| *** galstrom is now known as galstrom_zzz | 00:01 | |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Change to set the container network MTU https://review.openstack.org/204796 | 00:25 |
|---|---|---|
| *** alop has quit IRC | 00:27 | |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Added gate process and more logs https://review.openstack.org/204315 | 00:28 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Add swift-sync role and environment https://review.openstack.org/201136 | 00:28 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Fix glance_nfs_client setting https://review.openstack.org/204542 | 00:28 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Restart mysql when config changed https://review.openstack.org/204545 | 00:28 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Moved user_group_vars to defaults https://review.openstack.org/204320 | 00:28 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Fix haproxy service config when ssl is enabled https://review.openstack.org/202911 | 00:29 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Disable Unattended Upgrades https://review.openstack.org/203858 | 00:29 |
| cloudnull | all of the gate failures that we had due to earlier OS CI failures have been rebased. | 00:31 |
| * cloudnull going home. have a good night | 00:34 | |
| *** davidself has joined #openstack-ansible | 00:35 | |
| *** shangxdy has joined #openstack-ansible | 01:58 | |
| *** javeriak has joined #openstack-ansible | 02:20 | |
| *** git-harry has quit IRC | 02:23 | |
| *** git-harry has joined #openstack-ansible | 02:23 | |
| javeriak | hey... anyone around. | 02:42 |
| *** shangxdy has quit IRC | 03:16 | |
| openstackgerrit | Matthew Kassawara proposed stackforge/os-ansible-deployment: Reduce neutron configuration https://review.openstack.org/204597 | 03:18 |
| *** javeriak has quit IRC | 03:22 | |
| *** javeriak has joined #openstack-ansible | 03:23 | |
| *** javeriak has quit IRC | 03:59 | |
| *** javeriak has joined #openstack-ansible | 04:10 | |
| *** mordred has quit IRC | 04:11 | |
| *** prometheanfire has quit IRC | 04:11 | |
| *** prometheanfire has joined #openstack-ansible | 04:12 | |
| *** daneyon_ has joined #openstack-ansible | 04:22 | |
| *** mordred has joined #openstack-ansible | 04:24 | |
| *** daneyon has quit IRC | 04:24 | |
| *** javeriak has quit IRC | 04:56 | |
| *** shausy has joined #openstack-ansible | 05:14 | |
| jwitko | hola jav | 05:19 |
| *** shausy has quit IRC | 05:32 | |
| *** shausy has joined #openstack-ansible | 05:33 | |
| mattt | morning | 06:12 |
| *** javeriak has joined #openstack-ansible | 06:51 | |
| *** javeriak_ has joined #openstack-ansible | 06:55 | |
| *** javeriak has quit IRC | 06:55 | |
| *** daneyon_ has quit IRC | 07:20 | |
| *** daneyon has joined #openstack-ansible | 07:21 | |
| pellaeon | andymccr: I have some problem relating to yesterday's question | 07:39 |
| pellaeon | now my playbook repo is at 16ac504 | 07:39 |
| pellaeon | when I run setup-everything, keystone containers stuck: | 07:40 |
| pellaeon | failed: [infra2_keystone_container-bf3a1760] => (item=keystone) => {"attempts": 5, "cmd": "/usr/local/bin/pip install keystone", "failed": true, "item": "keystone"} | 07:40 |
| *** subscope has quit IRC | 07:41 | |
| pellaeon | I issued "pip install keystone" manually in the keystone containers | 07:41 |
| pellaeon | it returns: | 07:42 |
| pellaeon | # pip install keystone | 07:42 |
| pellaeon | Ignoring indexes: https://pypi.python.org/simple | 07:42 |
| pellaeon | Requirement already up-to-date: keystone in /usr/local/lib/python2.7/dist-packages | 07:42 |
| pellaeon | Collecting oslo.i18n<1.6.0,>=1.5.0 (from keystone) Could not find a version that satisfies the requirement oslo.i18n<1.6.0,>=1.5.0 (from keystone) (from versions: 2.1.0) | 07:42 |
| pellaeon | No matching distribution found for oslo.i18n<1.6.0,>=1.5.0 (from keystone) | 07:42 |
| pellaeon | using pip install -v I see that it is due to oslo.i18n@1.5.0 not found at http://172.29.236.19:8181/os-releases/master/ | 07:43 |
| pellaeon | and I dug into the code and found that the repo container clones data from rpc-repo.rackspace.com | 07:44 |
| pellaeon | which inside http://rpc-repo.rackspace.com/os-releases/master/ doesn't have oslo.i18n@1.5.0 either | 07:45 |
| pellaeon | I'd like to know how rpc-repo.rackspace.com is updated | 07:46 |
| pellaeon | I don't know if its due to me running on a unmerged commit ? | 07:49 |
| *** vdo has joined #openstack-ansible | 07:55 | |
| *** subscope has joined #openstack-ansible | 07:56 | |
| *** git-harry has quit IRC | 07:59 | |
| *** jmccrory has quit IRC | 07:59 | |
| *** git-harry has joined #openstack-ansible | 08:01 | |
| *** jmccrory has joined #openstack-ansible | 08:06 | |
| *** pilgrimstack has joined #openstack-ansible | 08:07 | |
| mattt | pellaeon: keystone should be able to use a later oslo.i18n, i wonder if your git checkout is out of date | 08:09 |
