Am | the calendar invite has been b3rnard0'd | 15:07 |
| b3rnard0 | hey! | 15:07 |
| b3rnard0 | let me check it | 15:08 |
| mattt | my calendar said 1600 | 15:08 |
| mattt | it is now 1600 | 15:08 |
| BjoernT | according outlook it is now 1600 utc | 15:08 |
| BjoernT | but google says 1500 | 15:08 |
| mattt | but that's not true :) | 15:08 |
| mattt | it is 1500 UTC | 15:08 |
| BjoernT | i trust google | 15:09 |
| mattt | said no one ever | 15:09 |
| palendae | BjoernT, mattt: I bet you're seeing the invite that was made based on CST | 15:09 |
| b3rnard0 | well something happened as it was right last week | 15:09 |
| mattt | b3rnard0: we flipped to summer time on sunday | 15:10 |
| sigmavirus24 | palendae: it /is/ based on UTC | 15:10 |
| palendae | Ah | 15:10 |
| BjoernT | Fun fact is I re did my invite in the UTC time zone and outlook is wrong | 15:10 |
| sigmavirus24 | BjoernT: yep | 15:10 |
| b3rnard0 | i see sigmavirus24 thursday invite at 10am CST | 15:10 |
| palendae | laffo | 15:10 |
| b3rnard0 | oh outlook | 15:10 |
| Sam-I-Am | outlook does gmt with daylight savings | 15:11 |
| BjoernT | Microsoft and Apple usually have time zone issues, lol | 15:11 |
| Sam-I-Am | its special | 15:11 |
| cloudnull | https://review.openstack.org/#/c/169355/ <- icehouse 9.0.7 review, Cores please give it a look when you can. | 15:11 |
| palendae | " Coordinated Universal Time (UTC) " Wouldn't that be CUT? Everything about this is wrong! | 15:12 |
| palendae | mattt: So your clocks are at 1600? | 15:12 |
| sigmavirus24 | palendae: do you know what CERN stands for? | 15:13 |
| palendae | Something goofy in swiss? | 15:13 |
| mattt | palendae: yeah we're currently GMT+1 | 15:13 |
| sigmavirus24 | CERN = European Organization for Nuclear Research | 15:13 |
| palendae | derived from the name "Conseil Européen pour la Recherche Nucléaire" | 15:14 |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment: [WIP] Heat config changes for kilo https://review.openstack.org/167947 | 15:14 |
| sigmavirus24 | palendae: "SI" is International System of Units | 15:14 |
| palendae | Guess they just didn't want to update the logo | 15:14 |
| Sam-I-Am | sigmavirus24: "si" means "yes" to b3rnard0 | 15:14 |
| sigmavirus24 | palendae: not everything is made or altered for arrogant self-important americans ;) | 15:15 |
| palendae | sigmavirus24: I meant CERN | 15:15 |
| palendae | They had a working acronym in French | 15:15 |
| palendae | And left it | 15:15 |
| b3rnard0 | Sam-I-Am: que? | 15:15 |
| palendae | Even though they change the english | 15:15 |
| Sam-I-Am | b3rnard0: what am i queuing? | 15:16 |
| sigmavirus24 | b3rnard0: isn't it qué? | 15:16 |
| Sam-I-Am | b3rnard0: like rabbitmq? | 15:16 |
| b3rnard0 | si | 15:17 |
| sigmavirus24 | b3rnard0: isn't it sÃ? | 15:17 |
| b3rnard0 | wat? | 15:17 |
| Sam-I-Am | Le Système International d'Unités ? | 15:18 |
| *** KLevenstein has joined #openstack-ansible | 15:19 | |
| *** KLevenstein is now known as KLev-afk | 15:20 | |
| odyssey4me | I'm confused. When is the meeting? | 15:20 |
| palendae | odyssey4me: In 40 minutes | 15:20 |
| odyssey4me | I'm at GMT+2 now. | 15:20 |
| odyssey4me | Right, so the calendar invites for Rackers are broken. b3rnard0 make the fix when you learn how to outlook :p | 15:21 |
| d34dh0r53 | b3rnard0: what were you saying about logging? | 15:26 |
| sigmavirus24 | d34dh0r53: there was a log-rotate LP bug reported last night by paul | 15:26 |
| sigmavirus24 | d34dh0r53: https://bugs.launchpad.net/openstack-ansible/+bug/1438417 | 15:27 |
| openstack | Launchpad bug 1438417 in openstack-ansible "No logrotate config in containers" [Undecided,New] | 15:27 |
| sigmavirus24 | I'm not sure if that's what b3rnard0 was talking about | 15:27 |
| sigmavirus24 | But you reminded me that I saw it | 15:27 |
| odyssey4me | sigmavirus24 d34dh0r53 good call on that bug, that's pretty nasty | 15:28 |
| odyssey4me | assuming it's true ;) | 15:28 |
| d34dh0r53 | b3rnard0 is suspiciously quiet | 15:28 |
| d34dh0r53 | odyssey4me: I think it is in juno/icehouse | 15:28 |
| Sam-I-Am | d34dh0r53: he's busy piecing together the outhouse invite | 15:29 |
| Sam-I-Am | mattt: neutron stuffs? | 15:29 |
| *** stevemar has joined #openstack-ansible | 15:29 | |
| mattt | Sam-I-Am: what's that? | 15:33 |
| Sam-I-Am | the problem you were having | 15:33 |
| Sam-I-Am | this is an aio, right? | 15:33 |
| mattt | Sam-I-Am: i've not been successful getting networking going on AIO or multi node build (all on cloud servers) | 15:34 |
| Mudpuppy | Need duct tape around my head today heheh | 15:34 |
| Sam-I-Am | mattt: whats your ml2 conf look like? | 15:35 |
| stevelle | odyssey4me: sigmavirus24 pretty sure we put log rotation in master with rework to rsyslog. needed in juno though | 15:36 |
| stevelle | d34dh0r53: ^ | 15:36 |
| sigmavirus24 | stevelle: that's what d34dh0r53 said ;) | 15:36 |
| stevelle | all agreement | 15:36 |
| mattt | Sam-I-Am: http://paste.openstack.org/show/197714/ | 15:36 |
| sigmavirus24 | <d34dh0r53> odyssey4me: I think it is in juno/icehouse [10:29:04|31-03-2015] | 15:36 |
| *** daneyon_ has joined #openstack-ansible | 15:36 | |
| Sam-I-Am | mattt: vxlan tenant network? | 15:37 |
| mattt | Sam-I-Am: yepper | 15:39 |
| *** daneyon has quit IRC | 15:39 | |
| Sam-I-Am | hmm, file looks ok | 15:40 |
| Sam-I-Am | mattt: nothing in any of the debug? neutron just doesnt Do The Thing? | 15:44 |
| sigmavirus24 | Sam-I-Am: mattt related to https://bugs.launchpad.net/openstack-ansible/+bug/1436916 perhaps? | 15:45 |
| openstack | Launchpad bug 1436916 in openstack-ansible "num_sync_threads set to 0 when api_workers set to 0" [Undecided,New] | 15:45 |
| Apsu | mattt: In the agents container, what's the output of "ip -o l | grep tap | grep -v master"? | 15:45 |
| mattt | Apsu: http://paste.openstack.org/show/197718/ | 15:48 |
