| *** yamahata has joined #tripleo | 00:10 | |
| *** openstackstatus has quit IRC | 00:23 | |
| *** openstackstatus has joined #tripleo | 00:24 | |
| *** ChanServ sets mode: +v openstackstatus | 00:24 | |
| *** dsneddon_ has quit IRC | 00:27 | |
| *** bnemec has quit IRC | 00:28 | |
| *** dsneddon_ has joined #tripleo | 00:28 | |
| *** prad has joined #tripleo | 00:31 | |
| *** dsneddon_ has quit IRC | 00:48 | |
| *** dsneddon_ has joined #tripleo | 00:49 | |
| *** panda has quit IRC | 00:55 | |
| *** panda has joined #tripleo | 00:56 | |
| openstackgerrit | Steven Dake proposed a change to stackforge/kolla: Make keystone cycle until mariadb is avialable https://review.openstack.org/127093 | 01:03 |
|---|---|---|
| openstackgerrit | Richard Su proposed a change to openstack/tripleo-image-elements: SELinux: Fix keepalived killall denials https://review.openstack.org/127095 | 01:05 |
| sdake | larsks around? | 01:16 |
| larsks | Here! Just left you a comment on ^^^^... | 01:16 |
| sdake | cool | 01:18 |
| sdake | how do you access the creds | 01:18 |
| larsks | Which creds? Where? | 01:18 |
| sdake | now that I have keystone up and running i get invalid credentials | 01:18 |
| sdake | http://paste.fedoraproject.org/140496/28175391 | 01:19 |
| sdake | I do that | 01:19 |
| sdake | rather I source that | 01:19 |
| sdake | and keystone endpoint-list doesn't work | 01:19 |
| sdake | am I doing something wrong, or did creds change or something? | 01:19 |
| larsks | Just a sec... | 01:19 |
| larsks | I haven't tried authenticating as admin before (I've been using e.g., glance credentials from glance-api.conf, or using SERVICE_TOKEN). Let me try auth'ing as admin now and see... | 01:21 |
| sdake | I just want to ge tt o the poinut I can auth with nova | 01:21 |
| sdake | and glance | 01:21 |
| sdake | so if you could tell me how, that would rock :) | 01:21 |
| sdake | I had it working prior | 01:21 |
| larsks | I've been authenticating to glance using the credentials from glance-api.conf (admin_user, admin_password, etc). | 01:23 |
| larsks | That works fine. | 01:23 |
| sdake | i'll have to look at the creds stored there | 01:23 |
| larsks | Indeed I am also unable to authenticate as admin using the "kolla" password. Let me take a quick look at that. | 01:23 |
| sdake | what is odd is it was workign before | 01:25 |
| sdake | code churn ftl | 01:25 |
| larsks | Maybe. I'm making sure I'm running kolla master locally, then I will spend a little time debugging. | 01:26 |
| sdake | I git checkout on my master node | 01:26 |
| sdake | and use kubecfg from there | 01:27 |
| *** takadayuiko has left #tripleo | 01:27 | |
| *** lazy_prince has quit IRC | 01:27 | |
| larsks | Yup. | 01:27 |
| openstackgerrit | Richard Su proposed a change to openstack/tripleo-image-elements: SELinux: swift denied access to /var/lib/swift/.local https://review.openstack.org/127104 | 01:27 |
| sdake | I expect people will actually try our first milestone | 01:28 |
| sdake | so it would be good to fix all these annoying ordering problems | 01:28 |
| sdake | so whta we have is pretty solid | 01:28 |
| sdake | scollier said we are at the tip of the spear | 01:29 |
| sdake | ;-) | 01:29 |
| *** rhallisey has quit IRC | 01:30 | |
| larsks | Yeah, I saw that. My dev code is mostly protected from ordering problems at this point using api checks (or mysql checks, like in that comment). | 01:32 |
| larsks | I can merge my stuff, we we can make everything to the exit/reschedule dance. | 01:32 |
| larsks | s/we we/or we/ | 01:33 |
| larsks | Stooopid fingers. | 01:33 |
| *** jerryz has joined #tripleo | 01:34 | |
| sdake | yes please get reviews out so folks can review :) | 01:34 |
| sdake | api checks = way to roll | 01:34 |
| sdake | I was thinkjing about doing that tothe code base tonight, good thing you told me you had it done already :) | 01:34 |
| jerryz | lifeless: ping? | 01:35 |
| sdake | i am going to eat in about 15 mins or so, so if you have any ideas on the creds, type em in | 01:35 |
| sdake | if i'm afk, I'll take a look when i get back | 01:35 |
| sdake | I want to try out nova and glance tonight see where they s tand | 01:36 |
| larsks | Bandwidth to the beaker environment seems to be awful right now (docker pull takes foreeeeeeeever). | 01:36 |
| *** sdake_ has quit IRC | 01:41 | |
| *** cwolfe__ has joined #tripleo | 01:42 | |
| *** Andy_Yan has joined #tripleo | 01:43 | |
| *** killer_prince has joined #tripleo | 01:49 | |
| *** killer_prince is now known as lazy_prince | 01:49 | |
| larsks | sdake: So, I deployed a fresh mariadb and a fresh keystone, and I was able to authenticate using admin/kolla without a problem: http://chunk.io/f/81903a61193d4027ae108ec907958bb2 | 01:51 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: update existing keystone objects https://review.openstack.org/127108 | 01:53 |
| *** lazy_prince has quit IRC | 02:05 | |
| *** lazy_prince has joined #tripleo | 02:05 | |
| *** nosnos has joined #tripleo | 02:06 | |
| *** weshay has quit IRC | 02:25 | |
| openstackgerrit | James Polley proposed a change to openstack/tripleo-image-elements: Make deploy-ci-overcloud script configurable https://review.openstack.org/116534 | 02:26 |
| *** julim has quit IRC | 02:27 | |
| sdake | hmm, so problem on my end then? | 02:28 |
| openstackgerrit | James Polley proposed a change to openstack/tripleo-image-elements: Make deploy-ci-overcloud script configurable https://review.openstack.org/116534 | 02:29 |
| tchaypo | StevenK: could you take a look at https://review.openstack.org/#/c/116534/ ? | 02:30 |
| tchaypo | specifically lxsli's comment about indentation at the bottom of https://review.openstack.org/#/c/116534/4/elements/tripleo-cd/deploy-ci-overcloud | 02:30 |
| larsks | If you previous ran keystone before the change to 'kolla' as the default password, and you still have the same mariadb container running, it would still be using the old password. The change I just pushed will force a password update when we start the container and avoid that problem. | 02:31 |
| tchaypo | if you know what he means maybe we can avoid needing to wait till tomorrow | 02:31 |
| larsks | sdake: ^^^ | 02:31 |
| larsks | You could use SERVICE_ENDPOINT/SERVICE_TOKEN to reset the admin password. | 02:31 |
| *** dsneddon_ has quit IRC | 02:33 | |
| sdake | larsks I killed my entire setup earlier same problem | 02:34 |
| sdake | say, do we need the VOLUME in mariadb container | 02:34 |
| larsks | sdake: It is not operationally required, no, but it's a good idea. | 02:34 |
| larsks | sdake: Re: auth problem, not sure. Works for me with images built from kolla master. | 02:35 |
| *** dsneddon_ has joined #tripleo | 02:35 | |
| larsks | Are your images fresh? I haven't been building new release images (so I'm using fresh builds, but they're in my larsks/... namespace). | 02:36 |
| sdake | I built all new images for everything | 02:37 |
| sdake | on kolla nameespace | 02:37 |
| larsks | If you do reset the password, are you then able to authenticate? | 02:42 |
| larsks | That would make it clear if this is a password problem or something else... | 02:42 |
| *** jerryz has quit IRC | 02:42 | |
| *** openstackgerrit has quit IRC | 02:45 | |
| *** rainya has joined #tripleo | 02:55 | |
| *** untriaged-bot has joined #tripleo | 03:00 | |
