| *** huikang has quit IRC | 00:05 | |
| *** huikang has joined #openstack-kolla | 00:11 | |
| *** Pavo has quit IRC | 00:21 | |
| *** tonanhngo has joined #openstack-kolla | 00:25 | |
| *** Pavo has joined #openstack-kolla | 00:26 | |
| *** tonanhngo has quit IRC | 00:26 | |
| *** hfu has joined #openstack-kolla | 00:28 | |
| *** huikang has quit IRC | 00:37 | |
| *** v1k0d3n has joined #openstack-kolla | 00:37 | |
| *** severion has joined #openstack-kolla | 00:39 | |
| *** v1k0d3n has quit IRC | 00:43 | |
| *** tonanhngo has joined #openstack-kolla | 00:43 | |
| *** tonanhngo has quit IRC | 00:44 | |
| *** v1k0d3n has joined #openstack-kolla | 00:45 | |
| *** severion has quit IRC | 00:48 | |
| *** tonanhngo has joined #openstack-kolla | 01:04 | |
| *** tonanhngo has quit IRC | 01:05 | |
| *** huikang has joined #openstack-kolla | 01:09 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Add octavia docker image https://review.openstack.org/399896 | 01:12 |
|---|---|---|
| kollabot1 | Gerrit Code Review | 01:12 |
| *** zhubingbing has joined #openstack-kolla | 01:16 | |
| *** ccesario has quit IRC | 01:26 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: DO_NOT_MERGE: TEST MASTER BRANCH https://review.openstack.org/399897 | 01:57 |
| kollabot1 | Gerrit Code Review | 01:57 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Run init-once in kolla-ansible folder https://review.openstack.org/399898 | 02:01 |
| kollabot1 | Gerrit Code Review | 02:01 |
| *** tonanhngo has joined #openstack-kolla | 02:03 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Add octavia docker image https://review.openstack.org/399896 | 02:04 |
| kollabot1 | Gerrit Code Review | 02:04 |
| *** tonanhngo has quit IRC | 02:04 | |
| *** ccesario has joined #openstack-kolla | 02:17 | |
| *** Pavo has quit IRC | 02:21 | |
| *** tonanhngo has joined #openstack-kolla | 02:24 | |
| *** Pavo has joined #openstack-kolla | 02:25 | |
| *** tonanhngo has quit IRC | 02:26 | |
| *** rhallisey has quit IRC | 02:33 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla-ansible: Enable cinder backup tab in horizon https://review.openstack.org/399900 | 02:41 |
| kollabot1 | Gerrit Code Review | 02:41 |
| *** ccesario has quit IRC | 02:44 | |
| *** tonanhngo has joined #openstack-kolla | 02:44 | |
| *** tonanhngo has quit IRC | 02:46 | |
| *** Jeffrey4l has quit IRC | 02:53 | |
| *** ccesario has joined #openstack-kolla | 02:56 | |
| *** tonanhngo has joined #openstack-kolla | 03:04 | |
| *** tonanhngo has quit IRC | 03:06 | |
| *** tonanhngo has joined #openstack-kolla | 03:24 | |
| *** tonanhngo has quit IRC | 03:26 | |
| *** fragatina has joined #openstack-kolla | 03:32 | |
| *** dave-mccowan has joined #openstack-kolla | 03:48 | |
| *** dave-mccowan has quit IRC | 03:52 | |
| *** tonanhngo has joined #openstack-kolla | 04:11 | |
| *** tonanhngo has quit IRC | 04:13 | |
| *** Pavo has quit IRC | 04:21 | |
| *** eaguilar has quit IRC | 04:23 | |
| *** Pavo has joined #openstack-kolla | 04:26 | |
| *** tonanhngo has joined #openstack-kolla | 04:26 | |
| *** tonanhngo has quit IRC | 04:29 | |
| *** tonanhngo has joined #openstack-kolla | 04:45 | |
| *** huikang has quit IRC | 04:46 | |
| *** tonanhngo has quit IRC | 04:46 | |
| *** tonanhngo has joined #openstack-kolla | 04:58 | |
| *** tonanhngo has quit IRC | 04:59 | |
| *** Jeffrey4l has joined #openstack-kolla | 05:01 | |
| *** tonanhngo has joined #openstack-kolla | 05:15 | |
| *** huikang has joined #openstack-kolla | 05:17 | |
| *** tonanhngo has quit IRC | 05:17 | |
| *** huikang has quit IRC | 05:18 | |
| *** zhubingbing has quit IRC | 05:29 | |
| *** unicell1 has joined #openstack-kolla | 05:59 | |
| *** unicell has quit IRC | 05:59 | |
| *** unicell has joined #openstack-kolla | 06:02 | |
| *** unicell1 has quit IRC | 06:04 | |
| *** AnswerGuy has quit IRC | 06:05 | |
| *** unicell1 has joined #openstack-kolla | 06:10 | |
| *** unicell has quit IRC | 06:11 | |
| *** tonanhngo has joined #openstack-kolla | 06:14 | |
| *** tonanhngo has quit IRC | 06:17 | |
| *** Pavo has quit IRC | 06:21 | |
| *** Pavo has joined #openstack-kolla | 06:25 | |
| *** tonanhngo has joined #openstack-kolla | 06:35 | |
| *** tonanhngo has quit IRC | 06:38 | |
| *** tonanhngo has joined #openstack-kolla | 06:54 | |
| *** tonanhngo has quit IRC | 06:55 | |
| *** tonanhngo has joined #openstack-kolla | 07:15 | |
| *** tonanhngo has quit IRC | 07:16 | |
| *** tonanhngo has joined #openstack-kolla | 07:35 | |
| *** tonanhngo has quit IRC | 07:38 | |
| *** senk has joined #openstack-kolla | 07:49 | |
| *** newmember has joined #openstack-kolla | 07:54 | |
| *** tonanhngo has joined #openstack-kolla | 07:54 | |
| *** tonanhngo has quit IRC | 07:57 | |
| *** x220_ has joined #openstack-kolla | 07:59 | |
| *** newmember has quit IRC | 08:01 | |
| *** sbezverk_ has joined #openstack-kolla | 08:03 | |
| *** sbezverk has quit IRC | 08:03 | |
| *** tonanhngo has joined #openstack-kolla | 08:09 | |
| *** tonanhngo has quit IRC | 08:11 | |
| *** Pavo has quit IRC | 08:21 | |
| *** Pavo has joined #openstack-kolla | 08:25 | |
| *** tonanhngo has joined #openstack-kolla | 08:27 | |
| *** tonanhngo has quit IRC | 08:27 | |
| *** nihilifer has joined #openstack-kolla | 08:44 | |
| *** tonanhngo has joined #openstack-kolla | 08:45 | |
| *** tonanhngo has quit IRC | 08:47 | |
| *** tonanhngo has joined #openstack-kolla | 09:04 | |
| *** tonanhngo has quit IRC | 09:05 | |
| *** senk has quit IRC | 09:06 | |
| *** x220_ has quit IRC | 09:10 | |
| *** tonanhngo has joined #openstack-kolla | 09:24 | |
| *** tonanhngo has quit IRC | 09:27 | |
| *** tonanhngo has joined #openstack-kolla | 09:45 | |
| *** tonanhngo has quit IRC | 09:45 | |
| *** mgoddard has joined #openstack-kolla | 09:50 | |
| *** papacz has joined #openstack-kolla | 09:54 | |
| *** mgoddard has quit IRC | 09:54 | |
| *** senk has joined #openstack-kolla | 09:59 | |
| *** magicboiz has joined #openstack-kolla | 10:01 | |
| *** tonanhngo has joined #openstack-kolla | 10:04 | |
| *** tonanhngo has quit IRC | 10:05 | |
| *** magicboiz has quit IRC | 10:08 | |
| *** oxkipo has joined #openstack-kolla | 10:15 | |
| oxkipo | Hi I cannot attach a cinder lvm volume to a instance. Also kolla/ubuntu-source-iscsid:3.0.1 this containers status is only created. Please may someone help me? | 10:16 |
| *** Pavo has quit IRC | 10:21 | |
| *** tonanhngo has joined #openstack-kolla | 10:25 | |
| *** Pavo has joined #openstack-kolla | 10:26 | |
| *** tonanhngo has quit IRC | 10:26 | |
| oxkipo | please may someone help me with this? | 10:28 |
| *** tonanhngo has joined #openstack-kolla | 10:45 | |
| *** tonanhngo has quit IRC | 10:47 | |