| mattt | Apsu: not sure if you saw what i wrote earlier | 15:49 |
| mattt | Apsu: so i compared a kilo install to juno, and i noticed when i booted an instance on kilo i didn't have the same bridge fdb entries | 15:49 |
| mattt | Apsu: once i manually created them and rebooted the instance it was able to get an ip from dhcp | 15:49 |
| mattt | sigmavirus24: hehe nah | 15:49 |
| mattt | sigmavirus24: got burned by that one already, 100% sure it's not that | 15:49 |
| Apsu | mattt: I saw, yeah. I find that odd, but the fdb entries are an L2pop thing, and aren't explicitly required for operation. What is required, however, is that the taps are in the bridges | 15:50 |
| sigmavirus24 | mattt: just making sure =P | 15:50 |
| Apsu | And that tap you just pasted isn't in a bridge :) | 15:50 |
| mattt | Apsu: this is the issue you were wrestling before right? | 15:50 |
| Apsu | Yep | 15:51 |
| Apsu | I wrote a fuzzing script to attempt to figure out what's going on, and I discovered some fun facts | 15:51 |
| palendae | You discovered a den of lies | 15:51 |
| Apsu | ^ | 15:52 |
| Apsu | Apparently when you tell the kernel "do this network thing", and it returns from the syscall | 15:52 |
| *** jwagner_away is now known as jwagner | 15:53 | |
| Apsu | It doesn't immediately do the thing you told it to do | 15:55 |
| Apsu | So... creating an interface then immediately checking for it can fail if you do enough of them fast enough. | 15:55 |
| mattt | Apsu: wait, regarding my env | 15:55 |
| Apsu | Likewise bringing it up/down, adding/removing from bridge | 15:55 |
| mattt | Apsu: ip -o l | grep tap | grep master |wc -l => 4 | 15:55 |
| mattt | Apsu: so i do have some tap devices in bridges | 15:56 |
| Apsu | Yes. | 15:56 |
| Apsu | Most of them go in, sometimes one or two go in then back out. | 15:56 |
| Apsu | That specific scenario is what I'm trying to nail down now | 15:56 |
| Sam-I-Am | Apsu: is this issue only with kilo? | 15:56 |
| palendae | Sam-I-Am: no | 15:57 |
| mattt | Apsu: i don't even know what that problematic tap device is tho | 15:57 |
| palendae | Sam-I-Am: It's a kernel-level thing | 15:57 |
| mattt | Apsu: the ones i'm concerned with are all in bridges | 15:57 |
| Apsu | Sam-I-Am: No. | 15:57 |
| Apsu | Unrelated to neutron. | 15:57 |
| Sam-I-Am | well, yeah... i'm just wondering if something changed in neutron code that makes it more evident | 15:57 |
| Apsu | mattt: My guess is that's not true. Do you have debug on for your neutron LB agent? | 15:58 |
| mattt | Apsu: it must be, because if i manually add the fdb entries network flows | 15:58 |
| mattt | if one of hte necessary taps weren't in the bridge surely that wouldn't happen | 15:58 |
| palendae | Sam-I-Am: What first made it apparent to us was enabling the neutron tempest tests on Juno | 15:58 |
| Sam-I-Am | ahh ok | 15:58 |
| *** galstrom is now known as galstrom_zzz | 15:58 | |
| Apsu | mattt: hrm. That just doesn't make sense to me. | 15:58 |
| mattt | Apsu: ok let me try to document this quickly (need to leave office) | 15:59 |
| Apsu | mattt: No problem. Not that I don't believe what you're seeing, I'm just trying to understand how it could make sense. | 15:59 |
| Sam-I-Am | this | 16:00 |
| cloudnull | bug triage meetings: cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, mancdaz, dolphm, _shaps_, BjoernT, claco, echiu, dstanek | 16:00 |
| Apsu | The FDB (Forwarding DataBase) is part of every linux bridge. It's filled in automatically with MAC addresses during the normal course of switch learning operation | 16:00 |
| dstanek | o/ | 16:00 |
| b3rnard0 | hello. here's the link to the bug notes: https://etherpad.openstack.org/p/openstack_ansible_bug_triage.2015-03-31-16.00 | 16:00 |
| hughsaunders | lo | 16:00 |
| Apsu | The ML2 L2pop feature tries to add MAC addresses to the FDB that it already knows about, since Neutron built the network topology | 16:00 |
| Apsu | That way it saves on broadcast/flood behavior | 16:01 |
| Apsu | But that doesn't mean it Requires preloading the FDB. It should always work with/without it. | 16:01 |
| Sam-I-Am | "tries" | 16:01 |
| d34dh0r53 | \0/ | 16:01 |
| *** javeriak has joined #openstack-ansible | 16:02 | |
| cloudnull | helo . | 16:02 |
| cloudnull | first up: https://bugs.launchpad.net/openstack-ansible/+bug/1438498 | 16:02 |
| openstack | Launchpad bug 1438498 in openstack-ansible "Memcache connection limit too low" [Undecided,New] | 16:02 |
| *** tbyrom138 has joined #openstack-ansible | 16:03 | |
| cloudnull | its set to 1024. do we want to parameterize that? | 16:03 |
| cloudnull | seems like an easy fix in master/juno but is it important to us? | 16:03 |
| odyssey4me | is that hard set in our defaults, in a template or in a file (like an upstart file) ? | 16:04 |
| d34dh0r53 | I'm a big fan of parameterize all of the things | 16:05 |
| *** galstrom_zzz is now known as galstrom | 16:05 | |
| d34dh0r53 | less chance for snowflakes that way | 16:06 |
| odyssey4me | d34dh0r53 rumour has it that snowflakes are trendy :p | 16:06 |
| cloudnull | odyssey4me its an unset variable. | 16:06 |
| d34dh0r53 | odyssey4me: :) what all the cool kids are doing? | 16:06 |
| cloudnull | -c #N | 16:06 |
| cloudnull | -c <num> max simultaneous connections (default: 1024) | 16:06 |
| cloudnull | the template https://github.com/stackforge/os-ansible-deployment/blob/master/playbooks/roles/memcached_server/templates/memcached.conf | 16:07 |
| odyssey4me | ah, so something we haven't needed to set before now - in that case we may as well parameterise and add as a best practise | 16:07 |
| cloudnull | would need to be changed. | 16:07 |
| dstanek | do you also need to up the ulimit if you go higher than the 1024? | 16:08 |
| Apsu | Probabl | 16:09 |
| Apsu | y | 16:09 |
| odyssey4me | good point - sounds like some triage is required | 16:09 |