| untriaged-bot | No untriaged bugs so far! \o/ | 03:00 |
| *** untriaged-bot has quit IRC | 03:00 | |
| *** pcrews has quit IRC | 03:03 | |
| *** ramishra has joined #tripleo | 03:03 | |
| sdake | how do I do that (rese tpwd) larsks | 03:10 |
| *** rushiagr_away is now known as rushiagr | 03:12 | |
| *** lazy_prince is now known as killer_prince | 03:13 | |
| larsks | sdake: set SERVICE_TOKEN and SERVICE_ENDPOINT correctly, and then: keystone user-password-update --pass kolla admin | 03:13 |
| larsks | sdake: sleeping time here. talk to you tomorrow! | 03:14 |
| sdake | I dont see service_token | 03:14 |
| sdake | ok enjoy sleepytime :) | 03:14 |
| sdake | see you monday | 03:14 |
| larsks | Those are environment variables you need to set, not settings in a config file. SERVICE_TOKEN needs the value of admin_token from keystone.conf, and SERVICE_ENDPOINT needs to point at port 35357. | 03:17 |
| larsks | g'night! | 03:17 |
| sdake | i am just gonna reboot and start fresh | 03:18 |
| sdake | enjoy | 03:18 |
| *** penick has joined #tripleo | 03:26 | |
| *** chuckC_ has joined #tripleo | 03:27 | |
| *** penick_ has joined #tripleo | 03:27 | |
| *** killer_prince has quit IRC | 03:28 | |
| *** lazy_prince has joined #tripleo | 03:29 | |
| *** penick has quit IRC | 03:30 | |
| *** penick_ is now known as penick | 03:30 | |
| *** loki184 has joined #tripleo | 03:33 | |
| *** ChuckC__ has joined #tripleo | 03:34 | |
| *** nosnos has quit IRC | 03:35 | |
| *** ramishra has quit IRC | 03:37 | |
| *** ramishra has joined #tripleo | 03:37 | |
| *** ChuckC__ has quit IRC | 03:37 | |
| StevenK | Bleh. Ironic gets a node locked message, and doesn't try again so nova retuns no host and then errors :-/ | 03:44 |
| tchaypo | nova should retry 3 times | 03:46 |
| tchaypo | or whatever the scheduler retry is set to, shouldn't it? | 03:46 |
| StevenK | Huh | 03:48 |
| StevenK | There's a traceback in nova-compute | 03:48 |
| *** ramishra has quit IRC | 03:48 | |
| StevenK | Oct 9 02:22:41 ubuntu nova-compute: 2014-10-09 02:22:41.884 4384 TRACE nova.openstack.common.periodic_task Unauthorized: The request you have made requires authentication. (HTTP 401) | 03:48 |
| greghaynes | ouch | 03:48 |
| tchaypo | seems familiar. | 03:49 |
| StevenK | Which then leads to: | 03:49 |
| StevenK | Oct 9 02:33:51 ubuntu nova-compute: 2014-10-09 02:33:51.341 4384 TRACE nova.openstack.common.loopingcall InstanceDeployFailure: Failed to provision instance 65871cb3-3e02-484f-a6b9-552950a74cc9: Failed to deploy. Error: HTTPNotFound (HTTP 404) | 03:49 |
| tchaypo | assuming this was driven by heat | 03:50 |
| StevenK | Undercloud bring up | 03:50 |
| tchaypo | how long before the first CREATE_IN_PROGRESS and the first CREATE_FAILED ? | 03:50 |
| StevenK | 3 to 4 minutes, I guess? | 03:50 |
| tchaypo | not keystone auth timeout then | 03:50 |
| StevenK | I'm trying to get hetrogenous flavor support working | 03:51 |
| StevenK | | b14bae25-98ce-4fdf-a4df-f7544b6adf3d | baremetal_3072_10_30_1 | 3072 | 10 | 30 | | 1 | 1.0 | True | | 03:51 |
| *** penick has quit IRC | 03:53 | |
| *** dsneddon_ has quit IRC | 03:57 | |
| *** tzumainn has quit IRC | 03:58 | |
| *** dsneddon_ has joined #tripleo | 03:58 | |
| *** spzala has quit IRC | 04:00 | |
| *** rushiagr is now known as rushiagr_away | 04:08 | |
| *** ramishra has joined #tripleo | 04:13 | |
| *** ramishra has quit IRC | 04:24 | |
| *** pradeep has joined #tripleo | 04:32 | |
| *** nosnos has joined #tripleo | 04:35 | |
| *** Andy_Yan has quit IRC | 04:36 | |
| *** rushiagr_away is now known as rushiagr | 04:37 | |
| *** dsneddon_ has quit IRC | 04:39 | |
| *** dsneddon_ has joined #tripleo | 04:40 | |
| *** rainya has quit IRC | 04:40 | |
| *** loki184 has quit IRC | 04:42 | |
| *** loki184 has joined #tripleo | 04:47 | |
| *** lazy_prince has quit IRC | 04:50 | |
| *** loki184 has quit IRC | 04:53 | |
| *** Andy_Yan has joined #tripleo | 04:56 | |
| *** lazy_prince has joined #tripleo | 04:57 | |
| *** sdake has quit IRC | 05:12 | |
| greghaynes | huh, im now seeing that dib fail where it tries to unmount a loop dev that doesnt exist | 05:15 |
| *** rainya has joined #tripleo | 05:16 | |
| StevenK | greghaynes: There's a patch up for it | 05:16 |
| greghaynes | yea, I wonder what changed to cause that - I assumed that the person who was running into it was doing something funky | 05:17 |
| *** rainya has quit IRC | 05:18 | |
| * greghaynes +A's that patch | 05:18 | |
| tchaypo | StevenK: at least you have a simple failure. I've got a case where things retry for 4 hours until token timeout but I can't figure out what's causing the failure that gets retried | 05:19 |
| *** dsneddon_ has quit IRC | 05:20 | |
| *** jml has quit IRC | 05:21 | |
| *** dsneddon_ has joined #tripleo | 05:22 | |
| *** jcoufal has joined #tripleo | 05:22 | |
| *** jml has joined #tripleo | 05:26 | |
| tchaypo | greghaynes: you were right btw | 05:29 |
| tchaypo | on my undercloud node (which is a few weeks old at this point, perhaps things have changed) the logs don't get rotated | 05:29 |
| StevenK | Huh. Using current -incubator master I get no nova instance, the OS::Server::Nova marked as CREATE_FAILED and no errors in the nova log | 05:31 |
| tchaypo | wheee | 05:31 |
| StevenK | I am going to become a potato farmer, I swear. | 05:31 |
| pleia2 | potatoes++ | 05:32 |
| greghaynes | tchaypo: fun fun | 05:38 |
| greghaynes | tchaypo: pretty sure nothing has changed in that front | 05:38 |
| * tchaypo cries | 05:39 | |
| tchaypo | i go away for a week and come back | 05:39 |
| tchaypo | and all my problems failed to be magically fixed | 05:39 |
| tchaypo | can we hire those cobbler's elves? | 05:39 |
| greghaynes | well, I was playing with trying to ship all our logs to a single box | 05:40 |
| greghaynes | (speaking of crying) | 05:40 |
| greghaynes | I have given up trying to do this the systemd way for now | 05:40 |
| tchaypo | no more poettering around? | 05:41 |
| greghaynes | for now | 05:41 |
| greghaynes | maybe one of the redhatters can do it, though | 05:42 |
| StevenK | tchaypo: Booo | 05:46 |
| StevenK | greghaynes: I have *no* idea where that uname is coming from | 05:47 |
| StevenK | greghaynes: (Re: os-collect-config) | 05:47 |
| tchaypo | no loki | 05:49 |
| *** yamahata has quit IRC | 05:51 | |
| *** yamahata has joined #tripleo | 05:51 | |
| tchaypo | which means I can't chat about 106727 | 05:52 |
| greghaynes | StevenK: awesome! | 05:52 |
| *** loki184 has joined #tripleo | 05:59 | |
| StevenK | WAT | 06:01 |
| StevenK | Oh | 06:02 |
| StevenK | There's another call in there | 06:02 |
| StevenK | The call we append is the second one | 06:03 |
| *** ramishra has joined #tripleo | 06:04 | |
| StevenK | greghaynes: I hate requests | 06:05 |
| tchaypo | I'm starting to hate it too | 06:06 |
| tchaypo | because of the urllib3 thing | 06:06 |
| StevenK | greghaynes: http://paste.openstack.org/show/119810/ | 06:06 |
| tchaypo | nova-compute logs show it spinning waiting for ironic to say nodes are ready | 06:08 |
| tchaypo | ironic-api logs just show it busy being polled by nova.. | 06:08 |
| *** ramishra has quit IRC | 06:09 | |
| *** loki184 has quit IRC | 06:09 | |
| *** loki184 has joined #tripleo | 06:09 | |
| tchaypo | ah, here we go, i'm seeing it try to schedule noes.. | 06:12 |
| *** mrunge has joined #tripleo | 06:13 | |