| oxkipo | please may someone help me with cinder? | 10:52 |
| *** papacz1 has joined #openstack-kolla | 11:01 | |
| *** papacz2 has joined #openstack-kolla | 11:04 | |
| *** papacz has quit IRC | 11:04 | |
| *** tonanhngo has joined #openstack-kolla | 11:05 | |
| *** papacz1 has quit IRC | 11:06 | |
| *** tonanhngo has quit IRC | 11:06 | |
| *** hfu has quit IRC | 11:13 | |
| *** hfu has joined #openstack-kolla | 11:15 | |
| *** hfu has quit IRC | 11:16 | |
| *** oxkipo has quit IRC | 11:16 | |
| *** hfu has joined #openstack-kolla | 11:16 | |
| *** hfu has quit IRC | 11:16 | |
| *** hfu has joined #openstack-kolla | 11:17 | |
| *** hfu has quit IRC | 11:17 | |
| *** tonanhngo has joined #openstack-kolla | 11:35 | |
| *** tonanhngo has quit IRC | 11:38 | |
| *** kristian2709 has joined #openstack-kolla | 11:43 | |
| kristian2709 | Hi please may someone help me fix cinder not assigning volumes to an instance? | 11:45 |
| *** tonanhngo has joined #openstack-kolla | 11:55 | |
| *** tonanhngo has quit IRC | 11:58 | |
| *** senk has quit IRC | 12:02 | |
| *** portdirect_away is now known as portdirect | 12:11 | |
| portdirect | kristian2709: sup? | 12:11 |
| *** david-lyle has joined #openstack-kolla | 12:12 | |
| kristian2709 | portdirect: I cannot attach a volume to the instance. It doesnt give me any error | 12:12 |
| portdirect | kristian2709: ok - at what point do you think its failing? can you access the logs for cinder (what backend are you using - my experience with ceph is very limited :( ) and nova-compute for the node running the vm? | 12:14 |
| *** tonanhngo has joined #openstack-kolla | 12:15 | |
| *** david-lyle has quit IRC | 12:17 | |
| *** tonanhngo has quit IRC | 12:17 | |
| kristian2709 | portdirect: my kolla installation is aio on a bare metal node with cinder and lvm configured on host | 12:20 |
| *** Pavo has quit IRC | 12:21 | |
| portdirect | kristian2709: ok, can you look at the logs for cinder volume, and nova-compute and see at what point it is failing? | 12:22 |
| kristian2709 | ok | 12:22 |
| *** hfu has joined #openstack-kolla | 12:24 | |
| *** Pavo has joined #openstack-kolla | 12:26 | |
| kristian2709 | portdirect: strange now it works | 12:26 |
| kristian2709 | dunno how | 12:26 |
| kristian2709 | portdirect: please may you help me fix this issue? | 12:40 |
| kristian2709 | vhttps://bugs.launchpad.net/nova/+bug/1642419 | 12:40 |
| openstack | Launchpad bug 1642419 in OpenStack Compute (nova) "GPU Passthrough isn't working" [Medium,New] | 12:40 |
| kollabot1 | Bug #1642419 “GPU Passthrough isn't working” : Bugs : OpenStack Compute (nova) | 12:40 |
| *** tonanhngo has joined #openstack-kolla | 12:44 | |
| *** tonanhngo has quit IRC | 12:47 | |
| portdirect | kristian2709: whoop! thats great - I'll have a look at the bug now | 12:56 |
| portdirect | kristian2709: K - I've only done GPU passthrough with nova-docker before (and not with kolla) any possibility you could get the logs from nova-compute in a gist to look at as well - whihc may be able to give me a better idea about whether this is an issue at the kolla or nova end? | 13:00 |
| portdirect | kristian2709: you running this on a gaming rig!? | 13:03 |
| kristian2709 | portdirect: yes | 13:04 |
| *** tonanhngo has joined #openstack-kolla | 13:04 | |
| kristian2709 | also the logs should be there | 13:04 |
| portdirect | whoops - missed that - 2 sec | 13:06 |
| *** tonanhngo has quit IRC | 13:06 | |
| kristian2709 | np | 13:07 |
| portdirect | kristian2709: k, a few things I'd try: | 13:09 |
| kristian2709 | ok | 13:09 |
| kristian2709 | tell mw | 13:09 |
| kristian2709 | *me | 13:09 |
| portdirect | get nova_compute running in the host-pid namespace: https://github.com/openstack/kolla-ansible/blob/master/ansible/roles/nova/tasks/start_compute.yml#L49 | 13:10 |
| kollabot1 | kolla-ansible/start_compute.yml at master · openstack/kolla-ansible · GitHub | 13:10 |
| portdirect | also try both libvirt and nova_compute in the host ipc namespace | 13:10 |
| *** skramaja has quit IRC | 13:11 | |
| *** uclug has joined #openstack-kolla | 13:16 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Install mkfs.vfat tool in ironic-conductor image https://review.openstack.org/399957 | 13:16 |
| kollabot1 | Gerrit Code Review | 13:16 |
| *** hfu has quit IRC | 13:19 | |
| *** hfu has joined #openstack-kolla | 13:20 | |
| *** hfu has quit IRC | 13:20 | |
| *** hfu has joined #openstack-kolla | 13:21 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: DO_NOT_MERGE: TEST NEWTON BRANCH https://review.openstack.org/386981 | 13:21 |
| kollabot1 | Gerrit Code Review | 13:21 |
| *** hfu has quit IRC | 13:21 | |
| *** hfu has joined #openstack-kolla | 13:21 | |
| *** uclug has quit IRC | 13:22 | |
| *** hfu has quit IRC | 13:22 | |
| *** hfu has joined #openstack-kolla | 13:22 | |
| *** hfu has quit IRC | 13:23 | |
| *** hfu has joined #openstack-kolla | 13:23 | |
| *** hfu has quit IRC | 13:23 | |
| *** zhubingbing has joined #openstack-kolla | 13:26 | |
| *** eaguilar has joined #openstack-kolla | 13:29 | |
| *** sp__ has joined #openstack-kolla | 13:29 | |
| openstackgerrit | Merged openstack/kolla: Use upper constraints for all jobs in tox.ini https://review.openstack.org/384221 | 13:35 |
| kollabot1 | Gerrit Code Review | 13:35 |
| *** sdake has joined #openstack-kolla | 13:48 | |
| *** sdake_ has joined #openstack-kolla | 13:52 | |
| *** hfu has joined #openstack-kolla | 13:52 | |
| *** sdake has quit IRC | 13:53 | |
| *** sp__ has quit IRC | 13:58 | |
| *** imcsk8_ has quit IRC | 14:08 | |
| *** imcsk8 has joined #openstack-kolla | 14:08 | |
| *** zhubingbing has quit IRC | 14:10 | |
| *** rhallisey has joined #openstack-kolla | 14:11 | |
| *** david-lyle has joined #openstack-kolla | 14:14 | |
| *** tonanhngo has joined #openstack-kolla | 14:14 | |
| *** tonanhngo has quit IRC | 14:16 | |
| *** david-lyle has quit IRC | 14:18 | |
| *** Pavo has quit IRC | 14:21 | |
| *** Pavo has joined #openstack-kolla | 14:26 | |
| *** tonanhngo has joined #openstack-kolla | 14:34 | |
| *** tonanhngo has quit IRC | 14:36 | |
| *** senk has joined #openstack-kolla | 14:42 | |
| *** sdake has joined #openstack-kolla | 14:52 | |
| *** sdake has quit IRC | 14:54 | |
| *** sdake__ has joined #openstack-kolla | 14:54 | |
| *** tonanhngo has joined #openstack-kolla | 14:54 | |
| openstackgerrit | Merged openstack/kolla-ansible: typo error of keyring spelling https://review.openstack.org/399455 | 14:55 |
| kollabot1 | Gerrit Code Review | 14:55 |
| *** sdake__ has quit IRC | 14:55 | |
| *** tonanhngo has quit IRC | 14:55 | |
| *** sdake has joined #openstack-kolla | 14:55 | |
| *** sdake_ has quit IRC | 14:55 | |