| cloudnull | dstanek i don't believe so. nothing in man page indicates that you would have to. | 16:10 |
| dstanek | cloudnull: i thought the default was 1024 fds | 16:10 |
| Apsu | dstanek: It is. | 16:10 |
| Apsu | http://stackoverflow.com/questions/2054212/is-ulimit-needed-launching-memcached-with-c minor verification | 16:11 |
| Apsu | But it does make sense. | 16:11 |
| sigmavirus24 | lol memcached | 16:12 |
| cloudnull | so we'd need to update the limits.conf where memcached is installed such that the limit is set to the same value of the user defined connections when greater then 1024. | 16:13 |
| d34dh0r53 | that's how I read it | 16:13 |
| Apsu | yar | 16:13 |
| odyssey4me | sounds like a worthy bug note, with a ref to the url from Apsu | 16:14 |
| cloudnull | moving on | 16:15 |
| cloudnull | for juno/icehouse: https://bugs.launchpad.net/openstack-ansible/+bug/1438417 | 16:15 |
| openstack | Launchpad bug 1438417 in openstack-ansible "No logrotate config in containers" [Undecided,New] | 16:15 |
| dstanek | cloudnull: yep, exactly. although making it a little higher may not sure - i don't know how many fds it uses for other things although i suspect not many | 16:15 |
| * cloudnull dstanek thats a good point. if we set it high than 1024 we may want to add some percentage over the defined value. | 16:17 | |
| cloudnull | andymccr, odyssey4me, d34dh0r53: https://bugs.launchpad.net/openstack-ansible/+bug/1438417 <- what say you? | 16:17 |
| openstack | Launchpad bug 1438417 in openstack-ansible "No logrotate config in containers" [Undecided,New] | 16:17 |
| d34dh0r53 | dstanek: cloudnull we could add 1024 to it so that the OS still has it's old default, up to a point of course | 16:18 |
| d34dh0r53 | cloudnull: andymccr odyssey4me confirmed, high | 16:18 |
| andymccr | sounds high to me :D | 16:18 |
| odyssey4me | cloudnull target at icehouse/juno with invalid at master - priority sounds high (as per the others) | 16:19 |
| cloudnull | one. | 16:19 |
| cloudnull | *done | 16:19 |
| cloudnull | what will it take to get that in? | 16:19 |
| odyssey4me | although icehouse is in maintenance - not sure if it'll every get a new tag, or are we still on the path to the last tag? | 16:20 |
| palendae | odyssey4me: A review was cut for 9.0.7 today | 16:20 |
| cloudnull | its on the path to a last tag, until we make another one :) | 16:20 |
| *** daneyon has joined #openstack-ansible | 16:20 | |
| odyssey4me | I suppose we can merge a fix and do a new tag if someone decides it's important enough | 16:21 |
| d34dh0r53 | perhaps a known issues for icehouse? | 16:21 |
| odyssey4me | work involved is probably not too bad | 16:21 |
| d34dh0r53 | fixed in docs :) | 16:21 |
| odyssey4me | if this is done in master, it could be back ported or written in - I would expect a juno backport to be relatively simple | 16:21 |
| cloudnull | ok , so being that this is a hot issue, who wants it? | 16:22 |
| odyssey4me | icehouse, well, that's often complicated - so yeah I agree with d34dh0r53 that perhaps a docs entry is best | 16:22 |
| d34dh0r53 | I think the Juno patch will need to be written from scratch as the logging arch changed somewhat. Porting from Juno to Icehouse *shouldn't* be that hard, but I don't think this is worthy of a dot release for icehouse by itself | 16:23 |
| *** daneyon_ has quit IRC | 16:24 | |
| odyssey4me | d34dh0r53 agreed, sounds like you're volunteering? | 16:25 |
| *** tbyrom138 has quit IRC | 16:26 | |
| * d34dh0r53 hides behind something | 16:26 | |
| d34dh0r53 | yeah, I can take it | 16:27 |
| BjoernT | and don't forget all the nice logs like neutron-agent* lol, those are filling up pretty fast | 16:27 |
| cloudnull | so d34dh0r53 voluntold. | 16:27 |
| cloudnull | next: https://bugs.launchpad.net/openstack-ansible/+bug/1438118 | 16:28 |
| openstack | Launchpad bug 1438118 in openstack-ansible "some v10 Juno playbooks containing a wrong host pattern" [Undecided,New] | 16:28 |
| odyssey4me | netron logs are awesome DDoS tools | 16:28 |
| BjoernT | lol, /dev/urandom is too | 16:29 |
| cloudnull | this is a host pattern problem in juno. | 16:29 |
| cloudnull | we should fix that for 10.1.3 | 16:29 |
| cloudnull | mo | 16:29 |
| cloudnull | *imo | 16:29 |
| Apsu | odyssey4me: debug=True, enjoy your iowait! | 16:30 |
| odyssey4me | whoops cloudnull yeah that's a bad lysdexic typos | 16:30 |
| cloudnull | it effects icehouse too | 16:30 |
| * odyssey4me nods | 16:31 | |
| Sam-I-Am | whoops | 16:31 |
| BjoernT | yeah it's an openstack problem in general. No debug on, does not give you any detail to fix issues and debug on is too much... | 16:31 |
| odyssey4me | cloudnull I would guess that this doesn't have too much effect functionally, so this is a low/medium priority. | 16:33 |
| *** galstrom is now known as galstrom_zzz | 16:34 | |
| odyssey4me | I would also venture to say that an icehouse fix is not essential. | 16:34 |
| cloudnull | i agree low | 16:34 |
| cloudnull | but it is something that we should resolve. | 16:34 |
| odyssey4me | the fix is fairly trivial | 16:34 |
| cloudnull | it is | 16:34 |
| odyssey4me | any volunteers | 16:34 |
| cloudnull | low hanging fruit here folks :) | 16:35 |
| odyssey4me | svg perhaps you'd like to submit a fix seeing as you found it? | 16:35 |
| odyssey4me | otherwise, if no volunteers then I'm happy to put one together | 16:36 |
| palendae | I could do it, too | 16:36 |
| odyssey4me | sold, to the palendae :) | 16:36 |
| cloudnull | palendae in the house ! | 16:37 |
| Sam-I-Am | palendae: earn your curs^H^H^Hore | 16:37 |
| cloudnull | next : https://bugs.launchpad.net/openstack-ansible/+bug/1437978 | 16:37 |
| openstack | Launchpad bug 1437978 in openstack-ansible "RabbitMQ restarts without condition in juno/icehouse" [Undecided,New] | 16:37 |
| cloudnull | essentially anytime you run the rabbitmq plays, it restarts in icehouse/juno . | 16:37 |