| greghaynes | StevenK: wah | 06:14 |
| greghaynes | oh | 06:16 |
| greghaynes | wow | 06:16 |
| greghaynes | so its a requests bug | 06:16 |
| greghaynes | again | 06:16 |
| *** dshulyak_ has joined #tripleo | 06:16 | |
| greghaynes | oh | 06:17 |
| StevenK | greghaynes: Oh? | 06:20 |
| greghaynes | it seems like basically our moc'd popen is getting used also by platform module | 06:21 |
| greghaynes | which is amazing | 06:21 |
| StevenK | It's not suprising, we mock subprocess.Popen | 06:23 |
| greghaynes | I am curious how this ever passed though | 06:23 |
| greghaynes | yea, its just a hilarious fail | 06:23 |
| tchaypo | like this overcloud | 06:24 |
| tchaypo | we have the usual 4 nodes up and now I can't figure out what the others are doing | 06:24 |
| greghaynes | StevenK: so seems like we should just be filtering the calls result to assert on the os-c-c call, and maybe also ensure theres only one os-c-c call | 06:25 |
| * greghaynes can hack this up | 06:25 | |
| greghaynes | And maybe note down the anti pattern of assuming a large library doesnt make any popen calls ;) | 06:26 |
| StevenK | greghaynes: Haha, yes | 06:29 |
| StevenK | greghaynes: Maybe we just want to confirm expected_args in calls[x]['args'] | 06:30 |
| greghaynes | yep, filter out the os-c-c call in there, then probably make sure theres exactly one in there | 06:30 |
| *** pensu has joined #tripleo | 06:31 | |
| StevenK | greghaynes: I look forward to your comment involving 'anti-pattern' | 06:37 |
| *** mrunge has quit IRC | 06:39 | |
| *** pcrews has joined #tripleo | 06:39 | |
| tchaypo | ah, here we go | 06:40 |
| tchaypo | ironic shows lots of nodes in wait call-back | 06:40 |
| *** rdopieralski has joined #tripleo | 06:45 | |
| *** mrunge has joined #tripleo | 06:47 | |
| *** pcrews has quit IRC | 06:47 | |
| GheRivero | morning all | 06:56 |
| *** k4n0 has joined #tripleo | 06:56 | |
| *** jprovazn has joined #tripleo | 07:00 | |
| *** k4n0 has quit IRC | 07:02 | |
| greghaynes | StevenK: https://review.openstack.org/#/c/127146/ | 07:03 |
| *** ci-overcloud has joined #tripleo | 07:06 | |
| ci-overcloud | ************** ci-overcloud complete status=1 ************ | 07:06 |
| *** ci-overcloud has quit IRC | 07:06 | |
| SpamapS | it lives! | 07:07 |
| tchaypo | interesting | 07:07 |
| tchaypo | I wasn't expecting that to fail for hours yet | 07:07 |
| tchaypo | lo ghe | 07:07 |
| SpamapS | clearly you haven't used tripleo-cd very much. :) | 07:07 |
| tchaypo | SpamapS: just because you're PTL doesn't mean you can't sleep | 07:07 |
| tchaypo | oh I see what happened here | 07:08 |
| tchaypo | I set the timeout to low | 07:08 |
| tchaypo | the stacik still only has 4 nodes in create_complete and the rest are still spinning | 07:08 |
| tchaypo | but the script got bored and left | 07:08 |
| StevenK | greghaynes: Failure on py33 :-( | 07:09 |
| greghaynes | baw | 07:09 |
| GheRivero | I love to see the ci-overcloud alive! | 07:10 |
| greghaynes | StevenK: you can test it locallt/ | 07:10 |
| StevenK | greghaynes: http://paste.openstack.org/show/119816 should fix that failure, however | 07:10 |
| greghaynes | StevenK: I wasnt able to get py33 test working locally | 07:10 |
| tchaypo | it's not alive, that :( | 07:10 |
| greghaynes | oh, unrelated fail? | 07:10 |
| StevenK | Yeah | 07:10 |
| GheRivero | tchaypo: too bad. Anything i can do to help? | 07:11 |
| StevenK | greghaynes: TypeError: object of type 'filter' has no len() | 07:11 |
| StevenK | But that's a Python 3 trap | 07:12 |
| tchaypo | I don't know. The thing I'm currently trying to debug is why these nodes are in wait call-back | 07:12 |
| greghaynes | wat | 07:12 |
| StevenK | greghaynes: filter() in Python 2 returns a list. filter() in Python 3 returns an iterator | 07:13 |
| greghaynes | ogod | 07:13 |
| greghaynes | ITS A TRAP | 07:13 |
| StevenK | greghaynes: And 'mock' in the commit message :-P | 07:13 |
| *** k4n0 has joined #tripleo | 07:14 | |
| greghaynes | StevenK: what did you have to do to make tox py33 test work locally | 07:14 |
| greghaynes | well | 07:14 |
| StevenK | - proc_calls = filter(lambda call: call['args'] == expected_cmd, calls) | 07:14 |
| StevenK | + proc_calls = [call for call in calls if call['args'] == expected_cmd] | 07:14 |
| greghaynes | eventually python code is going to be mostly made up of list comprehensions | 07:15 |
| StevenK | Haha | 07:16 |
| StevenK | greghaynes: http://paste.openstack.org/show/119817 | 07:16 |
| *** jerryz has joined #tripleo | 07:16 | |
| StevenK | greghaynes: I have a chroot for py33 on Trusty, with a PPA enabled, 'deb http://ppa.launchpad.net/fkrull/deadsnakes/ubuntu trusty main' | 07:17 |
| greghaynes | ok, uploaded | 07:18 |
| StevenK | Heh, now I can't review it | 07:18 |
| greghaynes | I think you can | 07:19 |
| greghaynes | this rule is codified somewhere | 07:19 |
| greghaynes | but I think its that you need one outsider to +A, essentially | 07:19 |
| *** ifarkas has joined #tripleo | 07:19 | |
| StevenK | py27, pep8 pass | 07:20 |
| StevenK | py26, py27, pep8 pass | 07:20 |
| *** mrunge has quit IRC | 07:21 | |
| StevenK | greghaynes: We win | 07:22 |
| *** jistr has joined #tripleo | 07:23 | |
| greghaynes | w00t | 07:23 |
| *** andreaf has quit IRC | 07:26 | |
| tchaypo | gah. | 07:27 |
| tchaypo | so from the logs it looks like ironic is doing all the right things and the nodes pxe_boot... | 07:27 |
| tchaypo | and then nothing | 07:27 |
| tchaypo | eventually irnoic gives up, turns them off, and tries a new node. | 07:27 |
| tchaypo | I think it would be handy to get on a node so i can see what it thinks is going on, but that's going to have to wait until tomorrow | 07:28 |
| *** mrunge has joined #tripleo | 07:29 | |
| *** killer_prince has joined #tripleo | 07:31 | |
| *** k4n0 has quit IRC | 07:42 | |
| *** dshulyak_ has quit IRC | 07:45 | |
| *** ramishra has joined #tripleo | 07:45 | |
| *** ramishra has quit IRC | 07:50 | |
| *** akrivoka has joined #tripleo | 07:50 | |
| *** dtantsur|afk is now known as dtantsur | 07:50 | |
| *** gfidente has quit IRC | 07:53 | |
| *** k4n0 has joined #tripleo | 07:54 | |
| *** dmellado_ has joined #tripleo | 07:56 | |
| *** dmellado_ has quit IRC | 07:57 | |
| *** lazy_prince has quit IRC | 07:58 | |
| *** killer_prince is now known as lazy_prince | 07:58 | |
| *** gfidente has joined #tripleo | 08:01 | |
| *** athomas has joined #tripleo | 08:03 | |
| *** Matt1 has joined #tripleo | 08:08 | |
| *** shardy_z is now known as shardy | 08:09 | |
| *** jistr has quit IRC | 08:12 | |
| *** gfidente has quit IRC | 08:24 | |
| *** lucasagomes has joined #tripleo | 08:25 | |
| *** takadayuiko has joined #tripleo | 08:26 | |
| *** derekh has joined #tripleo | 08:30 | |
| *** jistr has joined #tripleo | 08:31 | |
| *** regebro|afk is now known as regebro | 08:33 | |
| *** loki184 has quit IRC | 08:37 | |
| *** jp_at_hp has joined #tripleo | 08:40 | |
| *** akrivoka has quit IRC | 08:43 | |
| *** ramishra has joined #tripleo | 08:46 | |
| *** ramishra has quit IRC | 08:51 | |
| *** akrivoka has joined #tripleo | 08:52 | |
| *** takadayuiko is now known as takadayuiko_away | 08:58 | |
| *** akrivoka has quit IRC | 08:59 | |
| *** akrivoka has joined #tripleo | 08:59 | |
| *** loki184 has joined #tripleo | 09:01 | |
| *** gfidente has joined #tripleo | 09:19 | |