| *** subscope has quit IRC | 08:09 | |
| odyssey4me | pellaeon which branch/tag did you checkout? | 08:15 |
| *** javeriak_ has quit IRC | 08:15 | |
| mattt | pellaeon: may be worht updating your repo server also | 08:15 |
| pellaeon | mattt: I need the fix in this commit https://review.openstack.org/#/c/204542/ | 08:17 |
| pellaeon | so I'm now on 6ac504 | 08:18 |
| pellaeon | I did run the repo-clone-mirror.yml playbook | 08:19 |
| andymccr | pellaeon: that should't change any of the requirements, so thats pretty odd. I had a fresh install yesterday which setup oslo.i18n (2.1.0) and keystone's requirements.txt suggests the version just needs to be >=1.5.0 | 08:20 |
| odyssey4me | pellaeon so there are two options for deployers - one to clone the upstream mirror, which for any branch may or may not work on a given day (only tags are stable) | 08:20 |
| odyssey4me | the other option is to build the repo locally instead, which is how the gate tests do it | 08:21 |
| odyssey4me | you'll notice a spec has been submitted to remove option one: https://review.openstack.org/203706 | 08:21 |
| pellaeon | odyssey4me: I see | 08:22 |
| pellaeon | andymccr: got some ideas on where I should check ? | 08:23 |
| pellaeon | hmm | 08:24 |
| pellaeon | I just did a `pip install -I keystone` and the issue was gone | 08:25 |
| *** subscope has joined #openstack-ansible | 08:25 | |
| *** jmccrory has quit IRC | 08:26 | |
| pellaeon | odyssey4me: thanks for the explanation though :-) | 08:26 |
| *** jmccrory has joined #openstack-ansible | 08:27 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Document required repository hosts config info https://review.openstack.org/202258 | 08:27 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Parse latest Tempest results https://review.openstack.org/204948 | 08:28 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Target AIO swift vars at specific containers https://review.openstack.org/204949 | 08:28 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Fix glance_nfs_client setting https://review.openstack.org/204954 | 08:52 |
| *** shausy has quit IRC | 09:01 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Fix glance_nfs_client setting https://review.openstack.org/204954 | 09:15 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Fix glance_nfs_client setting https://review.openstack.org/204542 | 10:03 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust SSH key creation method for repo servers https://review.openstack.org/204996 | 10:30 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust SSH key creation method for repo servers https://review.openstack.org/204996 | 10:45 |
| *** javeriak has joined #openstack-ansible | 10:49 | |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust SSH key creation method for Nova compute https://review.openstack.org/205008 | 11:09 |
| *** davidself has quit IRC | 11:44 | |
| *** tlian has joined #openstack-ansible | 12:13 | |
| *** javeriak has quit IRC | 12:14 | |
| *** javeriak has joined #openstack-ansible | 12:16 | |
| *** prad has joined #openstack-ansible | 12:41 | |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust SSH key creation method for Nova compute https://review.openstack.org/205008 | 12:49 |
| cloudnull | andymccr: ^ for that change can we do that for Keystone too? | 13:00 |
| cloudnull | its a much better way, good work on that | 13:01 |
| cloudnull | btw | 13:01 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Add swift-sync role and environment https://review.openstack.org/201136 | 13:04 |
| andymccr | cloudnull: yeh im doing keystone now :) got side tracked | 13:05 |
| andymccr | then i think thats the last one | 13:06 |
| cloudnull | yea i think so too | 13:06 |
| cloudnull | thats a great change! btw. simple and requires no additional scaffolding. | 13:06 |
| cloudnull | andymccr: os_horizon does it for ssl keys too it seems | 13:08 |
| cloudnull | do you think this could apply there too ? | 13:08 |
| cloudnull | playbooks/roles/os_horizon/tasks/horizon_ssl_key_distribute.yml:17 | 13:08 |
| andymccr | i'll take a look | 13:09 |
| javeriak | hey guys, question, whats the best way to inlclude a host to a predefined host_group, say i want to add compute hosts to the nova_metadata group | 13:12 |
| javeriak | i can't seem to figure out the connection to how they're defined in rpc_enviroment.yml to how they get placed in the rpc_inventory.json | 13:13 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Keystone Federation Service Provider Configuration https://review.openstack.org/194395 | 13:16 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust SSH key creation method for Keystone https://review.openstack.org/205056 | 13:17 |
| cloudnull | javeriak: add a belongs_to entry here https://github.com/stackforge/os-ansible-deployment/blob/juno/etc/rpc_deploy/rpc_environment.yml#L216-L220 for the container group your creating IE metadata_containers | 13:24 |