| odyssey4me | yeah, that sucks | 16:38 |
| cloudnull | we can potentially backport the logic from master. | 16:38 |
| odyssey4me | for production that can create a mess | 16:38 |
| odyssey4me | I'd vote for medium priority. | 16:39 |
| odyssey4me | Maybe low. | 16:39 |
| cloudnull | yup triaged medium | 16:39 |
| cloudnull | imo its a medium, its not idempotent and could cause production issues. | 16:39 |
| openstackgerrit | Miguel Alejandro Cantu proposed stackforge/os-ansible-deployment-specs: Add spec for BP implement-ceilometer https://review.openstack.org/169417 | 16:40 |
| Apsu | quiet openstackgerrit, we're busy | 16:40 |
| cloudnull | next: https://bugs.launchpad.net/openstack-ansible/+bug/1436999 | 16:40 |
| openstack | Launchpad bug 1436999 in openstack-ansible "LVM/DM/UDEV out of sync inside cinder-volumes container" [Undecided,New] | 16:40 |
| cloudnull | BjoernT around ? | 16:40 |
| Apsu | cloudnull: Saw that one, and I agree with BjoernT. In fact, I think it's surprising we haven't had more issues with udev being off in containers in fact lol | 16:42 |
| BjoernT | yes | 16:42 |
| * cloudnull hands mic to BjoernT | 16:42 | |
| *** KLev-afk is now known as KLevenstien | 16:43 | |
| *** KLevenstien is now known as KLevenstein | 16:43 | |
| BjoernT | so I will update the issue. I did some testing and was not able to reproduce it, in either configuration | 16:43 |
| cloudnull | so we'll need to do more work to confirm that issue. | 16:44 |
| BjoernT | I'm pretty sure that a static /dev, like we have right now is causing issues. Since the /dev/mapper tree is persistent across reboots | 16:44 |
| BjoernT | "/dev" is not managed by udev inside the container unless you set a parameter | 16:44 |
| BjoernT | obtain_device_list_from_udev rendered lvm useless | 16:45 |
| BjoernT | so I't a mystery how this can happen but we have a big issue here imho | 16:45 |
| cloudnull | so can we get some folks to go dive into this and track down what the root cause is ? | 16:46 |
| Apsu | What was the original issue causing us to disable udev? I mean, did we see an actual critical issue occurring, or just were trying to avoid something we'd heard about? | 16:46 |
| Apsu | Sorry, that's out of scope. I'll followup elsewhere | 16:46 |
| BjoernT | cpu load cause by udevd on each container | 16:46 |
| BjoernT | especially on infra nodes | 16:46 |
| cloudnull | Apsu triggering udev in the container causes it to trigger on the host. | 16:46 |
| cloudnull | x32+ containers makes the load hard to manage. | 16:47 |
| BjoernT | so to make it short, some dev help would be appreciated to reproduce it | 16:47 |
| *** bilal has joined #openstack-ansible | 16:47 | |
| cloudnull | also with all of udev enabled lvm becomes unavailable within the container. | 16:47 |
| *** jaypipes has joined #openstack-ansible | 16:48 | |
| odyssey4me | Sounds to me like Apsu and BjoernT could perhaps collaborate on this - although it does seem like an edge case. | 16:48 |
| cloudnull | yea Apsu BjoernT can you guys sync up and bang on it a bit ? | 16:49 |
| odyssey4me | Pure conjecture here, so correct me with pleasure. | 16:49 |
| Apsu | cloudnull: According to Ubuntu devs, their apparmor policy prevents udev trigger cascades, and they run udev in all their lxc containers. So... yeah. I'll follow up with BjoernT and look at it | 16:49 |
| BjoernT | nice | 16:49 |
| cloudnull | next: https://bugs.launchpad.net/openstack-ansible/+bug/1436916 | 16:50 |
| openstack | Launchpad bug 1436916 in openstack-ansible "num_sync_threads set to 0 when api_workers set to 0" [Undecided,New] | 16:50 |
| cloudnull | mattt ^ | 16:50 |
| odyssey4me | patch in already: https://review.openstack.org/168287 | 16:51 |
| odyssey4me | the whole api threads = 1 if it's 0 isunnecessary as it's already done at the top of that template | 16:53 |
| cloudnull | sounds good. | 16:54 |
| cloudnull | so last one: https://bugs.launchpad.net/openstack-ansible/+bug/1436647 | 16:55 |
| openstack | Launchpad bug 1436647 in openstack-ansible "Make chunk size configurable for swift-backed glance" [Undecided,New] | 16:55 |
| andymccr | assign that to me | 16:57 |
| andymccr | its super quick | 16:57 |
| andymccr | probably medium prio | 16:57 |
| odyssey4me | odd, why're we differing from the default swift which AFAIK is 5G, just like cloudfiles? | 16:58 |
| andymccr | odyssey4me: its a glance thing | 16:58 |
| cloudnull | andymccr done. | 16:58 |
| cloudnull | so we're done here. | 16:58 |
| andymccr | boom! | 16:58 |
| cloudnull | thanks everyone. | 16:58 |
| odyssey4me | ah, glance, how do I love thee... let me count the ways... | 16:58 |
| Sam-I-Am | woohoo | 16:58 |
| d34dh0r53 | notes -> https://etherpad.openstack.org/p/openstack_ansible_bug_triage.2015-03-31-16.00 | 16:58 |
| andymccr | odyssey4me: dont count the ways you will not be done before 2020! | 16:58 |
| cloudnull | just as a side note, we have a couple of specs that have been submitted, if we can get some reviews on them sometime soonish it would be great. | 16:59 |
| d34dh0r53 | minor grievance time? | 16:59 |
| d34dh0r53 | Based on this: https://wiki.openstack.org/wiki/Bugs I don't believe that we're using the Triaged tag appropriately | 17:00 |
| d34dh0r53 | s/tag/status/ | 17:00 |
| sigmavirus24 | huh | 17:00 |
| openstackgerrit | Miguel Grinberg proposed stackforge/os-ansible-deployment: WIP Horizon config changes for kilo https://review.openstack.org/167460 | 17:01 |
| odyssey4me | d34dh0r53 that's possible, but can you be more specific? | 17:02 |
| d34dh0r53 | We don't have specific analysis or fixes for the problem in most of our Triaged bugs | 17:03 |
| sigmavirus24 | odyssey4me: we're using triaged to mean that "The Bug team has seen this, confirmed it, and given it a priority" | 17:03 |
| odyssey4me | perhaps an item for thu's agenda should be whether we're adopting that as a set of definitions | 17:03 |