| lxsli | tchaypo: still here? | 09:22 |
| *** Andy_Yan has quit IRC | 09:24 | |
| *** athomas has quit IRC | 09:42 | |
| *** pelix has joined #tripleo | 09:44 | |
| *** jml has quit IRC | 09:47 | |
| *** jml has joined #tripleo | 09:49 | |
| *** athomas has joined #tripleo | 09:49 | |
| lxsli | gfidente: ping? | 09:54 |
| *** ramishra has joined #tripleo | 09:56 | |
| gfidente | pong lxsli | 10:00 |
| gfidente | :) | 10:01 |
| gfidente | I know what you want to talk about | 10:01 |
| gfidente | :) | 10:01 |
| *** ramishra has quit IRC | 10:01 | |
| *** ramishra has joined #tripleo | 10:06 | |
| *** jerryz has quit IRC | 10:10 | |
| *** jang1 has joined #tripleo | 10:11 | |
| lxsli | gfidente: :) can you put /tmp/exports on pastebin or something? | 10:11 |
| gfidente | lxsli, yes, thanks for helping with that | 10:11 |
| *** ramishra has quit IRC | 10:14 | |
| gfidente | http://paste.openstack.org/show/119833/ | 10:15 |
| lxsli | cheers | 10:16 |
| *** bashok has joined #tripleo | 10:18 | |
| *** yamahata has quit IRC | 10:19 | |
| lxsli | I like how it offers octal character codes without specifying the charset | 10:19 |
| gfidente | well so it says the exports man page | 10:23 |
| gfidente | but it's not that easy to 'port' those into something meaningful for the bash for or the install command | 10:23 |
| gfidente | so they are valid in the exports file, but not as argument of the install command, unless you translate them into the bash valid oct notation | 10:24 |
| *** ramishra has joined #tripleo | 10:24 | |
| lxsli | I'd be happy to support the quoted and unquoted forms, then document the octal form isn't supported | 10:24 |
| gfidente | quotes have kind of similar issue, I could parse those correctly | 10:25 |
| gfidente | but if you pass something in quotes to install, it creates a dir whose name contains the quotes | 10:27 |
| lxsli | Yes, quotes are interpreted by the shell | 10:28 |
| lxsli | so EG if A=\"abc\", we have to `tr -d '"' <<< $A` then do `install ... "$A"` | 10:29 |
| *** mkerrin has quit IRC | 10:29 | |
| *** pradeep has quit IRC | 10:31 | |
| *** pradeep has joined #tripleo | 10:31 | |
| lxsli | http://paste.openstack.org/show/119841/ | 10:35 |
| lxsli | it's not perfect but then parsing paired delimiters with regexen isn't possible | 10:35 |
| lxsli | oh I forgot to strip quotes, one second | 10:36 |
| *** mkerrin has joined #tripleo | 10:37 | |
| lxsli | http://paste.openstack.org/show/119842/ | 10:37 |
| *** lazy_prince is now known as killer_prince | 10:56 | |
| *** akrivoka has quit IRC | 11:01 | |
| *** ramishra has quit IRC | 11:06 | |
| *** loki184 has quit IRC | 11:14 | |
| gfidente | lxsli, so you put that in quotes | 11:16 |
| gfidente | install command will create a path that has quotes in the name | 11:17 |
| gfidente | try it with the entire for cycle | 11:17 |
| *** dmellado_ has joined #tripleo | 11:17 | |
| *** dmellado_ has quit IRC | 11:18 | |
| *** killer_prince has quit IRC | 11:23 | |
| *** lazy_prince has joined #tripleo | 11:24 | |
| *** rhallisey has joined #tripleo | 11:25 | |
| *** dmellado_ has joined #tripleo | 11:25 | |
| *** ramishra has joined #tripleo | 11:37 | |
| *** ramishra has quit IRC | 11:41 | |
| *** julim has joined #tripleo | 11:45 | |
| *** pensu has quit IRC | 11:46 | |
| *** StevenK has quit IRC | 11:50 | |
| *** StevenK has joined #tripleo | 11:51 | |
| *** dshulyak_ has joined #tripleo | 11:53 | |
| *** lazy_prince is now known as killer_prince | 12:08 | |
| *** weshay has joined #tripleo | 12:10 | |
| *** tzumainn has joined #tripleo | 12:12 | |
| *** rushiagr is now known as rushiagr_away | 12:20 | |
| *** openstackgerrit has joined #tripleo | 12:23 | |
| *** jprovazn has quit IRC | 12:23 | |
| *** spzala has joined #tripleo | 12:24 | |
| *** ramishra has joined #tripleo | 12:25 | |
| *** prad has quit IRC | 12:28 | |
| openstackgerrit | therese-mchale proposed a change to openstack/tripleo-image-elements: Use Percona Packages for Ubuntu and Debian https://review.openstack.org/90134 | 12:29 |
| *** jdob has joined #tripleo | 12:29 | |
| openstackgerrit | Ryan Hallisey proposed a change to stackforge/kolla: Add config setup, api services, and controller servies https://review.openstack.org/127220 | 12:29 |
| *** nosnos has quit IRC | 12:35 | |
| *** nosnos has joined #tripleo | 12:36 | |
| *** nosnos has quit IRC | 12:40 | |
| openstackgerrit | Ryan Hallisey proposed a change to stackforge/kolla: Add config setup, api services, and controller servies https://review.openstack.org/127220 | 12:48 |
| *** rushiagr_away is now known as rushiagr | 12:49 | |
| *** radez_g0n3 is now known as radez | 12:52 | |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: New node registration and upload dialogs https://review.openstack.org/126576 | 12:56 |
| *** pensu has joined #tripleo | 12:56 | |
| *** noslzzp has joined #tripleo | 13:01 | |
| *** ramishra has quit IRC | 13:03 | |
| *** akrivoka has joined #tripleo | 13:04 | |
| *** _crobertsrh is now known as crobertsrh | 13:05 | |
| *** jprovazn has joined #tripleo | 13:05 | |
| *** morazi has joined #tripleo | 13:05 | |
| openstackgerrit | Ryan Hallisey proposed a change to stackforge/kolla: bashate cleaning: nova-conductor/start.sh https://review.openstack.org/127228 | 13:07 |
| openstackgerrit | Ryan Hallisey proposed a change to stackforge/kolla: bashate cleanup: nova-scheduler/start.sh https://review.openstack.org/127229 | 13:07 |
| openstackgerrit | Ryan Hallisey proposed a change to stackforge/kolla: Nova-conductor will create the database https://review.openstack.org/127230 | 13:07 |
| openstackgerrit | A change was merged to openstack/tripleo-heat-templates: Sort inputs and parameters https://review.openstack.org/126926 | 13:08 |
| *** killer_prince is now known as lazy_prince | 13:09 | |
| *** dprince has joined #tripleo | 13:09 | |
| *** sdake has joined #tripleo | 13:12 | |
| *** sdake has quit IRC | 13:12 | |
| *** sdake has joined #tripleo | 13:12 | |
| *** pas-ha has joined #tripleo | 13:13 | |
| pas-ha | hi guys, I have question about disk-image-builder | 13:13 |
| pas-ha | it seems that when creating an image, it updates everything it can | 13:14 |
| pas-ha | is there a setting to disable update of components already installed in the existing image and only install the new ones? | 13:14 |
| *** spzala has quit IRC | 13:21 | |
| lazy_prince | pas-ha: not that i know off.. | 13:22 |
| lazy_prince | pas-ha: isn't it good as it updates to cover for all security issues.. | 13:23 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Add Ironic Discoverd setting to Tuskar-UI https://review.openstack.org/126916 | 13:24 |
| openstackgerrit | A change was merged to openstack/tuskar-ui: Updated auto-discovery process with new api https://review.openstack.org/126790 | 13:33 |
| *** akrivoka has quit IRC | 13:34 | |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: clean up base Dockerfile https://review.openstack.org/127242 | 13:36 |
| pas-ha | lazy_prince, not really. I have a lab where the stock cloud fedora 20 starts fine, but image created from it (only adds 3 packages) does not, some microcode issue - not finding the Sound Card 0_0 | 13:38 |
| *** bashok has quit IRC | 13:39 | |
| jpeeler | pas-ha: sounds like a perfect job for virt-customize | 13:39 |