| cloudnull | then create two entries that look like this in the physical skel https://github.com/stackforge/os-ansible-deployment/blob/juno/etc/rpc_deploy/rpc_environment.yml#L296-L301 | 13:25 |
| cloudnull | for metadata_containers and metadata_hosts | 13:25 |
| odyssey4me | cloudnull andymccr the ssl cert distribution could perhaps be done better using the ansible synchronise module, but it won't quite be the same as the ssh keys as the ssh keys has a special ansible module :/ | 13:29 |
| andymccr | we're rsync'ing? | 13:30 |
| andymccr | happy to change that up too | 13:30 |
| andymccr | i found the synchronize module worked way better | 13:30 |
| andymccr | i was having issues with the ring sync on swift plays before | 13:30 |
| odyssey4me | andymccr no, we're not (although I think we might be for the fernet token keys?) | 13:30 |
| andymccr | grep -ir "rsync" * ?! happy times | 13:30 |
| cloudnull | i dont like the syncronize module in our case. it pulls to the deployment host | 13:30 |
| andymccr | cloudnull: you can "delegate_to" | 13:31 |
| javeriak | cloudnull: okay, leme give that a try, or perhaps its just easier to have the metadata playbook run on compute hosts as well | 13:31 |
| odyssey4me | for all the ssl cert distributions we're using memcache, with the exception of the case where a user-provided cert is there... in which case we copy from the deployment host to the targets | 13:31 |
| odyssey4me | so it's only for self-signed certs that we ship them through memcache | 13:32 |
| cloudnull | which makes sense. pulling back to the deployment host we have to assume the directory structure of the system that things are running on which we have no idea, could be windows, mac, bsd, linux. in the case of a user provided cert its user provided, so its known where they are. | 13:34 |
| odyssey4me | yep | 13:34 |
| cloudnull | because theres no try / finally syntax in ansible, at least none that im aware of, we cant guarentee that the cert pulled back to the host is removed from the deployment host | 13:35 |
| odyssey4me | if the memcache model can be replaced with a delegated synchronise, that'd be great - but I recall miguelgrinberg and sigmavirus24_awa both getting stuck when trying to get that right, so they switched to the memcache method | 13:35 |
| andymccr | cloudnull: you are correct, do or do not, there is no try. | 13:35 |
| odyssey4me | hahaha | 13:36 |
| cloudnull | hahah | 13:36 |
| odyssey4me | but yes, I do not like pulling things back to the deployment host - that brings bad juju | 13:36 |
| cloudnull | juju is also bad . | 13:36 |
| cloudnull | :) | 13:36 |
| andymccr | synchronize module can be done in a way that it doesnt ship back to the host | 13:36 |
| cloudnull | if that can happen im game for sure. | 13:36 |
| andymccr | you "delegate_to" one of the hosts | 13:37 |
| odyssey4me | go for it with the Horizon stuff that's recently merged in | 13:37 |
| andymccr | so it does "host[0] --> allotherhosts" | 13:37 |
| odyssey4me | the Keystone fixup of the SSL cert stuff is still pending :/ https://review.openstack.org/194474 | 13:37 |
| andymccr | https://review.openstack.org/#/c/201136/18/playbooks/roles/os_swift_sync/tasks/swift_rings_distribute.yml e.g. | 13:37 |
| odyssey4me | nice! | 13:38 |
| andymccr | just make sure youget the src/dest + push/pull mode correct or else massive fail ensues :D | 13:39 |
| cloudnull | thats sweet! | 13:39 |
| cloudnull | we should make that go too, then we can delete the carried module all together. | 13:40 |
| cloudnull | peoples please go review https://review.openstack.org/#/c/204712 it helps our project config when "DocImpact" is used. | 13:41 |
| andymccr | lgtm | 13:41 |
| pilgrimstack | hi all | 13:41 |
| openstackgerrit | Matt Thompson proposed stackforge/os-ansible-deployment-specs: Ceph block devices https://review.openstack.org/205062 | 13:42 |
| pilgrimstack | after downgrad like advised here, I still have the same problem | 13:42 |
| pilgrimstack | I have /etc/openstack_deploy/openstack_hostnames_ips.yml with null address | 13:43 |
| pilgrimstack | "osad_cinder_api_container-fc96b74b": { | 13:43 |
| pilgrimstack | "ansible_ssh_host": null, | 13:43 |
| pilgrimstack | "container_address": null | 13:43 |
| pilgrimstack | }, | 13:43 |
| pilgrimstack | someone has already see this case? | 13:45 |
| *** javeriak has quit IRC | 13:48 | |
| pilgrimstack | All containers are UP but there IPs are not listed in config files | 13:50 |
| *** KLevenstein has joined #openstack-ansible | 13:50 | |
| pilgrimstack | for information, I'm just after this step : openstack-ansible setup-hosts.yml | 13:51 |
| *** spotz_zzz is now known as spotz | 13:51 | |
| andymccr | pilgrimstack: are your networks specified correctly in your openstack_user_config.yml? (provider_networks, specifically the management network) | 13:55 |
| andymccr | also are the entries populated in the openstack_inventory.json? | 13:55 |