| sigmavirus24 | odyssey4me: that classification says it should be used if there's some kind of fix ready for it | 17:03 |
| odyssey4me | then we can happily allow bugs to be changed according to the appropriate status | 17:04 |
| * sigmavirus24 likes our definition better personally =P | 17:04 | |
| odyssey4me | I see there's been a similar debate on openstack-dev somewhere | 17:04 |
| d34dh0r53 | yeah, I agree this should be discussed in the Thu meeting | 17:04 |
| sigmavirus24 | d34dh0r53: add it to the agenda? | 17:04 |
| odyssey4me | sigmavirus24 I prefer confirmed for that status, but even then it's a stretch | 17:05 |
| odyssey4me | maybe incomplete should be set if it needs investigation | 17:05 |
| sigmavirus24 | odyssey4me: except that literally anyone can mark a bug as "confirmed" | 17:05 |
| odyssey4me | which many of our reported bugs do need | 17:05 |
| d34dh0r53 | sigmavirus24: do you have the link for the meeting wiki? | 17:05 |
| sigmavirus24 | d34dh0r53: it's in the topic? | 17:06 |
| odyssey4me | d34dh0r53 https://wiki.openstack.org/wiki/Meetings/openstack-ansible#Agenda_for_next_meeting | 17:06 |
| d34dh0r53 | facepalm | 17:06 |
| odyssey4me | hahaha | 17:06 |
| svg | odyssey4me: sure, that's a no brainer, and it'll let me get acquainted with the contribution process | 17:09 |
| d34dh0r53 | :) AFK Lunch | 17:09 |
| svg | If I understood correctltly, I should patch it in master, and tag it for juno? | 17:09 |
| odyssey4me | svg that looks like it's an issue for juno/icehouse only, so patch to the juno branch and cherry-pick ti icehouse... palendae can advise as he's volunteered, but it'd be great if you could take it on as we'd welcome more contributors | 17:11 |
| odyssey4me | if you do find similar patterns in master, then patch for them too | 17:12 |
| palendae | svg, odyssey4me I haven't started and am going to lunch shortly, so feel free :) | 17:12 |
| svg | Ah ok, I see the new info on the bug report. | 17:12 |
| odyssey4me | svg, feel free to reassign to yourself - many people in this channel can help you get through the hurdles of your first submission :) | 17:13 |
| svg | I'll try to get my act together this evening, as I didn't do anything on opentack contributions yet. | 17:13 |
| svg | will do, thx | 17:13 |
| svg | nice, welcoming community btw :_) | 17:13 |
| sigmavirus24 | svg: I'm willing to help with OpenStack/OSS contributions in general so feel free to ping me when you have time and we can figure something out | 17:14 |
| * sigmavirus24 regularly helps people get their first contributions into OSS projects | 17:14 | |
| svg | btw, anybody here that can get me in touch with sales/support for Europe? Part of the evaluation I'm doing, is getting fallback support, as we have as good as no openstack knowledge inhouse | 17:14 |
| svg | $boss sounded like 'mirantis will probably give better support' | 17:15 |
| svg | (he contacted rackspace via a web form, but didn't get any reply yet) | 17:15 |
| Sam-I-Am | svg: trying to figure it out | 17:18 |
| svg | thx Sam-I-Am | 17:19 |
| odyssey4me | heh svg us devs don't get involved in the sales process much :p - we'll need to find the right stuff to point you at | 17:19 |
| svg | I hope for you you don't :) | 17:20 |
| Sam-I-Am | haha | 17:20 |
| svg | but anything that gets us somewhere fater than a 'webform' | 17:20 |
| svg | faster | 17:20 |
| odyssey4me | I am surprised that http://www.rackspace.co.uk/cloud/private (number at the bottom, live web chat option at the top) hasn't got $boss far? | 17:21 |
| javeriak | hey guys, i got reviews back for my spec adding modular neutron plays https://review.openstack.org/#/c/169189/2 , so I've assigned juno as the series goal to the bp | 17:23 |
| svg | odyssey4me: he emailed | 17:23 |
| javeriak | and from the latest comment i take it, this can still go in for 10.x ? | 17:25 |
| javeriak | and I can file it again for master | 17:25 |
| svg | odyssey4me: I'm doing a webchat now, hopefully I can get him th einfo he wants, thx | 17:27 |
| odyssey4me | javeriak the general idea is usually to focus initial attempts at new ways of doing things in master, then to look at back-porting | 17:28 |
| odyssey4me | however, it's not exclusive and if you're prepared to forward-port then it may be entertained | 17:29 |
| javeriak | yep my current focus is supporting these changes for 10.x, and I would ofcourse also like to add them to master in parallel, would this require opening a seperate bp to track the master effort? | 17:35 |
| *** sdake_ has joined #openstack-ansible | 17:37 | |
| odyssey4me | javeriak to be honest this is new territory, perhaps cloudnull can advise or perhaps it's best discussed at the next community meeting | 17:39 |
| odyssey4me | for now I'd suggest that the blueprint cover the intention clearly | 17:39 |
| stevelle | seems to me a single blueprint should be sufficient unless it turns out that the implementation is fundamentally different and requires a different strategy | 17:41 |
| *** sdake has quit IRC | 17:41 | |
| odyssey4me | stevelle I would agree personally. | 17:42 |
| *** sdake has joined #openstack-ansible | 17:48 | |
| *** sdake_ has quit IRC | 17:52 | |
| *** javeriak has quit IRC | 18:06 | |
| *** bilal has quit IRC | 18:11 | |
| sigmavirus24 | odyssey4me: stevelle blueprints are almost always exclusively targetted towards master | 18:14 |
| sigmavirus24 | stable/maintenance branches usually don't receive features/additions | 18:14 |
| cloudnull | sigmavirus24 i'd agree however we removed stable from the name of the branches because we were attempting to provide a mechanism to allow features to be added into a branch that may have already been released. | 18:18 |
| sigmavirus24 | blueprints are also designed to only target one series of development | 18:19 |