| openstackgerrit | Radomir Dopieralski proposed a change to openstack/tuskar-ui: New node registration and upload dialogs https://review.openstack.org/126576 | 13:39 |
| pas-ha | jpeeler, thanks for the hint, will check it out | 13:40 |
| jpeeler | pas-ha: this may be helpful (trim as necessary, especially --update) https://github.com/larsks/heat-kubernetes/blob/master/get-image.sh | 13:42 |
| *** julim has quit IRC | 13:47 | |
| *** spzala has joined #tripleo | 13:48 | |
| *** akrivoka has joined #tripleo | 13:51 | |
| *** pas-ha has quit IRC | 13:56 | |
| *** lazy_prince is now known as killer_prince | 13:57 | |
| *** pradeep has quit IRC | 13:58 | |
| openstackgerrit | A change was merged to openstack/os-collect-config: Exit with error status on failed command https://review.openstack.org/111334 | 14:08 |
| *** pensu has quit IRC | 14:12 | |
| *** bnemec has joined #tripleo | 14:26 | |
| *** pensu has joined #tripleo | 14:26 | |
| openstackgerrit | Lennart Regebro proposed a change to openstack/tuskar-ui: Add paging to the nodes list https://review.openstack.org/127259 | 14:30 |
| lxsli | gfidente: http://paste.openstack.org/show/119915/ :D | 14:31 |
| openstackgerrit | Lennart Regebro proposed a change to openstack/tuskar-ui: Add paging to the nodes list https://review.openstack.org/127259 | 14:31 |
| lxsli | skip to the bottom if you want | 14:31 |
| gfidente | uahuhuah thanks | 14:31 |
| gfidente | will look into that | 14:31 |
| gfidente | but feel free to take over the change if you want | 14:31 |
| lxsli | OK I won't take it over but I'll push a new version :) | 14:32 |
| *** pensu has quit IRC | 14:32 | |
| gfidente | ah you want to merge that first and than update? | 14:33 |
| lxsli | yeah two ticks | 14:33 |
| gfidente | fine for me and thanks for that | 14:33 |
| *** ramishra has joined #tripleo | 14:37 | |
| openstackgerrit | Tomas Sedovic proposed a change to openstack/tripleo-heat-templates: Compute and controller templates without merge.py https://review.openstack.org/123100 | 14:37 |
| shardy | dumb question - what do I source to get seed credentials? | 14:39 |
| *** pensu has joined #tripleo | 14:39 | |
| shardy | devtest_undercloud.sh is bailing saying invalid credentials | 14:39 |
| lxsli | scripts/devtest_variables.sh and seedrc iirc | 14:41 |
| *** rainya has joined #tripleo | 14:41 | |
| *** k4n0 has quit IRC | 14:42 | |
| *** ramishra has quit IRC | 14:42 | |
| shardy | lxsli: Thanks, I've tried those and still get Invalid OpenStack Nova credentials so something bust be borken, will dig a bit deeper | 14:43 |
| shardy | OS_PASSWORD in seedrc is "unset" so maybe that's related | 14:44 |
| lxsli | I think that's right actually | 14:45 |
| lxsli | yeah it's 'unset' in my seedrc too | 14:46 |
| *** rdopieralski has quit IRC | 14:49 | |
| openstackgerrit | Peter Belanyi proposed a change to openstack/tripleo-heat-templates: Add converted version of block and object storage https://review.openstack.org/123713 | 14:50 |
| *** pcrews has joined #tripleo | 14:52 | |
| *** yamahata has joined #tripleo | 14:58 | |
| openstackgerrit | Dougal Matthews proposed a change to openstack/python-tuskarclient: Update the default version to use the new API https://review.openstack.org/127188 | 14:58 |
| *** pradeep has joined #tripleo | 14:59 | |
| *** untriaged-bot has joined #tripleo | 15:00 | |
| untriaged-bot | No untriaged bugs so far! \o/ | 15:00 |
| *** untriaged-bot has quit IRC | 15:00 | |
| *** killer_prince has quit IRC | 15:00 | |
| *** prad has joined #tripleo | 15:01 | |
| openstackgerrit | Alexis Lee proposed a change to openstack/tripleo-image-elements: Allow any # of named nfs shares, with options https://review.openstack.org/125968 | 15:03 |
| openstackgerrit | Lennart Regebro proposed a change to openstack/tuskar-ui: Add paging to the nodes list https://review.openstack.org/127259 | 15:03 |
| *** spzala has quit IRC | 15:05 | |
| *** dsneddon_ has quit IRC | 15:07 | |
| *** jistr has quit IRC | 15:08 | |
| *** dsneddon_ has joined #tripleo | 15:09 | |
| *** spzala has joined #tripleo | 15:10 | |
| openstackgerrit | Dougal Matthews proposed a change to openstack/python-tuskarclient: Enable gating on H904 and fix existing violations https://review.openstack.org/127267 | 15:13 |
| *** killer_prince has joined #tripleo | 15:14 | |
| *** killer_prince is now known as lazy_prince | 15:14 | |
| devananda | jang: hi! if you're still tracking the ipmi-proxy work, there's a discussion about it happening in #openstack-ironic | 15:14 |
| openstackgerrit | Lars Kellogg-Stedman proposed a change to stackforge/kolla: remove erroneous bind_host from config https://review.openstack.org/127269 | 15:15 |
| openstackgerrit | A change was merged to stackforge/kolla: clean up base Dockerfile https://review.openstack.org/127242 | 15:16 |
| *** pradeep1 has joined #tripleo | 15:20 | |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Add svc-map to cinder-api https://review.openstack.org/104198 | 15:21 |
| *** pradeep has quit IRC | 15:21 | |
| *** julim has joined #tripleo | 15:22 | |
| openstackgerrit | Dougal Matthews proposed a change to openstack/python-tuskarclient: Enable gating on H405 and fix exiting violations https://review.openstack.org/127272 | 15:22 |
| *** akrivoka has quit IRC | 15:26 | |
| *** corvus is now known as jeblair | 15:30 | |
| *** TheJulia has joined #tripleo | 15:36 | |
| *** dtantsur is now known as dtantsur|afk | 15:37 | |
| *** ramishra has joined #tripleo | 15:37 | |
| *** pradeep has joined #tripleo | 15:38 | |
| *** pradeep1 has quit IRC | 15:38 | |
| *** spzala_ has joined #tripleo | 15:39 | |
| *** spzala has quit IRC | 15:41 | |
| *** spzala_ is now known as spzala | 15:41 | |
| openstackgerrit | Ladislav Smola proposed a change to openstack/tuskar-ui: Setup swift endpoint correctly https://review.openstack.org/127276 | 15:42 |
| *** ramishra has quit IRC | 15:43 | |
| openstackgerrit | Giulio Fidente proposed a change to openstack/tripleo-heat-templates: Passthrough and allNodesConfig for BlockStorage nodes https://review.openstack.org/112075 | 15:44 |
| *** pradeep has quit IRC | 15:48 | |
| *** pradeep1 has joined #tripleo | 15:48 | |
| *** dsneddon_ has quit IRC | 15:49 | |
| *** dsneddon_ has joined #tripleo | 15:50 | |
| *** dshulyak_ has quit IRC | 16:00 | |
| *** pradeep has joined #tripleo | 16:01 | |
| *** pradeep1 has quit IRC | 16:01 | |
| *** derekh has quit IRC | 16:05 | |
| *** yamahata has quit IRC | 16:06 | |
| *** dshulyak_ has joined #tripleo | 16:17 | |
| *** ChuckC__ has joined #tripleo | 16:20 | |
| *** pradeep1 has joined #tripleo | 16:24 | |
| *** pradeep has quit IRC | 16:26 | |
| *** athomas has quit IRC | 16:26 | |
| *** dshulyak_ has joined #tripleo | 16:26 | |
| *** dshulyak_ has quit IRC | 16:27 | |
| *** ChuckC__ has quit IRC | 16:29 | |
| *** blinky_ghost has joined #tripleo | 16:30 | |
| *** ifarkas has quit IRC | 16:31 | |
| blinky_ghost | Hi all, I'm testing tripleo in a virtual env. When I run instack-deploy-overcloud my instances get stuck at "spawning" and I see this error on the logs: 'InstanceDeployFailure: Timeout reached while waiting for PXE deploy of instance a968fdb5-3b47-4b8b-9230-a40a3d0ac7f6\n' What am I missing? thanks | 16:31 |
| *** rhallisey has quit IRC | 16:32 | |
| *** ChuckC___ has joined #tripleo | 16:33 | |
| *** ChuckC___ has quit IRC | 16:33 | |