| pilgrimstack | null address too in openstack_inventory.json | 13:56 |
| pilgrimstack | provider_networks => i'm double-checking | 13:56 |
| andymccr | and another quesiton - are the ips for the physical hosts specified in the inventory/hostname_ips | 13:57 |
| *** yaya has joined #openstack-ansible | 13:57 | |
| *** javeriak has joined #openstack-ansible | 13:59 | |
| *** dgrauet has joined #openstack-ansible | 13:59 | |
| pilgrimstack | I don't have any file named inventory/hostname_ips | 14:05 |
| *** meteorfox has quit IRC | 14:06 | |
| *** meteorfox has joined #openstack-ansible | 14:06 | |
| andymccr | pilgrimstack: sorry i meant in your openstack_inventory or openstack_hostname_ips files | 14:07 |
| *** sigmavirus24_awa is now known as sigmavirus24 | 14:08 | |
| lbragstad | o/ ahoy ansible gurus! I think I remember asking this question before but unfortunately I don't remember the answer. What is the reason behind using `xtrabackup` versus `xtrabackup-v2` for galera's SST method? https://github.com/stackforge/os-ansible-deployment/blob/master/playbooks/roles/galera_server/templates/cluster.cnf.j2#L37 | 14:11 |
| andymccr | lbragstad: im not sure - from what i can see it should probably be xtrabackup-v2 now since we're on a version > 5.5.33-x | 14:13 |
| *** Mudpuppy has joined #openstack-ansible | 14:13 | |
| lbragstad | andymccr: gotcha, I had a feeling it was related to packaging | 14:13 |
| *** Mudpuppy has quit IRC | 14:13 | |
| lbragstad | andymccr: did you all try `rsync`? | 14:14 |
| lbragstad | as the sst method? | 14:14 |
| *** Mudpuppy has joined #openstack-ansible | 14:14 | |
| *** javeriak has quit IRC | 14:15 | |
| pilgrimstack | andymccr: Yes, I have the IP of the pysical hosts | 14:16 |
| andymccr | pilgrimstack: ok good - so it sounds like the mapping of networks isn't correct for the containers so it is failing on that. The settings to check for are in openstack_user_config.yml so the "cidr_networks" which should have the management network specified as a cidr, and then in provider_networks you should have an entry that references "ip_from_q: management" and has a group_binds of "all_containers" wit | 14:17 |
| andymccr | h the "is_ssh_address: true" setting. | 14:17 |
| pilgrimstack | checking ;) | 14:18 |
| pilgrimstack | all seems good :/ http://pastebin.com/4UD4xeH5 | 14:24 |
| cloudnull | Andymccr lbragstad https://review.openstack.org/#/c/178259/ is looking to move to mariadb 10 w/ xtrabackup-v2 as the default. At present the role is hardcoded to xtrabackup which is solved in that review. | 14:24 |
| cloudnull | Afk a bit running out to get stronger coffee. | 14:24 |
| andymccr | pilgrimstack: is 'br-mgmt' the name of the bridge/device on the physical hosts? | 14:25 |
| pilgrimstack | yep | 14:26 |
| andymccr | hmm | 14:26 |
| andymccr | so are the containers themselves getting ips in that range? | 14:26 |
| pilgrimstack | checking | 14:26 |
| lbragstad | cloudnull: awesome, that makes sense | 14:26 |
| andymccr | so if you did lxc-ls -f do the containers have an address in the 192.168.3.0 range | 14:27 |
| pilgrimstack | nop, on 10.0.3.0/24 ... | 14:28 |
| pilgrimstack | dnsmasq seems provide this range | 14:29 |
| pilgrimstack | I don't know why... | 14:29 |
| pilgrimstack | Did I mise the dnsmasq configuration? I didn't see something like this in the doc | 14:30 |
| andymccr | pilgrimstack: nope thats the default lxc ip range so thats normal | 14:30 |
| andymccr | but you should have a second ip if everything is working properly | 14:30 |
| pilgrimstack | every container should have eht0 on 10.0.3.0/24 and eth1 on the cidr_management, that's what you mean? | 14:32 |
| andymccr | correct | 14:32 |
| andymccr | e.g. lxc-ls -f on a host of mine with a management range of 10.241.0.0/24: galera_container-f3f16f1b RUNNING 10.0.3.146, 10.241.0.227 | 14:32 |
| pilgrimstack | mmmh, strange, it seems that my containers are lazy and don't what to use the second IF | 14:34 |
| andymccr | well the inventory would get created first (which tells the containers which ip to take) so im guessing the issue is further up than the containers themselves | 14:36 |
| pilgrimstack | Can you tell me what you have in your /etc/lxc/lxc-openstack.conf, plz ? | 14:37 |
| andymccr | just the default eth0 lxc network | 14:37 |
| pilgrimstack | ok | 14:38 |
| pellaeon | Hello i've been having trouble with my keystone_container | 14:39 |
| pellaeon | it always stuck at the task os_keystone | Ensure admin tenant | 14:39 |
| pellaeon | error message: http://pastebin.com/gWFis9vX | 14:40 |
| andymccr | pellaeon: are you using the setup-everything.yml play? | 14:40 |
| lbragstad | looks like it can't find the domain to use? | 14:41 |
| *** TheIntern has joined #openstack-ansible | 14:41 | |
| pellaeon | andymccr: it will fail no matter I use the setup-everything.yml or os-keystone-install.yml | 14:41 |