| sigmavirus24 | We can use it to track two, but it'll probably confuse people later down the road if we do that | 18:20 |
| cloudnull | ie in future majors we'd want to provide "x" post the initial release. the issue is that 10/9.x are rax technical debt so as a rax product do we want to have it included and when? | 18:20 |
| sigmavirus24 | (to be clear, I'm not arguing to not back/forward port something because of LaunchPad, just that if we want to track the work in a less confusing way, we should probably not use one) | 18:21 |
| palendae | cloudnull: For javeriak's first BP, the modular networking support, I think would be useful. However, my personal opinion is that it should be current forward | 18:21 |
| palendae | Though it appears he has use for it in 10. | 18:21 |
| palendae | x | 18:22 |
| sigmavirus24 | Also note that while we all acknowledge 9.x and 10.x as technical debt, no one is aiming to refactor them so that they aren't | 18:23 |
| cloudnull | as i read the bp i think he is developing/has developed the functionality for "10.x" so we might want to see if we can team up with him and get the functionality into master. | 18:23 |
| cloudnull | sigmavirus24 right no refactoring the stable branches. | 18:25 |
| cloudnull | we've already been down that road =) | 18:25 |
| cloudnull | there be dragons. | 18:25 |
| sigmavirus24 | Right | 18:25 |
| * palendae aims to refactor | 18:26 | |
| sigmavirus24 | palendae: dont' make me shoot you with a blue shell | 18:26 |
| palendae | You mean a gold shell? | 18:27 |
| BjoernT | i take that too | 18:27 |
| *** javeriak has joined #openstack-ansible | 18:44 | |
| palendae | If some others could look at https://review.openstack.org/#/c/169069/ and comment/review, I'd appreciate it | 18:45 |
| Sam-I-Am | palendae: ha, you almost got me. | 18:49 |
| palendae | Sam-I-Am: ? | 18:49 |
| *** prometheanfire has joined #openstack-ansible | 18:50 | |
| Sam-I-Am | palendae: conflict, sir. | 18:51 |
| prometheanfire | is there a list of software that's needed by os-ansible? lxc and the like? (not the stuff from requirements.txt) | 18:51 |
| palendae | Sam-I-Am: You're a conflict | 18:51 |
| prometheanfire | trying to see how to package this (if that's even applicable) | 18:51 |
| Sam-I-Am | prometheanfire: the docs contain a list of packages needed on the hosts | 18:51 |
| Sam-I-Am | then the ansible stuff grabs packages for each service | 18:51 |
| Sam-I-Am | palendae: thanks for the compliment | 18:52 |
| prometheanfire | ah, guess it assumes a specific distro then | 18:52 |
| palendae | prometheanfire: There's http://git.openstack.org/cgit/stackforge/os-ansible-deployment/tree/scripts/bootstrap-aio.sh#n87 | 18:52 |
| Sam-I-Am | it runs on ubuntu | 18:52 |
| prometheanfire | right, forgot about that | 18:52 |
| Sam-I-Am | you could probably make it work on another distro | 18:52 |
| palendae | http://docs.rackspace.com/rpc/api/v9/bk-rpc-installation/content/sec_overview_requirements.html | 18:52 |
| prometheanfire | ya, I'm sure I can | 18:53 |
| prometheanfire | you need to define the container image location I'm guessing | 18:53 |
| prometheanfire | or is that run from the host? | 18:53 |
| *** stevemar has quit IRC | 18:54 | |
| Sam-I-Am | prometheanfire: containers are built on the fly | 18:54 |
| prometheanfire | using what as the source? | 18:55 |
| prometheanfire | that's my question? | 18:55 |
| prometheanfire | can I run this on gentoo? :P | 18:55 |
| palendae | http://git.openstack.org/cgit/stackforge/os-ansible-deployment/tree/playbooks/roles/lxc_container_create/tasks/container_create.yml#n57 | 18:55 |
| Sam-I-Am | openstack and python deps are python rolled into wheels packages | 18:56 |
| Sam-I-Am | the system deps are dpkg | 18:56 |
| palendae | You can run it, I don't know that it'll succeed :) | 18:56 |
| * prometheanfire is considering using this as the 'official' deployment way for gentoo | 18:56 | |
| palendae | Because of what he said | 18:56 |
| palendae | Lots of references to deb and dpkg hard coded | 18:56 |
| prometheanfire | is dpkg run in the container, that's my question | 18:56 |
| Sam-I-Am | maybe there's a package-ish module for ansible | 18:56 |
| prometheanfire | if so, no problem | 18:56 |
| Sam-I-Am | apt is, yes | 18:56 |
| Sam-I-Am | ssh into container, do things. | 18:57 |
| prometheanfire | so the host only need lxc things | 18:57 |
| palendae | prometheanfire: Yes, see http://git.openstack.org/cgit/stackforge/os-ansible-deployment/tree/playbooks/roles/lxc_container_create/tasks/container_create.yml#n64 | 18:57 |
| palendae | Though that actually just sets up apt sources | 18:57 |
| sigmavirus24 | prometheanfire: the lxc template is ubuntu which I'm not sure gentoo will like | 18:58 |
| prometheanfire | if it's in the container that's fine | 18:58 |
| prometheanfire | lxc template can be whatever it wants though, distro should not mater | 18:58 |
| palendae | Yeah | 18:58 |
| palendae | I see what you're getting at | 18:58 |
| palendae | Gentoo hosts with ubuntu containers? | 18:58 |
| prometheanfire | basically | 18:58 |
| sigmavirus24 | Gentoo deployers will love that /s | 18:59 |
| prometheanfire | I can work on submitting review to abstract other things :P | 18:59 |
| * prometheanfire waves at sigmavirus24 | 18:59 | |
| sigmavirus24 | Well as long as you don't try to use the downstream packages. Those are almost always broken | 18:59 |
| palendae | http://git.openstack.org/cgit/stackforge/os-ansible-deployment/tree/playbooks/roles/openstack_hosts/tasks/openstack_host_packages.yml runs apt on a loop | 18:59 |
| sigmavirus24 | =P | 18:59 |
| prometheanfire | my packages are good :P | 18:59 |
| sigmavirus24 | That's what they all say | 19:00 |
| prometheanfire | then again, I basically just package straight from pip | 19:00 |
| prometheanfire | the deps stay the same, etc | 19:00 |
| sigmavirus24 | prometheanfire: I'm mostly ribbing you | 19:00 |