| *** ChuckC___ has joined #tripleo | 16:33 | |
| *** ChuckC___ has quit IRC | 16:34 | |
| *** ChuckC__ has joined #tripleo | 16:35 | |
| *** ramishra has joined #tripleo | 16:38 | |
| *** ChuckC__ has quit IRC | 16:40 | |
| *** ChuckC__ has joined #tripleo | 16:41 | |
| *** ramishra has quit IRC | 16:42 | |
| openstackgerrit | A change was merged to openstack/diskimage-builder: Handle non-existing *generic kernel and initrd https://review.openstack.org/103062 | 16:44 |
| *** pradeep has joined #tripleo | 16:47 | |
| *** pradeep1 has quit IRC | 16:48 | |
| *** dshulyak_ has joined #tripleo | 16:54 | |
| *** lucasagomes has quit IRC | 16:55 | |
| *** pradeep1 has joined #tripleo | 16:56 | |
| *** ChuckC__ has quit IRC | 16:57 | |
| *** pradeep has quit IRC | 16:57 | |
| *** ChuckC__ has joined #tripleo | 16:57 | |
| *** rhallisey has joined #tripleo | 17:02 | |
| *** andreaf has joined #tripleo | 17:05 | |
| *** jp_at_hp has quit IRC | 17:07 | |
| *** gfidente has quit IRC | 17:08 | |
| bnemec | blinky_ghost: There must be another error in the Nova or Ironic logs that will tell you what actually failed. | 17:12 |
| *** jml has quit IRC | 17:19 | |
| *** ChuckC__ has quit IRC | 17:19 | |
| blinky_ghost | bnemec: I don't think it's using ironic | 17:20 |
| *** pradeep has joined #tripleo | 17:21 | |
| *** jml has joined #tripleo | 17:21 | |
| *** pradeep1 has quit IRC | 17:22 | |
| *** pradeep1 has joined #tripleo | 17:25 | |
| *** pradeep has quit IRC | 17:28 | |
| *** pradeep has joined #tripleo | 17:31 | |
| *** pradeep1 has quit IRC | 17:32 | |
| bnemec | blinky_ghost: Oh, yeah I suppose you're using the icehouse packages. Then it would be nova-baremetal logs instead. | 17:33 |
| blinky_ghost | bnemec: I don't see anything on nova-baremetal-manage.log or nova-baremetal-deploy-helper.log, I'll try to activate debug. Speaking in Juno can you point me a howto to test this? | 17:35 |
| *** jcoufal has quit IRC | 17:35 | |
| bnemec | blinky_ghost: For Juno you have to use the source install instructions in the instack-undercloud Git repo. | 17:37 |
| openstackgerrit | Daneyon Hansen proposed a change to stackforge/kolla: Closes-Bug: 1379442 https://review.openstack.org/127298 | 17:38 |
| blinky_ghost | bnemec: Can I try this on Centos7 box or do I have to use Fedora? | 17:38 |
| bnemec | blinky_ghost: I would suggest Fedora. I don't think we've tested it on centos at all. | 17:38 |
| blinky_ghost | bnemec: you're talking about this repo: https://github.com/agroup/instack-undercloud/blob/master/README-source.md? I see that http://file.rdu.redhat.com/~jslagle/tripleo-images-juno-source/ is not available | 17:40 |
| bnemec | blinky_ghost: Yeah, there's a lot of flux with it right now. I'm not sure where the images are - we're mostly building our own to test different fixes. | 17:43 |
| blinky_ghost | bnemec: Ok, thanks | 17:44 |
| *** pradeep1 has joined #tripleo | 17:46 | |
| *** pradeep has quit IRC | 17:48 | |
| *** blinky_ghost has quit IRC | 17:49 | |
| *** cwolferh has joined #tripleo | 17:51 | |
| *** mrunge has quit IRC | 17:51 | |
| daneyon | any suggestions on troubleshooting a kolla kube container? I want to change .conf settings manually for testing before I go through the process of building a new test image/container. I'm unable to use something lile systemd to restart the service. | 17:53 |
| *** cwolfe__ has quit IRC | 17:54 | |
| *** dshulyak_ has quit IRC | 17:58 | |
| SpamapS | daneyon: restart the container? | 18:00 |
| *** pradeep has joined #tripleo | 18:00 | |
| openstackgerrit | Ben Nemec proposed a change to openstack/diskimage-builder: Check for epel before installing it https://review.openstack.org/127304 | 18:01 |
| *** pradeep1 has quit IRC | 18:02 | |
| *** rainya has quit IRC | 18:05 | |
| *** derekh has joined #tripleo | 18:08 | |
| daneyon | SpamapS when i restart the container, the mod's in .conf revert back to the original content. | 18:08 |
| *** andreaf has quit IRC | 18:08 | |
| daneyon | SpamapS: when i restart the container, the mod's in .conf revert back to the original content. | 18:08 |
| *** rainya has joined #tripleo | 18:08 | |
| *** pradeep1 has joined #tripleo | 18:15 | |
| daneyon | I found the issue with restarting the container, the start.sh always runs the config.sh script which reverts the configs. | 18:16 |
| *** pradeep has quit IRC | 18:16 | |
| *** pelix has quit IRC | 18:17 | |
| *** jprovazn has quit IRC | 18:23 | |
| *** pradeep1 has quit IRC | 18:28 | |
| *** pradeep has joined #tripleo | 18:28 | |
| *** dsneddon_ has quit IRC | 18:37 | |
| *** dsneddon_ has joined #tripleo | 18:38 | |
| *** ramishra has joined #tripleo | 18:40 | |
| *** derekh has quit IRC | 18:44 | |
| *** ramishra has quit IRC | 18:44 | |
| *** andreaf has joined #tripleo | 18:47 | |
| *** spzala has quit IRC | 18:50 | |
| *** dshulyak_ has joined #tripleo | 18:50 | |
| *** dprince has quit IRC | 18:52 | |
| *** noslzzp has quit IRC | 18:56 | |
| SpamapS | daneyon: well there you go. :) | 18:57 |
| daneyon | SpamapS: I think running the config script is something we want to make configurable. I'll submit a bp after i figure out why the glance container(s) are not working. | 18:59 |
| *** rushiagr is now known as rushiagr_away | 19:00 | |
| SpamapS | hrm | 19:00 |
| SpamapS | daneyon: why not edit the source of the configs rather than the in-container config? | 19:00 |
| *** ChuckC__ has joined #tripleo | 19:00 | |
| daneyon | SpamapS: I guess it's a natural troubleshooting workflow for me. | 19:06 |
| *** dshulyak_ has quit IRC | 19:07 | |
| SpamapS | daneyon: well both patterns are probably needed in the long run. | 19:07 |
| daneyon | It appears that even when i go to the minion where the container is running, log into the container and comment-out the config script reference in the start script,a new container ID gets created sometimes? | 19:09 |
| *** jcoufal has joined #tripleo | 19:10 | |
| daneyon | SpamapS: Re editing the source, doesn't this also require that i rebuild the image? If that's the case it can definitly extend the time to dev/troubleshoot as apposed to just testing different settings are restarting a container. | 19:10 |
| SpamapS | daneyon: Yes. Though can't you somehow just boot into the image, change it, and make a new docker container? | 19:13 |
| * SpamapS is pretty docker stupid so maybe not | 19:13 | |
| larsks | daneyon: SpamapS you can use "nsenter" to run commands (including a shell) inside the container namespace. | 19:13 |
| larsks | That's not a docker thing; "nsenter" is part of coreutils. | 19:14 |
| larsks | And you could use "docker commit" to generate a new image, but that way leads to madness (because there is no record of how you reached your end state). | 19:14 |
| larsks | In general, the model will always be (a) update the dockerfile and context, (b) rebuild image, (c) boot new container. | 19:14 |
| *** spzala has joined #tripleo | 19:15 | |
| *** pradeep has quit IRC | 19:17 | |
| *** pradeep has joined #tripleo | 19:17 | |
| SpamapS | larsks: how long are image builds running? | 19:18 |
| SpamapS | diskimage-builder runs are like 5 minutes for all-in-one qcow2 images on a reasonably fast machine | 19:18 |
| larsks | Depends on what you're doing. Each step in a dockerfile is cached, so if you're only changing something at the end it can be very quick. | 19:18 |