| openstackgerrit | Merged openstack/kolla: Remove Ansible references from tox.ini https://review.openstack.org/399597 | 15:02 |
| kollabot1 | Gerrit Code Review | 15:02 |
| *** sdake_ has joined #openstack-kolla | 15:06 | |
| *** sdake has quit IRC | 15:10 | |
| *** hfu has quit IRC | 15:15 | |
| *** tonanhngo has joined #openstack-kolla | 15:16 | |
| *** hfu has joined #openstack-kolla | 15:16 | |
| *** hfu has quit IRC | 15:17 | |
| *** tonanhngo has quit IRC | 15:17 | |
| *** senk has quit IRC | 15:17 | |
| *** magicboiz has joined #openstack-kolla | 15:20 | |
| *** eaguilar has quit IRC | 15:28 | |
| *** tyrola has joined #openstack-kolla | 15:29 | |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla-kubernetes: Mariadb Kubernetes Operator https://review.openstack.org/399263 | 15:32 |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla-kubernetes: Kubernetes Operator base https://review.openstack.org/399262 | 15:32 |
| kollabot1 | Gerrit Code Review | 15:32 |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla-kubernetes: Change services dir to templates dir https://review.openstack.org/399971 | 15:32 |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla-kubernetes: Service and Mariadb operator template https://review.openstack.org/399972 | 15:32 |
| kollabot1 | Gerrit Code Review | 15:32 |
| kollabot1 | Gerrit Code Review | 15:32 |
| kollabot1 | Gerrit Code Review | 15:32 |
| *** eaguilar has joined #openstack-kolla | 15:38 | |
| *** eaguilar has quit IRC | 15:40 | |
| *** v1k0d3n has quit IRC | 15:45 | |
| *** v1k0d3n has joined #openstack-kolla | 15:45 | |
| *** eaguilar has joined #openstack-kolla | 15:48 | |
| *** eaguilar has quit IRC | 16:00 | |
| *** senk has joined #openstack-kolla | 16:06 | |
| portdirect | rhallisey: ping | 16:13 |
| *** tonanhngo has joined #openstack-kolla | 16:15 | |
| rhallisey | portdirect, hey | 16:15 |
| portdirect | sunday workers :/ | 16:15 |
| portdirect | looking at your operator work - can I play with any of it atm? | 16:16 |
| sbezverk_ | portdirect: with this weather on east coast, what else can you do ;-) | 16:16 |
| *** tonanhngo has quit IRC | 16:16 | |
| portdirect | come to scotland - I'll show you reall rain and misery :) | 16:17 |
| rhallisey | portdirect, ya sure | 16:17 |
| portdirect | any pointers? | 16:18 |
| sbezverk_ | portdirect: visiting Scotland is definetly on my to do list :-) | 16:18 |
| *** jrist has quit IRC | 16:19 | |
| rhallisey | portdirect, do you want to try it or develop it osme? | 16:20 |
| rhallisey | some* | 16:20 |
| portdirect | sbezverk_: hit me up anytime your through man, though I'd strongly recommed august for the Edinburgh Festival | 16:20 |
| rhallisey | the process to try it is a little annoying | 16:20 |
| *** jrist has joined #openstack-kolla | 16:20 | |
| rhallisey | you have to build the base | 16:20 |
| rhallisey | then the mariadb-op | 16:20 |
| rhallisey | then create the resources | 16:20 |
| rhallisey | I haven't fully tried every piece yet | 16:20 |
| rhallisey | because I don't think I have all of them done yet | 16:20 |
| rhallisey | I think I need a deployment for the operator | 16:21 |
| portdirect | rhallisey: figured :) | 16:21 |
| *** Pavo has quit IRC | 16:21 | |
| *** sdake_ has quit IRC | 16:21 | |
| *** Pavo has joined #openstack-kolla | 16:21 | |
| rhallisey | ya it's super rough atm | 16:21 |
| portdirect | that the early stages of a major WIP for you :) | 16:22 |
| rhallisey | ya | 16:22 |
| rhallisey | sbezverk_, how come we can't change the dir name? | 16:22 |
| rhallisey | helm patches haven't merged yet so I figured we'd make the change now | 16:22 |
| sbezverk_ | rhallisey: I am not saying we cannot, but if you check Kevin's PS for helm | 16:23 |
| portdirect | mostly interested to make sure that I know whats going on so when I charge into development I'm going the right direction. | 16:23 |
| rhallisey | sbezverk_, right he uses it | 16:23 |
| sbezverk_ | Kevin uses completely different approach | 16:23 |
| rhallisey | I'll check out what he has | 16:23 |
| Pavo | good morning everyone | 16:24 |
| sbezverk_ | rhallisey: and even that structure is not even finalized.. if changing folder is not a show stopper for you, maybe we should hold a bit.. | 16:25 |
| rhallisey | it's not the biggest deal, but it makes more sense to me in terms of structure | 16:26 |
| portdirect | rhallisey: I agree with that - as it's what they are - but makes it a bit tricky atm | 16:28 |
| rhallisey | sure | 16:28 |
| portdirect | Possibly the most usfull thing i could do right now is get kolla-build updated so we can build the images? | 16:30 |
| rhallisey | some things we need revolve around making sure the repos are good to work with both projects | 16:31 |
| portdirect | yup: I made a first stab here: https://review.openstack.org/#/c/395676/, but got some feedback from inc0 and kfox (which I agree with), though will leave tthe option of just building a single template dir for tings like koll-k8s only container builds | 16:33 |
| kollabot1 | Gerrit Code Review | 16:33 |
| *** tonanhngo has joined #openstack-kolla | 16:35 | |
| *** tonanhngo has quit IRC | 16:38 | |
| sbezverk_ | rhallisey: are you going to use helm to deploy a specific chart for a service example mariadb or some other form of template? | 16:42 |
| v1k0d3n | hey portdirect | 16:42 |
| v1k0d3n | is that halcyon stuff working out for you? do you think that would be good to write up in kolla docs as a decent dev solution for kolla? | 16:43 |
| v1k0d3n | sorry...kolla-k8s i mean | 16:43 |
| portdirect | v1k0d3n: halcyon's working great for me atm :) | 16:46 |
| *** g3ek has quit IRC | 16:46 | |
| *** haplo37 has quit IRC | 16:46 | |
| portdirect | though I've only really tried with with weave as that worked first time out of the box for me - i think using it was a basis flow local dev docs would be awesome | 16:47 |
| portdirect | *for not flow | 16:48 |
| rhallisey | sbezverk_, ya an operator will trigger a helm install | 16:53 |
| *** g3ek has joined #openstack-kolla | 16:55 | |
| *** haplo37 has joined #openstack-kolla | 16:56 | |
| v1k0d3n | operator as in person or operator container rhallisey ? | 16:56 |
| rhallisey | ugh that name | 16:56 |
| v1k0d3n | exactly :) | 16:56 |
| v1k0d3n | i am starting to wish i never brought up the subject. | 16:56 |
| portdirect | v1k0d3n: stop trolling :D | 16:56 |
| rhallisey | v1k0d3n, from what I understood about the context of sbezverk_, I'm referring to a kubernetes operator | 16:57 |
| v1k0d3n | i'm not. i'm seriously wondering, that's all. | 16:57 |
| rhallisey | we should re name it to admiral | 16:57 |
| v1k0d3n | ok | 16:57 |
| v1k0d3n | just checking | 16:57 |