| prometheanfire | I know | 19:00 |
| * sigmavirus24 hasn't tried out teh gentoo packages | 19:00 | |
| prometheanfire | I want to stop packaging openstack, more or less :P | 19:00 |
| sigmavirus24 | On the bright side, I never see bug reports from users of them | 19:00 |
| palendae | So setup-hosts.yml calls openstack-hosts-setup.yml, which calls the openstack_hosts role, with has a call to apt | 19:00 |
| sigmavirus24 | prometheanfire: that's a healthy life decision | 19:00 |
| palendae | That's the extent of what I see, though I haven't actually grepped | 19:01 |
| * sigmavirus24 is pretty sure the debian maintainer is going to not live very long at this rate | 19:01 | |
| prometheanfire | I'll still package the clients, but the way to run openstack servers on gentoo would be containers | 19:01 |
| prometheanfire | anyway, time for a meeting | 19:01 |
| prometheanfire | palendae: I'll have to check if ansible has that package resource, need to get more familiar with it anyway | 19:02 |
| prometheanfire | brb in an hour :( | 19:02 |
| *** stevemar has joined #openstack-ansible | 19:04 | |
| *** stevemar has quit IRC | 19:08 | |
| *** stevemar has joined #openstack-ansible | 19:08 | |
| *** hj-hp has quit IRC | 19:24 | |
| *** hj-hp has joined #openstack-ansible | 19:24 | |
| *** sdake_ has joined #openstack-ansible | 19:31 | |
| *** sdake__ has joined #openstack-ansible | 19:35 | |
| *** sdake has quit IRC | 19:35 | |
| d34dh0r53 | cloudnull: we good to merge 9.0.7? | 19:37 |
| *** sdake_ has quit IRC | 19:38 | |
| cloudnull | make it go | 19:39 |
| cloudnull | if there are at least 2 +2's | 19:39 |
| palendae | plussed a 2 | 19:39 |
| d34dh0r53 | gone, no refunds | 19:40 |
| *** Mudpuppy_ has joined #openstack-ansible | 19:47 | |
| *** galstrom_zzz is now known as galstrom | 19:47 | |
| *** Mudpuppy has quit IRC | 19:50 | |
| *** hj-hp has quit IRC | 19:50 | |
| openstackgerrit | Javeria Khan proposed stackforge/os-ansible-deployment-specs: Add modularize-neutron-plays specification https://review.openstack.org/169189 | 19:53 |
| *** jwagner is now known as jwagner_away | 19:54 | |
| *** Mudpuppy_ has quit IRC | 19:56 | |
| *** Mudpuppy has joined #openstack-ansible | 19:56 | |
| bgmccollum | should this even be in the icehouse branch? https://github.com/stackforge/os-ansible-deployment/blob/icehouse/etc/rpc_deploy/conf.d/swift.yml.example | 20:00 |
| *** hj-hp has joined #openstack-ansible | 20:01 | |
| cloudnull | bgmccollum nope. | 20:03 |
| bgmccollum | ;( | 20:03 |
| *** daneyon_ has joined #openstack-ansible | 20:07 | |
| *** Mudpuppy has quit IRC | 20:08 | |
| sigmavirus24 | bgmccollum: nice catch | 20:08 |
| bgmccollum | this too - https://github.com/stackforge/os-ansible-deployment/blob/icehouse/etc/rpc_deploy/rpc_user_config.yml#L63 | 20:08 |
| openstackgerrit | Javeria Khan proposed stackforge/os-ansible-deployment-specs: Add modularize-neutron-plays specification https://review.openstack.org/169189 | 20:10 |
| *** daneyon has quit IRC | 20:11 | |
| stevelle | javeriak: FYI there is one extra space on line 39 and 40 preceding the hyphen, over indenting the list items. Contrast with 43 and 44 and they don't line up. | 20:13 |
| *** Mudpuppy has joined #openstack-ansible | 20:14 | |
| javeriak | stevelle , thanks for the catch, still trying to get used to the sphinx syntax :) | 20:16 |
| *** jwagner_away is now known as jwagner | 20:16 | |
| svg | [preparing first contribution] making a patch for a registered bugm on say the juno branch, which needs a similar/related patch on the icehouse branch; do I need to do sth specific to have a relation with both patches ? or is mentioning the same `Closes-Bug:' in both commits enough? | 20:17 |
| openstackgerrit | Javeria Khan proposed stackforge/os-ansible-deployment-specs: Add modularize-neutron-plays specification https://review.openstack.org/169189 | 20:17 |
| stevelle | javeriak: it looks like you can use tox -e docs for a local test | 20:17 |
| javeriak | I've been basically using this http://rst.ninjs.org to test out the formatting | 20:19 |
| cloudnull | svg: `Closes-Bug:` in both patches should be enough. it the bug will require multiple patches to close for a single release branch you can use `Partial-Bug:` | 20:19 |
| svg | ok, thx; Partial-Bug then | 20:20 |
| cloudnull | javeriak if you just run "tox" it will test and do all the building. the tox.ini has been setup to do all of the build job for you. | 20:21 |
| javeriak | cloudnull, cool, just gave it a try, much easier | 20:22 |
| cloudnull | yea less steps :) | 20:23 |
| palendae | cloudnull: Would you be able to answer some of the questions on https://review.openstack.org/#/c/169069/ ? | 20:23 |
| cloudnull | inline comments ? | 20:24 |
| palendae | yeah | 20:24 |
| cloudnull | let e go look | 20:24 |
| svg | Is `git review' always that slow? | 20:26 |
| cloudnull | palendae stevelle we can get rid of that top brctl call/configure_bridge call the check in sysfs for the bridge should be enough for us to skip the container build process. | 20:26 |
| palendae | So lines 190-197 could be removed? | 20:27 |
| cloudnull | 191 - 196 | 20:28 |
| svg | Hey, git review fails on my stating I need to fill in my contct details, which I tried several times alreaady, but that always gets me Server Error - Cannot store contact information | 20:28 |
| palendae | Ah, yes | 20:28 |
| *** jwagner is now known as jwagner_away | 20:29 | |
| cloudnull | svg sorry for the reference doc but have you gone through - https://wiki.openstack.org/wiki/Documentation/HowTo/FirstTimers#Set_up_git_and_git_review | 20:29 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Updated for 9.0.7 tag https://review.openstack.org/169355 | 20:30 |