| SpamapS | with cached git and cloud image. :-P | 19:18 |
| larsks | If you need to re-run a large package install, it all depends on how fast yum can fetch packages. | 19:18 |
| *** dsneddon_ has quit IRC | 19:18 | |
| *** dsneddon_ has joined #tripleo | 19:20 | |
| daneyon | larsks: i've been using nsenter to get shell access of the containers that I'm troubleshooting. | 19:22 |
| *** pradeep1 has joined #tripleo | 19:24 | |
| *** ChuckC__ has quit IRC | 19:24 | |
| daneyon | larsks: I feel the model you mention make troubleshooting much longer, more steps, difficult, than it should be. I would see an ops team being unhappy with that process. | 19:24 |
| larsks | An ops team that is not tracking their changes in version control is probably going to run into bigger problems. | 19:25 |
| larsks | Because they will have working images that nobody can reproduce. | 19:25 |
| larsks | The time required to build new images if you're just changing config or shell scripts is typically very, very minimal. | 19:26 |
| *** pradeep has quit IRC | 19:26 | |
| *** radez is now known as radez_g0n3 | 19:26 | |
| SpamapS | It's a balance. | 19:26 |
| SpamapS | Some ops teams are disciplined enough to make local changes to debug, and then fold them back into VCS | 19:27 |
| SpamapS | some are not. | 19:27 |
| larsks | SpamapS: For those ops teams that are able to do that, "nsenter" is a great tool for fiddling with an existing image. | 19:27 |
| SpamapS | For those that are not, you need selectors so that they can flip just one container to the debugging image. | 19:27 |
| larsks | Well, that's already available. I mean, images are specified by name, so if you have myanme/someimage and myname/someimage:dev, you can easily use one instead of the other. | 19:28 |
| SpamapS | larsks: except that it sounds like the model being used has docker running the main executables directly, so that will cause configuration pain. | 19:28 |
| larsks | SpamapS: Maybe? If I am debugging, I start a shell with docker and then run things by hand. | 19:29 |
| SpamapS | Yeah it sounds like there are several angles to get at the problem. | 19:29 |
| *** ChuckC__ has joined #tripleo | 19:35 | |
| *** dsneddon_ has quit IRC | 19:40 | |
| *** dsneddon_ has joined #tripleo | 19:41 | |
| greghaynes | \O/ os-collect-config patches finally merged | 19:46 |
| greghaynes | SpamapS: Not sure if you saw, but os-c-c will now exit 1 during a failed --once run, might effect your ansible stuff | 19:47 |
| SpamapS | greghaynes: that would be a good thing | 19:48 |
| *** ChuckC__ has quit IRC | 19:48 | |
| greghaynes | SpamapS: Also, is it acceptable to push reviews at the stackforge tripleo-ansible repo or is that way too out of date at the moment | 19:48 |
| SpamapS | greghaynes: we have a few guards in place after o-c-c runs that might be unnecessary. :) | 19:48 |
| SpamapS | greghaynes: go ahead. I hope to start syncing things back from the HP repo next Monday. | 19:49 |
| greghaynes | ok, hopefully I should be back up to speed enough to help you with that by then :) | 19:50 |
| SpamapS | That would be great | 19:50 |
| SpamapS | I'm kind of conflicted on whether or not to strip out the JIRA tickets. | 19:50 |
| SpamapS | On one hand they are useless to upstream contributors. | 19:50 |
| SpamapS | On the other hand, I'd like the hashes to sync up. :-P | 19:50 |
| greghaynes | Yea, probably just depends on how much the code is getting changed to be pushed upstream... | 19:52 |
| greghaynes | if its mostly just a sync id say keep em | 19:52 |
| SpamapS | Yeah it should be a sync for 90% of changes. | 19:53 |
| SpamapS | There's some Helion specific documentation which won't get synced. | 19:53 |
| *** pradeep has joined #tripleo | 20:02 | |
| *** jml has quit IRC | 20:02 | |
| greghaynes | Oh hey, heat just sent out an email about planning their mid-cycle meetup.... this is an amazing idea | 20:03 |
| *** pradeep1 has quit IRC | 20:04 | |
| *** jml has joined #tripleo | 20:06 | |
| SpamapS | greghaynes: indeed, we need to do the same. | 20:07 |
| *** jcoufal has quit IRC | 20:13 | |
| greghaynes | How do we usually get feedback on possible hosting locations? | 20:15 |
| greghaynes | slagle: ^ youre a rher | 20:15 |
| slagle | greghaynes: i could ask around to see if rh would be willing to host again | 20:17 |
| greghaynes | Yes please :) | 20:18 |
| greghaynes | Ill send out an email with a pretty empty etherpad in a few | 20:18 |
| slagle | i'd just mail the list. when we did that 2 cycles ago, we ended up at Yahoo with awesome free food for a week | 20:18 |
| greghaynes | oh, right | 20:18 |
| slagle | in fact, we should make a decision on who has the free food | 20:19 |
| slagle | clearly rh would be out, unless we want to gorge on granola bars for 5 days | 20:19 |
| greghaynes | Ill add free food as a detail to list by each venue | 20:19 |
| greghaynes | well, rh had bbq | 20:19 |
| greghaynes | next door | 20:19 |
| slagle | you west coasters and your obsession with bbq :) | 20:20 |
| slagle | i guess i take it for granted | 20:20 |
| greghaynes | Yea, I moved away from Texas and now have to take advantage of it whenever I can | 20:21 |
| SpamapS | you do | 20:28 |
| SpamapS | We should try to host in Seattle | 20:28 |
| SpamapS | HP office in DT would accomodate maybe 20 visitors tho.. hm. | 20:29 |
| *** pradeep1 has joined #tripleo | 20:30 | |
| *** pradeep has quit IRC | 20:31 | |
| openstackgerrit | Daneyon Hansen proposed a change to stackforge/kolla: Closes-Bug: 1379442 https://review.openstack.org/127298 | 20:31 |
| slagle | i would go back to seattle. just for elysian. good beer | 20:32 |
| SpamapS | I dub slagle and greghaynes co-venue-czars. | 20:33 |
| SpamapS | slagle: yeah, but we _squished_ into the old office, and the new office is mostly the same size w.r.t. shared conference room space. | 20:34 |
| slagle | it was quite tight as i recall | 20:34 |
| *** GheRivero has left #tripleo | 20:34 | |
| greghaynes | email sent | 20:34 |
| *** GheRivero has joined #tripleo | 20:35 | |
| greghaynes | I would fully support seattle | 20:35 |
| greghaynes | you know, I wonder if I could get a space in portland | 20:35 |
| *** crobertsrh is now known as _crobertsrh | 20:35 | |
| greghaynes | It almost certainly wouldnt have free food though | 20:36 |
| *** rainya has quit IRC | 20:38 | |
| clarkb | psu | 20:40 |
| greghaynes | exactly | 20:40 |
| greghaynes | I mean, they just did debconf there | 20:41 |
| SpamapS | Free food is mostly just a win for whoever is not hosting on their corporate travel budgets. :p | 20:41 |
| *** jdob has quit IRC | 20:47 | |
| nibalizer | greghaynes: much easier for psu to hold events when term is not in session | 20:50 |
| greghaynes | ah | 20:50 |
| clarkb | greghaynea on bbq front I have you covered | 20:51 |
| clarkb | just not this weekend damn rain | 20:51 |
| greghaynes | this is true, so youll be providing the free brisket if we host in portland? | 20:52 |
| greghaynes | :) | 20:52 |
| clarkb | it is a possibility | 20:52 |
| greghaynes | \O/ | 20:52 |
| *** rainya has joined #tripleo | 20:56 | |
| *** jcoufal has joined #tripleo | 20:58 | |
| nibalizer | oh wow | 20:59 |
| *** untriaged-bot has joined #tripleo | 21:00 | |
| untriaged-bot | No untriaged bugs so far! \o/ | 21:00 |
| *** untriaged-bot has quit IRC | 21:00 | |