| v1k0d3n | our workflow will be 100% developer driven. | 16:57 |
| v1k0d3n | so developer will push code. container registry sees commit tag, and build new release tag. helm is updated, and triggers CI. once CI completes, queues up for human to push-button deploy the update. | 16:58 |
| v1k0d3n | all developer-driven. | 16:58 |
| v1k0d3n | may even use some jira component to approach CD changes. | 16:59 |
| v1k0d3n | still thinking through these steps, but that's a general logic. | 16:59 |
| v1k0d3n | why i asked anyway. | 16:59 |
| portdirect | v1k0d3n: The only thing 'missing' from halcyon in my opionion at the moment is a helper script to set up creds for kubectl and helm on the machine running vagrant - I could knock up a quick bash script for that if your up for it? | 16:59 |
| rhallisey | gitcha | 17:00 |
| rhallisey | gotcha* | 17:00 |
| v1k0d3n | portdirect: absolutely. | 17:00 |
| v1k0d3n | would definitely accept that. | 17:00 |
| portdirect | 5 mins :) | 17:01 |
| kristian2709 | portdirect: still the same error with CPUID | 17:03 |
| sbezverk_ | rhallisey: then it make sense to sync up with Kevin about templates.. | 17:09 |
| rhallisey | ya I'll study his patch | 17:09 |
| rhallisey | I haven't yet | 17:09 |
| rhallisey | just a little, but I need to more | 17:09 |
| *** tyrola has quit IRC | 17:11 | |
| *** bmace has quit IRC | 17:11 | |
| *** bmace has joined #openstack-kolla | 17:11 | |
| *** senk has quit IRC | 17:12 | |
| *** tonanhngo has joined #openstack-kolla | 17:14 | |
| v1k0d3n | so can anyone help me to understand an AIO deployment i am trying to stand up. | 17:16 |
| *** tonanhngo has quit IRC | 17:16 | |
| v1k0d3n | i guess it's been a little while (and new version) since my last attempt. | 17:16 |
| v1k0d3n | i have two interfaces: em1 and em2 (ubuntu). em1 is the network_interface and em2 is the neutron_external_interface. | 17:17 |
| v1k0d3n | em1 is 192.168.70.25/24. | 17:17 |
| v1k0d3n | em2 does not have an ip of course, but is attached to a "public" subnet of 192.168.4.x/22. | 17:17 |
| v1k0d3n | after deployment, i configure a flat interface to physnet1. all good there. | 17:18 |
| v1k0d3n | then i want to create my pub-sub and create 192.168.4.0 net with a dhcp pool. hosts get an ip just fine...but for some reason i can't communicate to them. something i'm missing here? | 17:19 |
| v1k0d3n | 192.168.4 and 192.168.70 are attached to the same router. should be fine there. | 17:20 |
| v1k0d3n | to be clear...can't ping 192.168.4.x hosts in the same broadcast domain. | 17:21 |
| v1k0d3n | tell me i'm doing something dumb here, please? :) | 17:21 |
| *** tonanhngo has joined #openstack-kolla | 17:24 | |
| *** tonanhngo has quit IRC | 17:26 | |
| *** magicboiz has quit IRC | 17:33 | |
| *** tonanhngo has joined #openstack-kolla | 17:36 | |
| * v1k0d3n realizes that it's a lazy sunday | 17:37 | |
| portdirect | kristian2709: sorry dude, I'm not familiar enough with IOMMU to help you any further in a constructive manner :( | 17:40 |
| v1k0d3n | sbezverk_: are you around, or is it too late for you? | 17:44 |
| *** kwazar has quit IRC | 17:55 | |
| *** timss has quit IRC | 17:55 | |
| *** kwazar has joined #openstack-kolla | 17:56 | |
| *** timss has joined #openstack-kolla | 17:56 | |
| sbezverk_ | v1k0d3n: hey, still around.. it is only 1pm here.. | 18:08 |
| kristian2709 | ok np | 18:08 |
| v1k0d3n | do you have any ideas or thoughts on above? | 18:10 |
| v1k0d3n | sbezverk_: | 18:10 |
| sbezverk_ | v1k0d3n: reading | 18:11 |
| v1k0d3n | thing that is strange....i see the mac address for the instances on the 192.168.4.x subnet. | 18:11 |
| *** david-lyle has joined #openstack-kolla | 18:11 | |
| sbezverk_ | v1k0d3n: how about security rules? | 18:12 |
| *** pbourke has quit IRC | 18:13 | |
| *** kristian2709 has quit IRC | 18:13 | |
| *** pbourke has joined #openstack-kolla | 18:13 | |
| v1k0d3n | yeah, that's always the first thing i look at. all good there. | 18:13 |
| sbezverk_ | v1k0d3n: can you configm that your public interface is pluged to ovs bridge | 18:14 |
| sbezverk_ | I meant confirm | 18:14 |
| sbezverk_ | v1k0d3n: usually in this situation I just go step by step with tcpdmp and see where packets gets dropped.. | 18:15 |
| sbezverk_ | v1k0d3n: instance tap interface, linux bridge interface in/out, ovs interface, external interface, somewhere you will see packets stop crossing.. | 18:16 |
| v1k0d3n | ah! got it. | 18:16 |
| v1k0d3n | all good sbezverk_ | 18:17 |
| sbezverk_ | v1k0d3n: what was it? | 18:17 |
| v1k0d3n | sbezverk_: really simple fix. had to add a route on the AIO host to send all 192.168.4.x traffic to em2. | 18:18 |
| v1k0d3n | in other words...really simple bonehead error :) | 18:19 |
| v1k0d3n | probably from lack of sleep or busy week... :) | 18:19 |
| sbezverk_ | v1k0d3n: good catch! | 18:19 |
| portdirect | Thats caught me out before | 18:19 |
| v1k0d3n | sbezverk_: do you know if AIO deployments go haywire on reboots? | 18:19 |
| v1k0d3n | or is there any special considerations to make for an AIO development host? | 18:20 |
| v1k0d3n | like there is with OSA | 18:20 |
| *** Pavo has quit IRC | 18:21 | |
| *** Pavo has joined #openstack-kolla | 18:21 | |
| sbezverk_ | v1k0d3n: Have not seen many issues with aio on reboot. | 18:26 |
| sbezverk_ | in fact I do not recall even a single one | 18:26 |
| v1k0d3n | sbezverk_: so everything just comes up, or do i have to restart the containers? | 18:26 |
| sbezverk_ | v1k0d3n: yep, if your docker autostarts, then everything should just come up | 18:27 |
| v1k0d3n | yeah that's great. | 18:27 |
| v1k0d3n | thanks sbezverk_ | 18:27 |
| sbezverk_ | v1k0d3n: np | 18:28 |
| *** hogepodge has quit IRC | 18:34 | |
| *** sdake has joined #openstack-kolla | 18:41 | |
| sdake | rhallisey we need operators for our operators - yo dawg! | 18:42 |
| rhallisey | :/ | 18:54 |
| *** senk has joined #openstack-kolla | 19:10 | |
| *** sdake has quit IRC | 19:18 | |
| v1k0d3n | prediction for 2017: operators will be the most overused and abused term of the year :) | 19:20 |
| v1k0d3n | btw...can anyone riddle me this...? | 19:22 |
| portdirect | hot off the press: kubernetes 1.5 introduces 'users'. | 19:22 |
| v1k0d3n | i have an AIO without TLS, and i want to keep all configuration options (including running machihne) and just add TLS using ansible. possible? | 19:22 |
| v1k0d3n | portdirect: ha! | 19:22 |
| v1k0d3n | i'm going to have our team build the "consultant" container. it will claim to magically "improve" your CI/CD pipline, leave you with more questions than answers, and it bills you at the rate of $400 per hour automatically. | 19:26 |
| v1k0d3n | and it can scale without anyone's knowledge. | 19:27 |