| svg | cloudnull: Yes, I have. The issue is me not being able to fille in my contact details on https://review.openstack.org/#/settings/contact | 20:31 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Updated for 9.0.7 tag https://review.openstack.org/169505 | 21:01 |
| *** daneyon_ has quit IRC | 21:02 | |
| openstackgerrit | Serge van Ginderachter proposed stackforge/os-ansible-deployment: Correct a syntax error in some hosts: patterns https://review.openstack.org/169506 | 21:02 |
| *** daneyon has joined #openstack-ansible | 21:02 | |
| openstackgerrit | Serge van Ginderachter proposed stackforge/os-ansible-deployment: Correct a syntax error in some hosts: patterns https://review.openstack.org/169507 | 21:04 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Repo Updated for 9.0.7 tag https://review.openstack.org/169505 | 21:09 |
| cloudnull | ^ this review is going to fail for icehouse until it gets replicated out to mirror.rax | 21:10 |
| *** hj-hp has quit IRC | 21:13 | |
| prometheanfire | cloudnull: chicken meet egg? | 21:17 |
| cloudnull | its a race ! | 21:17 |
| prometheanfire | cloudnull: you know if there's a package resource like puppets (where you can override name per distro) | 21:17 |
| prometheanfire | in ansible | 21:18 |
| cloudnull | example? | 21:18 |
| prometheanfire | packge 'apache' { | 21:18 |
| prometheanfire | rhel? name=httpd | 21:19 |
| *** BjoernT has left #openstack-ansible | 21:19 | |
| prometheanfire | debian? name=apache2 | 21:19 |
| prometheanfire | } | 21:19 |
| prometheanfire | peudo-y code | 21:19 |
| cloudnull | in our bits, thats in the apt_packages list | 21:19 |
| prometheanfire | right, I want that to be generic :P | 21:19 |
| cloudnull | IE https://github.com/stackforge/os-ansible-deployment/blob/master/playbooks/roles/os_cinder/defaults/main.yml#L143-L154 | 21:20 |
| cloudnull | so you'd have to add a new list. and tasks to support the new package lists. | 21:20 |
| prometheanfire | I get that | 21:20 |
| prometheanfire | it would probably be seperated out to lists per distro | 21:20 |
| prometheanfire | then load the correct list | 21:21 |
| prometheanfire | cinder_gentoo_packaged: or something | 21:21 |
| cloudnull | yup. | 21:21 |
| prometheanfire | the test pipeline would need gentoo hosts for this wouldn't it | 21:22 |
| cloudnull | yup | 21:22 |
| prometheanfire | lol | 21:22 |
| prometheanfire | guess that gives me more preasure to build official gentoo(hardened) images for OS | 21:23 |
| cloudnull | just adapt the AIO scripts and get it on an external CI within your own gentoo based deployment cluster. | 21:23 |
| cloudnull | BOOM! | 21:23 |
| prometheanfire | need an image first | 21:23 |
| cloudnull | openstack, in openstack, on gentoo. | 21:23 |
| cloudnull | it would be OO'G | 21:23 |
| prometheanfire | goog | 21:23 |
| prometheanfire | gentoo on openstack on openstack on gentoo | 21:24 |
| prometheanfire | need to add an le to the end | 21:24 |
| d34dh0r53 | LMAO | 21:24 |
| d34dh0r53 | OO'G | 21:24 |
| * cloudnull nobody gets my jokes.... | 21:24 | |
| mattt | haha, gentoo | 21:32 |
| *** hj-hp has joined #openstack-ansible | 21:32 | |
| mattt | with a lot of work i'm sure you could build a pretty nice platform | 21:32 |
| cloudnull | palendae whats the sate with https://bugs.launchpad.net/openstack-ansible/+bug/1430937 | 21:33 |
| openstack | Launchpad bug 1430937 in openstack-ansible juno "Enable NetworkBasicOps test in tempest commit checks" [Medium,New] | 21:33 |
| cloudnull | ^ isn't that in juno ? | 21:33 |
| cloudnull | Apsu: whats the state of https://bugs.launchpad.net/openstack-ansible/juno/+bug/1423783 | 21:34 |
| openstack | Launchpad bug 1423783 in openstack-ansible juno "Horizon is not generating CSS correctly" [Undecided,New] | 21:34 |
| cloudnull | Apsu palendae: whats the state of https://bugs.launchpad.net/openstack-ansible/+bug/1430937 https://bugs.launchpad.net/openstack-ansible/+bug/1425255 | 21:35 |
| openstack | Launchpad bug 1430937 in openstack-ansible juno "Enable NetworkBasicOps test in tempest commit checks" [Medium,New] | 21:35 |
| openstack | Launchpad bug 1425255 in openstack-ansible juno "Tempest NetworkBasicOps tests cannot ping from within instances" [Critical,New] - Assigned to Nolan Brubaker (nolan-brubaker) | 21:35 |
| Apsu | cloudnull: sec | 21:36 |
| Apsu | I did the master commit on the CSS one. Guess it got tagged for backport. | 21:36 |
| Apsu | As for the networkops tempest stuff, that's what I'm working on trying to find a workaround for with the kernel interface behavior | 21:37 |
| *** sdake has joined #openstack-ansible | 21:38 | |
| *** sdake__ has quit IRC | 21:42 | |
| *** Mudpuppy_ has joined #openstack-ansible | 21:44 | |
| *** KLevenstein has quit IRC | 21:44 | |
| *** Mudpuppy has quit IRC | 21:47 | |
| *** Mudpuppy_ has quit IRC | 21:48 | |
| *** KLevenstein has joined #openstack-ansible | 21:50 | |
| *** stevemar has quit IRC | 21:51 | |
| *** sdake_ has joined #openstack-ansible | 21:59 | |
| *** sdake has quit IRC | 22:03 | |
| *** javeriak has quit IRC | 22:04 | |
| *** javeriak has joined #openstack-ansible | 22:12 | |
| *** hj-hp has quit IRC | 22:18 | |
| *** KLevenstein has quit IRC | 22:23 | |
| *** hj-hp has joined #openstack-ansible | 22:26 | |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:42 | |
| *** devlaps has joined #openstack-ansible | 22:50 | |
| *** devlaps has quit IRC | 22:54 | |
| *** darrenc is now known as darrenc_afk | 22:59 | |
| *** galstrom is now known as galstrom_zzz | 23:02 | |
| *** jaypipes has quit IRC | 23:09 | |
| *** hj-hp has quit IRC | 23:15 | |
| *** hj-hp has joined #openstack-ansible | 23:16 | |
| *** hj-hp has quit IRC | 23:20 | |
| *** darrenc_afk is now known as darrenc | 23:22 | |
| *** jaypipes has joined #openstack-ansible | 23:23 | |
| *** javeriak has quit IRC | 23:32 | |
| *** javeriak has joined #openstack-ansible | 23:36 | |
| *** javeriak has quit IRC | 23:43 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!