| andymccr | lbragstad: yeah i think its because the setup-everything.yml play doesn't include haproxy (since we considered haproxy to be for testing/dev) | 14:41 |
| andymccr | so it will be trying to connect to a VIP that doesnt exist | 14:41 |
| lbragstad | andymccr: ahh, interesting | 14:41 |
| pellaeon | wait, some more details | 14:42 |
| pellaeon | I checked /var/log/keystone/ssl_access.log on the keystone containers | 14:42 |
| pellaeon | it showed this: | 14:43 |
| pellaeon | "GET /v2.0/domains HTTP/1.1" 404 344 "-" "python-keystoneclient" | 14:43 |
| pellaeon | I do indeed use haproxy btw | 14:44 |
| mattt | pellaeon: when you say get stuck, is it literally timing out or bailing w/ a specific message? | 14:44 |
| pellaeon | keystone domain API is in V3, V2 doesn't have that api | 14:44 |
| andymccr | pellaeon: so the haproxy install doesnt happen as part of setup-eveyrthing.yml, so try run the haproxy-install.yml playbook | 14:44 |
| andymccr | and then the keystone-install.yml playbook | 14:44 |
| andymccr | i imagine that will work | 14:44 |
| pellaeon | mattt: not timing out, it spits out error message: http://pastebin.com/gWFis9vX | 14:45 |
| mattt | pellaeon: can you rerun the haproxy-install.yml playbook just to rule that out? | 14:48 |
| pellaeon | mattt: running | 14:50 |
| mattt | pellaeon: can you also check your openrc in the keystone container to make sure it's using keystone v3 ? | 14:50 |
| pellaeon | mattt: oh-uh, it's using V2... | 14:51 |
| pellaeon | export OS_IDENTITY_API_VERSION=2 | 14:51 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Adjust SSH key creation method for Keystone https://review.openstack.org/205056 | 14:52 |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Fix haproxy service config when ssl is enabled https://review.openstack.org/202911 | 14:52 |
| pilgrimstack | andymccr: OMG… syntax problem, wrong indation in my yaml… | 14:54 |
| pilgrimstack | lot of comment beween two section and indentation broken | 14:55 |
| pilgrimstack | sorry for this time on this stupid thing | 14:55 |
| andymccr | haha no problem pilgrimstack - glad its fixed at least, if it wasnt something like that i was getting to the point where i had no ideas left :) | 14:56 |
| *** KLevenstein has quit IRC | 14:56 | |
| mattt | pellaeon: you're running master right? did you iniitially deploy some time again, run into issues, re-pull your checkout and try again? | 14:56 |
| mattt | trying to understand what's going on here :) | 14:56 |
| pellaeon | mattt: until yesterday I was running kilo branch | 14:57 |
| mattt | pellaeon: did you get a successful deploy on kilo? | 14:58 |
| pellaeon | I originally added include haproxy-install.yml to setup-infrastructure.yml | 14:58 |
| odyssey4me | pellaeon haproxy-install must run after setup-infrastructure, or at least at the end of it | 14:59 |
| pellaeon | but I now realized that it was overwritten by yesterday's checkout to master | 14:59 |
| pellaeon | It did deploy, though not fully working, I'm still tweaking the user_config | 15:00 |
| pellaeon | I mean, it's probably not the playbooks' fault, but my configuration | 15:00 |
| *** KLevenstein has joined #openstack-ansible | 15:01 | |
| pellaeon | odyssey4me: it's been quite some time I forgot where I put it actually, but I did follow the sequence on a post explaining haproxy on the forum, which I couldn't find now | 15:01 |
| odyssey4me | pellaeon a good way to know the process is to look at scripts/run-playbooks.sh which is the way the gate checks are done | 15:02 |
| pellaeon | odyssey4me: I see | 15:04 |
| *** weezS has joined #openstack-ansible | 15:05 | |
| *** KLevenstein has quit IRC | 15:05 | |
| pellaeon | now i'm re-running setup-hosts.yml, so it might take a while | 15:05 |
| *** KLevenstein has joined #openstack-ansible | 15:05 | |
| pellaeon | https://community.rackspace.com/products/f/45/t/4326 yep, this one's the haproxy post | 15:06 |
| pellaeon | now running setup-infrastructure... | 15:06 |
| *** fawadkhaliq has joined #openstack-ansible | 15:07 | |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Document required repository hosts config info https://review.openstack.org/205100 | 15:08 |
| *** vdo has quit IRC | 15:10 | |
| pellaeon | found this line in user_group_vars.yml : | 15:12 |
| pellaeon | keystone_service_internalurl: "{{ keystone_service_internaluri }}/v2.0" | 15:12 |
| pellaeon | which is the cause for openrc to export OS_IDENTITY_API_VERSION=2 | 15:13 |
| pellaeon | hmm I see you moved user_group_vars to playbooks/inventory/group_vars/hosts.yml | 15:15 |
| *** dgrauet has quit IRC | 15:17 | |
| *** gregdek has quit IRC | 15:23 | |
| *** javeriak has joined #openstack-ansible | 15:27 | |
| *** Bjoern_ has joined #openstack-ansible | 15:31 | |
| *** javeriak has quit IRC | 15:34 | |
| *** javeriak_ has joined #openstack-ansible | 15:38 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Adjust SSH key creation method for Nova compute https://review.openstack.org/205008 | 15:44 |