| v1k0d3n | i.e. bring in more "consulting containers". | 19:27 |
| v1k0d3n | it can even bring in 3rd party resources when it can't get the job done. | 19:27 |
| portdirect | I know a few VC's who may be interested in that.. | 19:28 |
| *** senk has quit IRC | 19:31 | |
| v1k0d3n | rhallisey: can i just change globals to include TLS and redeploy without ansible messing up any of my running workloads or config? | 19:32 |
| v1k0d3n | again, this is for that aio i am building for development. | 19:32 |
| rhallisey | that's tricky | 19:36 |
| rhallisey | idk if you reconfigure will work with enabling tls | 19:37 |
| sbezverk_ | portdirect: I am more concerned with this in 1.5 Rename PetSet to StatefulSet | 19:38 |
| portdirect | sbezverk_: i saw that; you know what logic's behind that? | 19:39 |
| rhallisey | let me think | 19:40 |
| rhallisey | .. | 19:40 |
| *** dave-mccowan has joined #openstack-kolla | 19:42 | |
| sbezverk_ | portdirect: nope | 19:43 |
| rhallisey | v1k0d3n, I think you will be ok | 19:44 |
| rhallisey | nova will get restarted | 19:44 |
| sbezverk_ | sbezverk_: I guess one way to verify is to load 1.5.0 beta 1 and test spec with kind: petset | 19:44 |
| rhallisey | but I don't think it will destroy infrastructure you have | 19:45 |
| v1k0d3n | rhallisey: so fyi...that completely killed my host :( | 20:07 |
| v1k0d3n | can't access anymore. | 20:07 |
| rhallisey | O.o | 20:07 |
| *** dave-mccowan has quit IRC | 20:08 | |
| rhallisey | the vms? or the bm machine? | 20:08 |
| v1k0d3n | well, i guess there's a reliance on haproxy? | 20:08 |
| rhallisey | yes, endpoints get reset | 20:09 |
| v1k0d3n | horizon. i have aio with haproxy turned off. | 20:09 |
| v1k0d3n | just pointing to the internet | 20:09 |
| v1k0d3n | interface, sorry. | 20:09 |
| v1k0d3n | so i guess maybe removing it may correct? anyway...trying it just in case. don't want to have to rebuild or anything. | 20:09 |
| v1k0d3n | totally should've thought about that when i was blowing through the changes. | 20:10 |
| * v1k0d3n feels like a bonehead. rtfm. | 20:12 | |
| rhallisey | hehe | 20:12 |
| v1k0d3n | to be fair...this is the first time i chose not to use haproxy. | 20:13 |
| v1k0d3n | trying to toy around with various options and reconfigure options. i want to rely on it more, and break it often. so this is good anyway. | 20:14 |
| *** Pavo has quit IRC | 20:21 | |
| v1k0d3n | rhallisey: yeah that wacked everything out. i think i'm forced to rebuild :( | 20:22 |
| *** hogepodge has joined #openstack-kolla | 20:22 | |
| rhallisey | ugh | 20:22 |
| rhallisey | darn | 20:22 |
| *** hogepodge has quit IRC | 20:22 | |
| v1k0d3n | nova errors...errors all over the place. | 20:23 |
| *** hogepodge has joined #openstack-kolla | 20:24 | |
| *** Pavo has joined #openstack-kolla | 20:26 | |
| *** dave-mccowan has joined #openstack-kolla | 20:46 | |
| *** kristian2709 has joined #openstack-kolla | 20:46 | |
| kristian2709 | Hi is anyone familiar with gpu passthrough? | 20:51 |
| *** Jeffrey4l has quit IRC | 21:35 | |
| kristian2709 | please may someone help me with pci passthrough? | 21:36 |
| v1k0d3n | kristian2709: sundays are really slow. the group will be more responsive tomorrow. | 21:52 |
| kristian2709 | ok we will see | 21:52 |
| v1k0d3n | sbezverk_: you still around? | 21:53 |
| v1k0d3n | more questions if you're still hanging out. | 21:53 |
| *** dave-mccowan has quit IRC | 21:59 | |
| *** portdirect is now known as portdirect_away | 22:09 | |
| *** Pavo has quit IRC | 22:21 | |
| *** rhallisey has quit IRC | 22:23 | |
| *** Pavo has joined #openstack-kolla | 22:26 | |
| *** sdake_ has joined #openstack-kolla | 22:28 | |
| *** eaguilar has joined #openstack-kolla | 22:30 | |
| *** eaguilar has quit IRC | 22:38 | |
| v1k0d3n | does anyone know how i might be able to redeploy kolla on the same host without reinstalling everything? | 22:38 |
| v1k0d3n | i tried TLS and that failed...tried to go back and nova stopped working. | 22:38 |
| v1k0d3n | i'd like to clean and redeploy, but that's causing issues. | 22:39 |
| sdake_ | v1k0d3n do you mean kolla-ansible? | 22:39 |
| v1k0d3n | there's got to be some cruft left over somewhere. | 22:39 |
| v1k0d3n | yes | 22:39 |
| sdake_ | v1k0d3n did you run kolla-ansible destroy inbetween? | 22:39 |
| sdake_ | note kolla-ansible takes a flag which does a super cleanup | 22:40 |
| v1k0d3n | another thing, i marked magnum and lbaas to be installed, but didn't seem to be enabled. | 22:40 |
| v1k0d3n | sdake_: i have done that yes. | 22:40 |
| sdake_ | also tls does work | 22:40 |
| sdake_ | ;) | 22:40 |
| v1k0d3n | not sure what you mean. i know it does. | 22:40 |
| sdake_ | so if you use destroy your system should be back in a pristine state | 22:40 |
| sdake_ | i mean in our immplementation | 22:41 |
| sdake_ | you said you tried tls and it failed | 22:41 |
| v1k0d3n | just didn't do it the first time of deploy, thihnking i could redeploy... | 22:41 |
| v1k0d3n | and that bombed | 22:41 |
| sdake_ | oh right reconfigure of tls | 22:41 |
| sdake_ | ya - reconfigure not meant for globals.yaml | 22:41 |
| v1k0d3n | i did try tls and it failed. right. | 22:41 |
| sdake_ | bit of a misnomer | 22:41 |
| sdake_ | reconfigure is limited to reconfiguring /etc/kolla/config | 22:41 |
| v1k0d3n | i asked ahead of time. was told should work. | 22:41 |
| v1k0d3n | so there is nothing here currently...destroyed. | 22:42 |
| sdake_ | v1k0d3n the answer is some things work some of the times | 22:42 |
| v1k0d3n | when i redeploy no love. | 22:42 |
| v1k0d3n | lol | 22:42 |
| v1k0d3n | ok | 22:42 |
| sdake_ | v1k0d3n so my answer is "globals.yaml cannot be reconfigured" | 22:42 |
| v1k0d3n | even after a destroy? | 22:42 |
| sdake_ | nope a destroy should do a ttal cleanup | 22:42 |
| sdake_ | so lets fi that | 22:42 |
| sdake_ | did you pass the inventory file when you destroyed? | 22:43 |
| v1k0d3n | AIO | 22:43 |
| v1k0d3n | AIO for development | 22:43 |
| sdake_ | got it | 22:43 |
| sdake_ | ok so can you tell me how redeploy is failling? | 22:43 |
| v1k0d3n | could of things...in my statement. | 22:44 |
| v1k0d3n | LBaaS never seemed to deploy. | 22:44 |
| v1k0d3n | so no magnum. | 22:44 |
| v1k0d3n | that's a fail. | 22:44 |
| sdake_ | upstream keeps doing funny things with lbaas | 22:44 |
| v1k0d3n | then when i redeploy, horizon fails, nova fails. endpoints are not working. | 22:44 |
| sdake_ | so there is no "redeploy" do you mean after a destroy you deploy and everything craters? | 22:45 |
| v1k0d3n | so i am hoping that when i pick 3.0.1 it should be stable. | 22:45 |
| v1k0d3n | i will be clear. | 22:45 |
| v1k0d3n | this is ALWAYS after a destroy. | 22:45 |
| v1k0d3n | yes | 22:45 |