| *** pradeep1 has quit IRC | 21:01 | |
| *** pradeep has joined #tripleo | 21:01 | |
| *** pradeep1 has joined #tripleo | 21:03 | |
| *** julim has quit IRC | 21:04 | |
| *** pradeep has quit IRC | 21:06 | |
| daneyon | What is the proper dev workflow for implementing changes to a kolla image? Should I use my own namespace for testing, revert back to kollaglue when testing is successful then push? | 21:06 |
| *** pradeep1 has quit IRC | 21:08 | |
| *** pensu has quit IRC | 21:10 | |
| larsks | daneyon: I am working almost entirely in my own namespace right now. At some point we want pushes to the kollaglue/ namespace to happen automatically only after tests pasts...but have neither the infrastructure for that nor the tests. | 21:11 |
| larsks | So in the future, the workflow will be: get things working locally, submit changes to git repository for review with "git review", automatic tests will pass (or not), new images will get built and pushed. | 21:12 |
| larsks | Or something like that. | 21:12 |
| larsks | Oh, and of course there's also peer review in that 'git review' process. | 21:12 |
| *** morazi_ has joined #tripleo | 21:29 | |
| *** morazi has quit IRC | 21:29 | |
| *** jcoufal has quit IRC | 21:29 | |
| *** rhallisey has quit IRC | 21:37 | |
| jpeeler | larsks: does crux not support adding just a role? | 21:41 |
| *** ramishra has joined #tripleo | 21:42 | |
| *** daneyon_ has joined #tripleo | 21:43 | |
| *** jang1 has quit IRC | 21:43 | |
| *** ramishra has quit IRC | 21:46 | |
| *** daneyon has quit IRC | 21:47 | |
| daneyon_ | larsks: Thanks for the explanation. I will use my own namespace then. | 21:55 |
| openstackgerrit | Ryan Brady proposed a change to openstack/tripleo-image-elements: Fix os-svc-install https://review.openstack.org/127363 | 22:13 |
| openstackgerrit | A change was merged to openstack/os-refresh-config: Remove bogus path in tox.ini's flake8 exclude https://review.openstack.org/124671 | 22:19 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: SELinux: Fix keepalived killall denials https://review.openstack.org/127095 | 22:22 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Make sure that keystonemiddleware package is installed https://review.openstack.org/124738 | 22:25 |
| openstackgerrit | A change was merged to openstack/tripleo-image-elements: Configures novnc proxy to listen on local-ipv4 https://review.openstack.org/125326 | 22:31 |
| *** morazi_ has quit IRC | 22:31 | |
| *** dsneddon_ has quit IRC | 22:40 | |
| *** dsneddon_ has joined #tripleo | 22:41 | |
| *** ramishra has joined #tripleo | 22:43 | |
| *** prad has quit IRC | 22:44 | |
| *** ramishra has quit IRC | 22:47 | |
| tchaypo | lxsli: Sorry | 22:57 |
| *** andreaf has quit IRC | 23:01 | |
| tchaypo | greghaynes: I can see in going to have to try to line up a venue in Sydney | 23:07 |
| tchaypo | Just to see if I can change your mind about a 3-day meetup | 23:07 |
| greghaynes | haha, I would totally support AU meetup :) | 23:07 |
| greghaynes | I tend to not get hit bad by the jet lag when going west that far but badly when I go east | 23:08 |
| greghaynes | so kind of would be win | 23:08 |
| SpamapS | tchaypo: does it help at all if it is on the west coast of the US? | 23:08 |
| SpamapS | 2 less hours TZ drift | 23:08 |
| greghaynes | also SEA/SFO probably has a lot more direct flights | 23:09 |
| SpamapS | yeah | 23:09 |
| SpamapS | I kind of wonder if we could find a way to host here in LA | 23:10 |
| tchaypo | 2 hours less tz drift but at 7 hours less travel | 23:10 |
| SpamapS | but would have to be paid space in a hotel. | 23:10 |
| tchaypo | *at least | 23:10 |
| SpamapS | Or I could fit 12 in my basement. ;) | 23:11 |
| mordred | SpamapS: the unfinished side of the Seattle office is quite huge -they did the UX meetup there last week when we were there | 23:12 |
| mordred | it's got tables and wifi | 23:12 |
| tchaypo | West coast is better; I would still prefer a 2-week trip | 23:12 |
| tchaypo | But west coast us is presumably worse for the Europeans | 23:12 |
| * greghaynes puts down SEA HP as a possibility | 23:14 | |
| tchaypo | If it's going to be 3 days of prefer wed-fri or Tuesday to Thursday | 23:16 |
| tchaypo | Starting on Monday means I lose the prior weekend | 23:16 |
| greghaynes | Yea, the long travel thing is a pretty legitimate point, and we have a lot of people from out of country | 23:17 |
| tchaypo | Ending on Friday means I lose the next one | 23:17 |
| SpamapS | mordred: _OH_ | 23:17 |
| SpamapS | mordred: I had no idea. | 23:17 |
| SpamapS | tchaypo: It's ok, we'll just sign you up for some HP internal-only stuff and you can spend 3 weeks in meetings. | 23:18 |
| openstackgerrit | Tapan Sharma proposed a change to openstack/tripleo-image-elements: Support Debian distro for iptables https://review.openstack.org/111369 | 23:19 |
| tchaypo | Actually, scratch that. I probavly prefer Mon-wed over wed-fri | 23:20 |
| *** dsneddon_ has quit IRC | 23:21 | |
| *** dsneddon_ has joined #tripleo | 23:23 | |
| tchaypo | on second thoughts | 23:24 |
| tchaypo | monday-wed is probably preferanly to wed-fri | 23:24 |
| tchaypo | going over I can leave on sunday and arrive sunday, so I get half a weekend | 23:24 |
| tchaypo | ending on friday, I leave the US on saturday and arrive home monday | 23:25 |
| tchaypo | I'd rather end wednesday, leave thursday, arrive home saturday morning | 23:25 |
| StevenK | My vote is for a week. Yes, it's draining, but it gives you three days to get things done. | 23:26 |
| StevenK | Monday is lost to jetlag and feeling terrible, Tuesday-Thursday for useful work, and Friday is mostly with people heading home | 23:26 |
| *** rainya has quit IRC | 23:41 | |
| tchaypo | actually what I really prefer is what I'm doing for paris - stay there for months afterwards | 23:41 |
| *** ramishra has joined #tripleo | 23:44 | |
| *** TheJulia has quit IRC | 23:48 | |
| openstackgerrit | greghaynes proposed a change to openstack-infra/tripleo-ci: Use os-cloud-config and os-apply-config from git https://review.openstack.org/127386 | 23:48 |
| * greghaynes wonders if that ^ will result in some unhappy reviews | 23:48 | |
| *** ramishra has quit IRC | 23:48 | |
| *** TheJulia has joined #tripleo | 23:50 | |
| greghaynes | maybe I should just do it for os-cloud-config since os-a-c isnt really much of a moving target | 23:54 |
| openstackgerrit | greghaynes proposed a change to openstack-infra/tripleo-ci: Use os-cloud-config from git https://review.openstack.org/127386 | 23:55 |
| StevenK | greghaynes: lifeless has made noises that the approach he wants to see is a local pip cache that has the versions we want and we pull from that | 23:56 |
| StevenK | greghaynes: Also, I'd rather only install os-cloud-config HEAD when we are testing os-cloud-config patches | 23:57 |
| greghaynes | in this case we already have os-cloud-config on disk so not sure why wed want a separate cache. With reguards to testing head or pip release - I agree that release is the way to go most everywhere but specifically for os-cloud-config its pretty anoying while were adding features that are tightly coupled with changes in t-i-e or t-i | 23:59 |
| greghaynes | basically means a minimum week long turnaround until you can get the feature passing CI | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!