| pellaeon | ran into another weird issue when running haproxy-install fatal: [haproxy1] => {'msg': 'AnsibleError: host not found: {', 'failed': True} | 15:53 |
| *** yaya has quit IRC | 15:58 | |
| cloudnull | meeting cloudnull, mattt, andymccr, d34dh0r53, hughsaunders, b3rnard0, palendae, Sam-I-Am, odyssey4me, serverascode, rromans, mancdaz, dolphm, _shaps_, BjoernT, claco, echiu, dstanek, jwagner, ayoung | 16:00 |
| cloudnull | in #openstack-meeting-4 | 16:00 |
| odyssey4me | sigmavirus24 as it turns out, I was wrong - horizon is actually behaving... the wrong endpoint usage was due to some sort of caching from before I changed the endpoint | 16:01 |
| *** yaya has joined #openstack-ansible | 16:02 | |
| odyssey4me | sigmavirus24 what it does mean though is that we need to ensure that the public endpoint is also v3, not v2 as it is currently - is there a reason why we only set the admin endpoint as v3 and left the internal/public endpoints as v2? | 16:02 |
| cloudnull | Meeting information https://wiki.openstack.org/wiki/Meetings/openstack-ansible | 16:02 |
| *** weezS has quit IRC | 16:03 | |
| sigmavirus24 | sorry | 16:03 |
| sigmavirus24 | odyssey4me: that was changed by someone else last minute before my patch was approved to avoid a merge conflict | 16:04 |
| sigmavirus24 | Like I said in our internal meeting, don't pull that shit | 16:04 |
| odyssey4me | sigmavirus24 ah, so the intent was to set them all to the v3 endpoint then? | 16:04 |
| sigmavirus24 | yes | 16:05 |
| odyssey4me | cool, thanks - I'll fix that up then | 16:05 |
| sigmavirus24 | :thumbsup: | 16:05 |
| *** TheIntern has quit IRC | 16:07 | |
| *** serverascode has quit IRC | 16:14 | |
| *** serverascode has joined #openstack-ansible | 16:16 | |
| *** Bjoern_ has quit IRC | 16:18 | |
| cloudnull | #topic Reviews | 16:18 |
| cloudnull | ha wrong windo | 16:18 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Keystone Federation Service Provider Configuration https://review.openstack.org/194395 | 16:18 |
| * cloudnull uses wrong window and can't spell | 16:19 | |
| b3rnard0 | yeah it's spelled winder | 16:19 |
| *** Bjoern_ has joined #openstack-ansible | 16:22 | |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust SSH key creation method for Nova compute https://review.openstack.org/205143 | 16:26 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust SSH key creation method for Keystone https://review.openstack.org/205144 | 16:26 |
| openstackgerrit | Andy McCrae proposed stackforge/os-ansible-deployment: Adjust SSH key creation method for repo servers https://review.openstack.org/205145 | 16:26 |
| *** weezS has joined #openstack-ansible | 16:26 | |
| *** pilgrimstack has quit IRC | 16:36 | |
| *** weezS has quit IRC | 16:40 | |
| *** TheIntern has joined #openstack-ansible | 16:43 | |
| *** TheIntern has quit IRC | 16:45 | |
| *** yaya has quit IRC | 16:46 | |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Updated kilo for new dev work - 07.20.2015 https://review.openstack.org/205155 | 16:50 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Updated kilo for new dev work - 07.23.2015 https://review.openstack.org/205155 | 16:51 |
| *** yaya has joined #openstack-ansible | 16:56 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Disable Unattended Upgrades https://review.openstack.org/203858 | 17:07 |
| *** alop has joined #openstack-ansible | 17:09 | |
| *** fawadkhaliq has quit IRC | 17:11 | |
| *** pilgrimstack has joined #openstack-ansible | 17:11 | |
| *** fawadkhaliq has joined #openstack-ansible | 17:11 | |
| *** javeriak has joined #openstack-ansible | 17:25 | |
| odyssey4me | sigmavirus24 cloudnull so we're moving wholly towards making use of v3 identity endpoints, but this won't work properly if the old v2 endpoints are still in the catalog right? | 17:28 |
| odyssey4me | do we need an upgrade action that removes the old ones? | 17:28 |
| sigmavirus24 | odyssey4me: hm | 17:29 |
| sigmavirus24 | that depends on what the user wants to happen after an upgrade | 17:29 |
| *** javeriak_ has quit IRC | 17:29 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Added gate process and more logs https://review.openstack.org/204315 | 17:31 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Fix Keystone URI/URL defaults https://review.openstack.org/205192 | 17:33 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Keystone Federation Service Provider Configuration https://review.openstack.org/194395 | 17:34 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Updated master for new dev work - 07.20.2015 https://review.openstack.org/199126 | 17:37 |
| odyssey4me | sigmavirus24 cloudnull better? https://review.openstack.org/205192 | 17:38 |
| openstackgerrit | Kevin Carter proposed stackforge/os-ansible-deployment: Add sample Keystone Federation SP configuration for ADFS https://review.openstack.org/203736 | 17:39 |
| *** TheIntern has joined #openstack-ansible | 17:43 | |