| sdake_ | mind showing me the log of where it craters during deploy - or does deploy work correctly | 22:46 |
| sdake_ | work/finish | 22:46 |
| v1k0d3n | currently destroyed. | 22:46 |
| sdake_ | cool run script | 22:46 |
| sdake_ | this will create a file called typescript in your cwd | 22:46 |
| sdake_ | then run kolla-ansible deploy | 22:46 |
| sdake_ | run with -vvv | 22:46 |
| sdake_ | once that craters or finishes, type exit | 22:47 |
| sdake_ | this will save a file called typescript | 22:47 |
| sdake_ | then paste the contets of typescript somewhere | 22:47 |
| sdake_ | if your on centos, you can yum install fpaste and fpaste typescript | 22:47 |
| sdake_ | if your on some other distro not sure how to past efrom cli but i know its possible ;) | 22:47 |
| v1k0d3n | i'm not sure that's making sense to me.... | 22:48 |
| sdake_ | there is a command called script | 22:48 |
| v1k0d3n | cwd? | 22:48 |
| sdake_ | cwd = current working directory | 22:48 |
| v1k0d3n | oh...yeah gotcha | 22:48 |
| sdake_ | typescript will contain the contents of your shell run | 22:48 |
| sdake_ | this is how i capture and share runs | 22:49 |
| v1k0d3n | and run script for kolla pull && kolla deploy? | 22:49 |
| sdake_ | some other people use tee and stuff | 22:49 |
| sdake_ | you run script once, it createa a subshell | 22:49 |
| v1k0d3n | also, what needs to be done to ensure lbaas and magnum? | 22:49 |
| v1k0d3n | that's sort of what i want to test. | 22:49 |
| sdake_ | lets tackle that next | 22:49 |
| sdake_ | i want to see where your at curently | 22:49 |
| sdake_ | i get conflicting reports on magnum working vs not working | 22:50 |
| v1k0d3n | ok, so i removed all containers and images from the box and rebooted. | 22:50 |
| sdake_ | it also wokrs for me | 22:50 |
| sdake_ | aer you using ceph per chance? | 22:50 |
| v1k0d3n | you're saying do a kolla-ansible pull | 22:50 |
| v1k0d3n | then kolla-ansible deploy -vvv | 22:50 |
| v1k0d3n | correct? | 22:50 |
| v1k0d3n | no ceph | 22:50 |
| sdake_ | did you run the script command first? | 22:50 |
| v1k0d3n | figured didn't need on an aio host | 22:50 |
| v1k0d3n | script? | 22:51 |
| sdake_ | type "script" | 22:51 |
| sdake_ | into your shell | 22:51 |
| sdake_ | then type echo "hello world" | 22:51 |
| sdake_ | then type exit | 22:51 |
| sdake_ | then type cat typescript | 22:51 |
| sdake_ | rather cat tyescript | 22:51 |
| v1k0d3n | i gotcha now. yup got it. | 22:52 |
| sdake_ | ok | 22:52 |
| v1k0d3n | so do a kolla-ansible pull and then deploy, correct? | 22:52 |
| sdake_ | so we are just giong to do the above exercise, except create a typescript for kolla-ansible deploy -vvv | 22:52 |
| v1k0d3n | well i want to get tls so need to check globals, and deploy certs | 22:52 |
| v1k0d3n | one sec | 22:52 |
| v1k0d3n | k pulling | 22:54 |
| v1k0d3n | as this is pulling and running, i am going to grab something to eat sdake_ you gonna be around in 30 mins? prob gonna take about 15-20 mins or so anyway. | 22:55 |
| sdake_ | if my batteries dont run out | 22:56 |
| sdake_ | 24% atm | 22:56 |
| v1k0d3n | power outlet? | 22:58 |
| *** sdake has joined #openstack-kolla | 22:58 | |
| v1k0d3n | well, wife wants to go with to get food apparently. changes my whole plan.... | 22:58 |
| v1k0d3n | so i will probably be faster and can keep troubleshooting since shes on phone. | 22:58 |
| v1k0d3n | ok, running deploy with -vvv now. | 22:59 |
| *** sdake_ has quit IRC | 23:02 | |
| *** eaguilar has joined #openstack-kolla | 23:04 | |
| *** tonanhngo has quit IRC | 23:05 | |
| *** bjolo has quit IRC | 23:07 | |
| *** bjolo has joined #openstack-kolla | 23:08 | |
| v1k0d3n | ok, so sdake same thing... | 23:11 |
| v1k0d3n | ran just fine... | 23:12 |
| v1k0d3n | cannot access horizon | 23:12 |
| v1k0d3n | redirects to TLS. | 23:12 |
| sdake | ok so that typescript file.. is where? | 23:15 |
| sdake | if you type exit | 23:15 |
| sdake | it will save it | 23:16 |
| sdake | the nyou can paste it :) | 23:16 |
| v1k0d3n | in cdir but that contains a lot | 23:16 |
| sdake | ya its a busy file | 23:16 |
| sdake | did it complete ? | 23:16 |
| sdake | if it completed the firs tthing i want to look at is the last 100 or so lines | 23:17 |
| sdake | so just paste that first if you like | 23:17 |
| v1k0d3n | are there any things you're looking for in particular? | 23:17 |
| sdake | what went wrong | 23:17 |
| v1k0d3n | first, the file is really large. the other thing is that passwords are in there. | 23:17 |
| sdake | ok well if yoru passwords are secure then thats not going to work | 23:18 |
| sdake | i had thought you used genpwd | 23:18 |
| sdake | so how about pasting the last 100 lines | 23:18 |
| v1k0d3n | horizon redirecting to SSL and fail | 23:18 |
| sdake | ther eshouldn't be anything there secure | 23:18 |
| sdake | i can't help with that information - i need full logs :) | 23:18 |
| sdake | but mayben ot even full | 23:18 |
| *** tonanhngo has joined #openstack-kolla | 23:18 | |
| sdake | i need to see if any ansible jobs are "failed" | 23:19 |
| sdake | use tail to display the last parts of typescript | 23:19 |
| Pavo | what you doing | 23:19 |
| sdake | pavo walking v1k0d3n through a kolla-ansible tls deploy | 23:20 |
| sdake | pavo your using tls right? | 23:21 |
| Pavo | yeah | 23:21 |
| v1k0d3n | sdake: ... | 23:21 |
| v1k0d3n | so horizon is ignoring TLS | 23:22 |
| sdake | pavo cool maybe you can help v1k0d3n when my power drops out :) | 23:22 |
| sdake | at 8% atm | 23:22 |
| sdake | v1k0d3n i hear that, i need more information | 23:22 |
| v1k0d3n | ok | 23:22 |
| v1k0d3n | hey if you're at 8% don't worry about it. | 23:22 |
| sdake | i reserve it for you :) | 23:23 |
| sdake | show me the line that says pass / fail / etc | 23:23 |
| sdake | at the end of the ansible output | 23:23 |
| v1k0d3n | i will torch this. or try something else. | 23:23 |
| sdake | nothing secure about it | 23:23 |
| v1k0d3n | there is no fail in any of the deploy logs | 23:23 |
| sdake | the last bit of output | 23:23 |
| sdake | it says number of jobs completed etc | 23:23 |
| sdake | rather tasks | 23:23 |
| *** tonanhngo has quit IRC | 23:23 | |
| v1k0d3n | something is cached. the endpoint is still there. when i curl the endpoint.... | 23:24 |
| sdake | you jsut deployed | 23:24 |
| v1k0d3n | {"version": {"status": "stable", "updated": "2016-10-06T00:00:00Z", "media-types": [{"base": "application/json", "type": "application/vnd.openstack.identity-v3+json"}], "id": "v3.7", "links": [{"href": "http://192.168.70.25:5000/v3/", "rel": "self"}]}} | 23:24 |
| sdake | are you running a regisry? | 23:25 |
| v1k0d3n | ...no? | 23:25 |
| sdake | cool | 23:25 |