| openstackgerrit | Matthew Thode proposed stackforge/os-ansible-deployment: Add rpc tunables for nova and neutron https://review.openstack.org/205204 | 18:04 |
| openstackgerrit | Matthew Thode proposed stackforge/os-ansible-deployment: Add rpc tunables for nova and neutron https://review.openstack.org/205205 | 18:05 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Keystone Federation Service Provider Configuration https://review.openstack.org/194395 | 18:43 |
| cloudnull | odyssey4me: lgtm | 18:46 |
| cloudnull | so the gate is down it seems. looks like infra will be working on it some more today/night | 18:49 |
| odyssey4me | cloudnull yep | 18:50 |
| cloudnull | that said, cores pleases review and setup commits that we are looking to wrap up for kilo/juno | 18:50 |
| cloudnull | obviously holding +w for gate, but reviews for inflight changes would be good. | 18:51 |
| *** bitblt has joined #openstack-ansible | 18:55 | |
| *** davi8784 has joined #openstack-ansible | 18:57 | |
| *** bitblt has quit IRC | 18:58 | |
| *** TheIntern has quit IRC | 18:59 | |
| *** annashen has joined #openstack-ansible | 18:59 | |
| *** javeriak has quit IRC | 19:04 | |
| *** javeriak has joined #openstack-ansible | 19:06 | |
| *** galstrom_zzz is now known as galstrom | 19:09 | |
| *** javeriak has quit IRC | 19:10 | |
| *** javeriak has joined #openstack-ansible | 19:11 | |
| *** davi8784 has quit IRC | 19:17 | |
| prometheanfire | so much duplication for for configs in the juno branch, where's the right place for nova server? | 19:20 |
| *** annashen has quit IRC | 19:31 | |
| *** annashen has joined #openstack-ansible | 19:32 | |
| cloudnull | prometheanfire: which nova config are you looking for ? | 19:38 |
| prometheanfire | nvm | 19:38 |
| cloudnull | you probably want https://github.com/stackforge/os-ansible-deployment/blob/juno/rpc_deployment/roles/nova_common/templates/nova.conf | 19:39 |
| prometheanfire | grep -r crosses symlinks | 19:39 |
| cloudnull | yes thats a confusing bit in Juno for sure. | 19:39 |
| prometheanfire | grep -d recurse doesn't | 19:39 |
| prometheanfire | helps a bit | 19:39 |
| openstackgerrit | Matthew Thode proposed stackforge/os-ansible-deployment: Add prc tunables to nova and neutron https://review.openstack.org/205256 | 19:41 |
| sigmavirus24 | prometheanfire: prc? | 19:41 |
| palendae | rpc | 19:42 |
| palendae | -2 | 19:42 |
| sigmavirus24 | -2 | 19:42 |
| sigmavirus24 | -4 | 19:42 |
| sigmavirus24 | -6 | 19:42 |
| sigmavirus24 | -10 | 19:42 |
| prometheanfire | LOL | 19:42 |
| sigmavirus24 | prometheanfire: what connections to China do you have? | 19:42 |
| prometheanfire | letmefixthatrealquick | 19:42 |
| sigmavirus24 | no | 19:42 |
| sigmavirus24 | -2 | 19:42 |
| openstackgerrit | Matthew Thode proposed stackforge/os-ansible-deployment: Add rpc tunables to nova and neutron https://review.openstack.org/205256 | 19:43 |
| prometheanfire | better? | 19:43 |
| palendae | no | 19:43 |
| prometheanfire | ok | 19:43 |
| prometheanfire | fine with me | 19:43 |
| sigmavirus24 | prometheanfire: btw | 19:44 |
| sigmavirus24 | `git review -X CHANGE_ID` is the preferred way to backport | 19:44 |
| palendae | prometheanfire: I had one more comment on it, otherwise it looks good | 19:44 |
| sigmavirus24 | (includes the original commit sha) | 19:44 |
| openstackgerrit | Matthew Thode proposed stackforge/os-ansible-deployment: Add rpc tunables for nova and neutron https://review.openstack.org/205204 | 19:47 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Keystone Federation Identity Provider Configuration https://review.openstack.org/194259 | 19:48 |
| prometheanfire | palendae: s/rpc/remote procedure call/ ? | 19:48 |
| palendae | prometheanfire: Nah; the question about source for the values | 19:48 |
| prometheanfire | palendae: the source are defaults | 19:49 |
| prometheanfire | palendae: I updated the master one | 19:49 |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Add tunables for nova and neutron https://review.openstack.org/205204 | 19:49 |
| palendae | prometheanfire: Right, could you document where that's stated? | 19:49 |
| palendae | Just looking at that file, I have no proof that that's really the upstream default | 19:49 |
| palendae | Similar to Jesse's question | 19:49 |
| palendae | odyssey4me: Why update that commit when prometheanfire is here? | 19:50 |
| prometheanfire | odyssey4me: :( | 19:50 |
| odyssey4me | palendae 'cuz he didn't fix it the first time :p | 19:50 |
| palendae | :| | 19:50 |
| prometheanfire | odyssey4me: it was commented on the first time? | 19:51 |
| odyssey4me | haha, fix away if you have another revision | 19:51 |
| odyssey4me | prometheanfire yes | 19:51 |
| prometheanfire | odyssey4me: why remove it? | 19:51 |
| palendae | Maybe I'm the odd one out, just changing someone's in-progress patch without saying anything is kind of...rude | 19:51 |
| odyssey4me | prometheanfire https://review.openstack.org/#/c/205204/1//COMMIT_MSG,cm | 19:51 |
| prometheanfire | we are specifically adding rpc tunables | 19:51 |
| palendae | *remote procedure call* | 19:51 |
| palendae | Not rackspace private cloud | 19:51 |
| prometheanfire | palendae: agreed | 19:52 |
| prometheanfire | about the in progress thing :( | 19:52 |
| odyssey4me | ah fair enough, sorry for my rudeness | 19:52 |
| palendae | Yeah, we do that a lot | 19:52 |
| prometheanfire | also, ok, about rackspace private cloud | 19:52 |
| palendae | This one cause a lot of confusion | 19:52 |
| palendae | Because I think people saw "RPC" and immediately thought private cloud | 19:52 |
| palendae | But it's RPC settings...that RPC would like to configure | 19:52 |
| odyssey4me | it was a rushed response on a train - apologies | 19:52 |
| prometheanfire | odyssey4me: np, just ping next time :P | 19:53 |
| odyssey4me | sure, perhaps answer within the review next time :p | 19:53 |
| palendae | I hope we don't make it a habit to just change contributor's patches | 19:54 |
| palendae | But that'll be my last comment | 19:54 |
| openstackgerrit | Matthew Thode proposed stackforge/os-ansible-deployment: Add remote procedure call tunables to nova and neutron https://review.openstack.org/205256 | 19:55 |
| prometheanfire | palendae: does the juno review look right now? | 19:55 |
| odyssey4me | lol prometheanfire almost - you have a different change id, and have thus faked the cherry-pick :p | 19:56 |
| odyssey4me | but meh - I would have expected the juno branch not to have been a cherry-pick anyway | 19:57 |
| prometheanfire | I do? | 19:57 |
| prometheanfire | oh, that's not a commit hash is it | 19:57 |
| prometheanfire | ok | 19:57 |
| odyssey4me | the change id and the commit hash are not the same | 19:57 |
| palendae | Yeah, it's a new change id | 19:57 |
| palendae | ChangeId tracks a given change across multiple commits | 19:58 |
| odyssey4me | your commit has is right, but the change id should be common if it's an actual cherry-pick | 19:58 |
| openstackgerrit | Matthew Thode proposed stackforge/os-ansible-deployment: Add remote procedure call tunables to nova and neutron https://review.openstack.org/205256 | 19:58 |
| *** javeriak has quit IRC | 20:01 | |
| openstackgerrit | Matthew Thode proposed stackforge/os-ansible-deployment: Add tunables for nova and neutron https://review.openstack.org/205270 | 20:06 |
| prometheanfire | ok, there's juno and kilo | 20:06 |
| odyssey4me | prometheanfire you can update the master commit message to match :) | 20:07 |
| prometheanfire | sure, whynot | 20:07 |
| prometheanfire | to match what? | 20:08 |
| prometheanfire | actually | 20:08 |
| odyssey4me | match the juno title - it makes sense | 20:08 |
| odyssey4me | is less vague | 20:08 |
| prometheanfire | ah, ok didn't know which was prefered :D | 20:08 |
| odyssey4me | (and apologies for breaking it in the first place) | 20:09 |
| prometheanfire | np, I think I saw that comment at one point but forgot | 20:09 |
| openstackgerrit | Matthew Thode proposed stackforge/os-ansible-deployment: Add remote procedure call tunables to nova and neutron https://review.openstack.org/205204 | 20:10 |
| openstackgerrit | Matthew Thode proposed stackforge/os-ansible-deployment: Add remote procedure call tunables to nova and neutron https://review.openstack.org/205270 | 20:11 |
| prometheanfire | and done | 20:11 |
| *** prad has quit IRC | 20:37 | |
| *** markvoelker has quit IRC | 20:42 | |
| *** jwagner is now known as jwagner_away | 20:50 | |
| *** yaya has quit IRC | 21:01 | |
| openstackgerrit | Jesse Pretorius proposed stackforge/os-ansible-deployment: Document required repository hosts config info https://review.openstack.org/205100 | 21:27 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 21:43 | |
| *** TheIntern has joined #openstack-ansible | 21:44 | |
| *** fawadk has joined #openstack-ansible | 21:47 | |
| *** fawadkhaliq has quit IRC | 21:48 | |
| *** Mudpuppy has quit IRC | 21:51 | |
| * cloudnull out | 22:07 | |
| cloudnull | have a good one. | 22:07 |
| *** cloudnull is now known as cloudull_zzz | 22:08 | |
| *** KLevenstein has quit IRC | 22:15 | |
| *** TheIntern has quit IRC | 22:19 | |
| *** annashen has quit IRC | 22:25 | |
| *** fawadk has quit IRC | 22:31 | |
| *** annashen has joined #openstack-ansible | 22:32 | |
| openstackgerrit | Merged stackforge/os-ansible-deployment: Add remote procedure call tunables to nova and neutron https://review.openstack.org/205204 | 22:39 |
| *** CheKoLyN has joined #openstack-ansible | 22:40 | |
| *** CheKoLyN has quit IRC | 22:40 | |
| *** spotz is now known as spotz_zzz | 22:43 | |
| *** galstrom is now known as galstrom_zzz | 22:52 | |
| *** Bjoern_ has quit IRC | 23:10 | |
| *** annashen has quit IRC | 23:31 | |
| *** pilgrimstack has quit IRC | 23:42 | |
| *** darrenc is now known as darrenc_afk | 23:44 | |
| *** darrenc_afk is now known as darrenc | 23:54 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!