| sdake | so i would expect the endpoint to respond unles you do a destroy | 23:25 |
| *** tonanhngo has joined #openstack-kolla | 23:26 | |
| v1k0d3n | you said you need full log. anything from the log in particular? | 23:26 |
| v1k0d3n | the log looks like a clean deployment. | 23:26 |
| sdake | v1k0d3n the last 3 lines after kolla-ansible deploy | 23:26 |
| sdake | it says TASKS: xyz | 23:26 |
| *** eaguilar has quit IRC | 23:27 | |
| sdake | there are two lines i'd take after this peice of ifnoratmion - either firewalld is a problem or your globals.yaml is misconfigured | 23:28 |
| sdake | but we need this pecie of information first to see if we move on to the next peice of debug | 23:28 |
| sdake | if there are failures in that last two lines thats a problem - and that needs fixing first | 23:28 |
| *** tonanhngo has quit IRC | 23:28 | |
| v1k0d3n | confused. | 23:29 |
| sdake | right, thtas why i debug 1 step at a time :) | 23:29 |
| v1k0d3n | it worked with exact same globals -- the TLS portion. | 23:29 |
| sdake | your next step is to produce the end output of the deploy operation :) | 23:30 |
| Pavo | kolla-build, kolla-ansible bootsctrap server -i /usr/share/kolla/ansible/inventory/multinode, kolla-ansible certificates, kolla-ansible deploy -i /usr/share/kolla/ansible/inventory/multinode | 23:30 |
| Pavo | thats all I do | 23:30 |
| sdake | pavo v1k0d3n is AIO atm | 23:30 |
| Pavo | ah haven't done a deployment of aio yet | 23:31 |
| Pavo | I would assume its easier though | 23:31 |
| *** Jeffrey4l has joined #openstack-kolla | 23:31 | |
| sdake | v1k0d3n are you in a vm? | 23:31 |
| Pavo | kolla-ansible certificates has to be done before deploy though | 23:31 |
| Pavo | for tls | 23:31 |
| v1k0d3n | no bare metal host | 23:31 |
| sdake | v1k0d3n you said it worked with same globals with tls, but then you changed (??X) and it stopped working | 23:32 |
| v1k0d3n | Pavo: i did certificates | 23:32 |
| sdake | what is the ??X above? | 23:32 |
| v1k0d3n | sdake: ..... again.... | 23:32 |
| v1k0d3n | i deployed without TLS (commented out...default). | 23:32 |
| v1k0d3n | worked fine. | 23:32 |
| Pavo | v1k0d3n do you have a certifcates folder in /etc/kolla? | 23:32 |
| v1k0d3n | destroyed. | 23:32 |
| v1k0d3n | cleared all images and containers | 23:32 |
| v1k0d3n | gen certs | 23:33 |
| v1k0d3n | redeploy | 23:33 |
| v1k0d3n | really, really simple. | 23:33 |
| v1k0d3n | that's it | 23:33 |
| sdake | you misse dthe step in there where you changed globals.yaml | 23:33 |
| sdake | what did you change it to? | 23:33 |
| sdake | destroy has nothing to do with it :) | 23:33 |
| sdake | especially if you rebooted | 23:34 |
| v1k0d3n | sorry. of course i changed globals. | 23:34 |
| sdake | right - can you apste that plz | 23:34 |
| v1k0d3n | kolla_enable_tls_external: "yes" | 23:34 |
| v1k0d3n | kolla_external_fqdn_cert: "{{ node_config_directory }}/certificates/haproxy.pem" | 23:34 |
| v1k0d3n | that's it. | 23:34 |
| v1k0d3n | i wanted to change dns, but i didn't...i'm trying to step through. | 23:35 |
| sdake | we really need to modify our docs to include tls setup | 23:35 |
| v1k0d3n | sdake: isn't that what this is? | 23:35 |
| v1k0d3n | http://docs.openstack.org/developer/kolla/advanced-configuration.html | 23:35 |
| kollabot1 | Advanced Configuration — kolla 4.0.0.0b2.dev45 documentation | 23:35 |
| Pavo | well you also have to have kolla_external_fqdn: as a domain name | 23:36 |
| v1k0d3n | http://docs.openstack.org/developer/kolla/advanced-configuration.html#self-signed-certificates | 23:36 |
| kollabot1 | Advanced Configuration — kolla 4.0.0.0b2.dev45 documentation | 23:36 |
| sdake | cool | 23:36 |
| Pavo | if no domain name then no certs gets made | 23:36 |
| sdake | didn't know that hit the docs | 23:36 |
| v1k0d3n | Pavo: to be fair....i actually changed the fqdn as well during one attempt | 23:36 |
| v1k0d3n | well, certs get generated, but i think against the IP address and not a FQDN | 23:36 |
| Pavo | what kolla deployment version are you deploying? | 23:37 |
| sdake | # This should be a VIP, an unused IP on your network that will float between | 23:37 |
| sdake | # the hosts running keepalived for high-availability. It defaults to the | 23:37 |
| v1k0d3n | not quite sure on this. | 23:37 |
| sdake | # kolla_internal_vip_address, allowing internal and external communication to | 23:37 |
| sdake | # share the same address. Specify a kolla_external_vip_address to separate | 23:37 |
| sdake | # internal and external requests between two VIPs. | 23:37 |
| sdake | #kolla_external_vip_address: "{{ kolla_internal_vip_address }}" | 23:37 |
| Pavo | I use stable/mitaka | 23:37 |
| v1k0d3n | ok... | 23:37 |
| v1k0d3n | so right.... | 23:37 |
| v1k0d3n | and i am not using HA | 23:37 |
| v1k0d3n | sorry...haproxy | 23:37 |
| sdake | tls is enabled via haproxy... | 23:37 |
| sdake | who on earth told you not to use haproxy? | 23:37 |
| sdake | i really wish people would stop giving out that advice | 23:38 |
| sdake | its bad! | 23:38 |
| v1k0d3n | right...but up above...even when i try to deploy W/O TLS it doesn't work now | 23:38 |
| v1k0d3n | and now we're back to the start... :) | 23:38 |
| Pavo | also v1k0d3n The kolla_internal_vip_address and kolla_external_vip_address must be different to enable TLS on the external network. | 23:38 |
| sdake | the docs are incomplete and should record this information | 23:38 |
| Pavo | that is from the docs sdake | 23:39 |
| Pavo | in the Note | 23:39 |
| sdake | pavo cool :) | 23:39 |
| v1k0d3n | ok, so i destroyed again. | 23:39 |
| sdake | how about pasting your globals.yaml this time :) | 23:39 |
| v1k0d3n | i need external and internal FQDN's | 23:39 |
| v1k0d3n | sdake: be glad to...but at this point, i want to know what it should actually be. | 23:40 |
| sdake | you said you have haproxy disabled | 23:40 |
| Pavo | but I do think the docs need to be completely rewritten in more of a guide stand point not just about the different areas of different settings, its need to be able to walk someone through a complete setup and deploy | 23:40 |
| sdake | that shouldn't be one of the things :) | 23:40 |
| sdake | if you had pasted the full logs this would have been the second thing i looked at | 23:41 |
| Pavo | no need for external and internal FQDNs, but you do need internal vip and external vip to be different | 23:41 |
| v1k0d3n | Pavo: OMG YES TO THIS!!! | 23:41 |
| v1k0d3n | user is guessing..."what is advanced?" "what is default"? | 23:41 |
| Pavo | exactly | 23:41 |
| v1k0d3n | how should user know from the start what is advanced? | 23:41 |
| Pavo | they are better than they were but still need alot more work | 23:42 |
| v1k0d3n | they will see..."enabled TLS" and think "oh, yeah...i wanted that". | 23:42 |
| v1k0d3n | then deploy...no precheck...no error...boom. | 23:42 |
| sdake | 0% | 23:42 |
| sdake | not sure when my laptop drops off | 23:42 |
| v1k0d3n | well, for AIO, it should just be....done. | 23:42 |
| sdake | tls has a 30-300% inpact on performance | 23:42 |
| v1k0d3n | no guessing. just defaults. TLS is security. include it. | 23:42 |
| sdake | so we don't enable it by default | 23:42 |
| v1k0d3n | AIO sdake | 23:43 |
| Pavo | this is my multinode globals | 23:43 |
| Pavo | http://pastebin.com/eZbDjtTE | 23:43 |
| kollabot1 | ################### # Kolla options ################### # Valid options are [ - Pastebin.com | 23:43 |
| v1k0d3n | nobody is looking for optimal performance for AIO | 23:43 |
| *** sdake has quit IRC | 23:43 | |
| Pavo | can someone try and get to https://ddi.hopto.org? | 23:44 |
| v1k0d3n | Pavo: no love to that link.... | 23:44 |
| Pavo | damnit | 23:44 |
| Pavo | I hate my ISP | 23:44 |
| v1k0d3n | who? | 23:45 |
| Pavo | can you try 1 more time | 23:45 |
| v1k0d3n | same. no go. | 23:45 |
| Pavo | wtf tcpdump shows you trying to get to it | 23:45 |
| Pavo | so port forwarding is working | 23:45 |
| Pavo | wonder if my FQDN is messed up | 23:46 |
| *** sdake has joined #openstack-kolla | 23:47 | |
| Pavo | I can get to it no problem from inside my network | 23:47 |
| sdake | v1k0d3n the problem with enabling TLS for AIO is it requires TWO vips not one | 23:47 |
| sdake | getting epople to understand ONE VIP is a challenge ;) | 23:47 |
| sdake | over 200 times i've debugged peoples envs that had TLS == host ip | 23:48 |
| sdake | i believe there is a precheck for that now ;-) | 23:48 |
| sdake | (note TLS==host ip doesn't work) | 23:48 |
| Pavo | true statement | 23:48 |
| sdake | learning kolla is incremental - not all at once :) | 23:49 |
| sdake | otherwise its overwhelming | 23:49 |
| sdake | not to say the docs don't suck | 23:49 |
| sdake | they indeed are not optimal | 23:49 |
| Pavo | I think the biggest issue with the docs right now is that both deployment methods are on the same page and is confusing | 23:50 |
| Pavo | they need to be seperated | 23:50 |
| sdake | you mean dev workflow and operator workflow pavo? | 23:50 |
| Pavo | 1 page for production deploymnet and a seperate page for dev deployment | 23:50 |
| sdake | writing docs is harder then writing implementations ;) | 23:50 |
| v1k0d3n | Pavo: http://pastebin.com/LdrTdd8J | 23:50 |
| kollabot1 | globals.yml - Pastebin.com | 23:50 |
| Pavo | and each one has a complete walk through on how to setup and deploy, then add advanced config examples | 23:51 |
| v1k0d3n | that's a WIP. | 23:51 |
| v1k0d3n | i have two interfaces.... | 23:51 |
| sdake | yikes enable_haproxy: NO is wrong | 23:51 |
| v1k0d3n | one, no ip address. | 23:51 |
| v1k0d3n | working on it sdake | 23:51 |
| v1k0d3n | there are examples out there without it. | 23:51 |
| Pavo | v1k0d3n #kolla_external_vip_address: "{{ kolla_internal_vip_address }}" needs to be different | 23:51 |
| v1k0d3n | ah...btw... sdake including cisco | 23:51 |
| v1k0d3n | cisco dev learning | 23:52 |
| v1k0d3n | which is where this came from orig | 23:52 |
| sdake | line 41 needs needs a specific VIP | 23:52 |
| sdake | v1k0d3n cisco is a big place, not everyone talks to eveyrone else ;) | 23:52 |
| Pavo | #kolla_enable_tls_external: "yes" needs to be uncommented | 23:52 |
| sdake | v1k0d3n anyway when i work here i work upstream | 23:52 |
| Pavo | #kolla_external_fqdn_cert: "{{ node_config_directory }}/certificates/haproxy.pem" needs to be uncommented | 23:53 |
| v1k0d3n | so one sec... | 23:53 |
| v1k0d3n | one at a time? | 23:53 |
| v1k0d3n | i will change haproxy now | 23:53 |
| Pavo | I would also change #enable_central_logging: "no" to enable_central_logging: "yes" | 23:53 |
| Pavo | so you can TS easier | 23:53 |
| v1k0d3n | kolla external vip | 23:53 |
| v1k0d3n | i have two nics up | 23:53 |
| v1k0d3n | one without an ip address | 23:53 |
| v1k0d3n | and the other is what i access horizon on. | 23:53 |
| *** kristian2709 has quit IRC | 23:54 | |
| sdake | right the external vip needs to be an unused address on your network | 23:54 |
| v1k0d3n | 192.168.70.25 is horizon | 23:54 |
| v1k0d3n | so that is external? | 23:54 |
| sdake | the internal vip needs to be an unsued address on your network | 23:54 |
| sdake | you need both for tls to work correctly | 23:54 |
| v1k0d3n | so both need to be unused addresses? | 23:54 |
| sdake | right - now you see why its not enabled by default ;) | 23:54 |
| Pavo | internal vip is network that only admins would use to access horizon, external vip would be an address that everyone else would use to access horizon | 23:54 |
| v1k0d3n | so i want to be able to access horizon on 70.25. | 23:55 |
| Pavo | and if you did a destroy you need to reboot because those vip address do not get unused until reboot | 23:55 |
| v1k0d3n | i have name resolution on that. | 23:55 |
| v1k0d3n | so i need to re-IP my host to .24 or something? | 23:55 |
| sdake | kolla_internal_vip_address: "192.168.70.25" | 23:56 |
| sdake | what is 70.25? | 23:56 |
| sdake | your host? | 23:56 |
| Pavo | the IP address on the nic that its connected doesn;t matter, you will use your internal and external vips to access horizon | 23:56 |
| v1k0d3n | yes, host is 192.168.70.25. | 23:56 |
| sdake | ok, that needs to be an unused address such as 192.168.70.200 | 23:56 |
| Pavo | ^ | 23:56 |
| sdake | (if 200 is in your network and unused) | 23:56 |
| v1k0d3n | ok, let me say this all in one string...maybe that is easier. | 23:56 |
| sdake | lets get aio with haproxy without tls working first ;) | 23:56 |
| v1k0d3n | i have DNS already set up to reach horizon...so the VIP needs to be 70.25....so i guess i need to re-IP address my host, is that correct? | 23:57 |
| Pavo | you can do kolla-ansible reconfigure to get tls working after you actually get everything else up | 23:57 |
| Pavo | v1k0d3nyes correct | 23:57 |
| sdake | right, if you re-ip your host, then .25 will be free for your internal vip - but your internal vip will not be TLS-ified | 23:57 |
| v1k0d3n | ok. let me readdress my host then. | 23:57 |
| Pavo | but make sure to reboot after you do | 23:58 |
| Pavo | so ARP can clear its cache | 23:58 |
| v1k0d3n | sdake: different issue. i am taking one step at a time. | 23:58 |
| v1k0d3n | addressing first. | 23:58 |
| v1k0d3n | Pavo: rebooting now | 23:59 |
| sdake | pavo he had no haproxy prior | 23:59 |
| sdake | a reboot wasnt' necessary | 23:59 |
| v1k0d3n | have ideas on how i can rewrite the docs. | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!