*** ssurana has quit IRC | 00:12 | |
sdake | ya docker is definatey imploding | 00:14 |
---|---|---|
*** jmccarthy has quit IRC | 00:15 | |
*** dims has quit IRC | 00:15 | |
*** jmccarthy has joined #kolla | 00:15 | |
-openstackstatus- NOTICE: Gerrit is offline for maintenance, ETA updated to 01:00 utc | 00:18 | |
*** ChanServ changes topic to "Gerrit is offline for maintenance, ETA updated to 01:00 utc" | 00:18 | |
*** aginwala has quit IRC | 00:25 | |
sdake | d_code when you get doen with dinner, docker was so corrupted I have to rebuild all my images | 00:27 |
sdake | docker 1.10 seems doa on centos | 00:27 |
sdake | perhaps its the storage backend | 00:27 |
sdake | i am trying 1.11-dev | 00:27 |
sdake | SamYaple ^^ | 00:27 |
sdake | if only someone had invented continuous integration | 00:28 |
sdake | docker only ci's ubuntu ufnortunately | 00:28 |
sdake | even though everyone rolls with centos | 00:28 |
*** sdake_ has joined #kolla | 00:35 | |
*** sdake has quit IRC | 00:35 | |
*** tzn has joined #kolla | 00:42 | |
sdake_ | pretty sweet that an overlayfs driver was added to rpm ;-) | 00:44 |
sdake_ | didn't think they would be able to fix that | 00:44 |
sdake_ | will = ways | 00:44 |
*** dims has joined #kolla | 00:46 | |
sdake_ | d_code this is probably worth a go - --storage-driver overlay in the docker command line in the systemd file | 00:46 |
*** tzn has quit IRC | 00:46 | |
sdake_ | i'll let you know if it deploys when i wrap up here | 00:46 |
sdake_ | devicemapper = POS | 00:46 |
sdake_ | d_code it does some loopback thing which a) sucks for performance b) sucks for functionality c) is chocked full of bugs | 00:47 |
sdake_ | overlayfs is commonly tested | 00:47 |
sdake_ | fantastic nova-toolbox has an exception during build | 00:53 |
*** dims has quit IRC | 00:53 | |
sdake_ | that probably explains why keystone is blocked on from source installs on centos | 00:53 |
*** neilus has joined #kolla | 01:03 | |
*** neilus has quit IRC | 01:10 | |
*** salv-orlando has joined #kolla | 01:14 | |
*** achanda has joined #kolla | 01:15 | |
*** salv-orlando has quit IRC | 01:21 | |
*** ChanServ changes topic to "Kolla IRC meetings on Wednesday 16:30 UTC (Day light savings doesn't apply) - Agenda @ https://wiki.openstack.org/wiki/Meetings/Kolla - IRC channel is *LOGGED* @ http://eavesdrop.openstack.org/irclogs/%23kolla/" | 01:22 | |
*** jasonsb has joined #kolla | 01:23 | |
sdake_ | the chances of survival are 725 to 1 | 01:27 |
*** sdake_ is now known as sdake | 01:27 | |
*** aginwala has joined #kolla | 01:40 | |
sdake | sweet kolla-toolbox building now | 01:41 |
*** openstackgerrit has quit IRC | 01:47 | |
*** openstackgerrit has joined #kolla | 01:47 | |
d_code | lol | 01:47 |
d_code | sdake: might docker 1.9 be better? 1.10 has a regression on cent? | 01:49 |
d_code | 1.9.1 | 01:49 |
openstackgerrit | Steven Dake proposed openstack/kolla: Make kolla-toolbox container build on CentOS bin https://review.openstack.org/279838 | 01:50 |
sdake | d_code I am trying overlayfs | 01:50 |
*** alisonh has quit IRC | 01:50 | |
sdake | if that works, I will change the docs to recommend overlay as the default storage driver | 01:50 |
d_code | cool | 01:50 |
sdake | devicemapper should never be the default driver | 01:50 |
sdake | only overlay or btrfs | 01:50 |
sdake | you will need that toolbox patch | 01:51 |
sdake | toolbox is busted, that is probably why you had the errors on centos in keystone | 01:51 |
sdake | but i'm nto sure yet, deploying now with that fix | 01:51 |
d_code | ah….I did see some recommendations to setup btrfs | 01:54 |
d_code | I’ll be back in about an hour and dig in some more | 01:54 |
sdake | docker fails on trying to umount the toolbox container's /dev/shm | 02:00 |
sdake | its not mounted | 02:00 |
sdake | docker tries to umount it | 02:00 |
sdake | docker returns an error 500 to the api | 02:00 |
sdake | our playbooks should forcibly remove existing contianers or soething | 02:02 |
sdake | because the start operation will fail if the container failed to start | 02:02 |
sdake | need to rebuild | 02:03 |
sdake | groan | 02:03 |
sdake | docker db corrupted yet again | 02:03 |
sdake | this is a kolla bug | 02:03 |
sdake | becausae of a chane in behavior of docker | 02:03 |
sdake | SamYaple didn't catch it because the kolla_toolbox always booted in his source only environment | 02:03 |
sdake | but if a container fails to start, it still is around | 02:04 |
sdake | if it fails to start, it should be removed | 02:04 |
sdake | atleast that is my current speculation | 02:04 |
sdake | i suspect once i get past this kolla toolbox issue the dhcp issue could occur | 02:06 |
sdake | and result in the same 500 error from docker | 02:06 |
*** alisonh has joined #kolla | 02:06 | |
sdake | anyway time for a break while my computer bakes the images bbiaf | 02:06 |
*** achanda has quit IRC | 02:14 | |
*** cfarquhar has quit IRC | 02:24 | |
*** aginwala has quit IRC | 02:28 | |
*** aginwala has joined #kolla | 02:33 | |
d_code | sdake: I agree with you on the removal of containers that failed to start. Without it, you get can easily get into a state where you lose idempotence guarantees | 02:40 |
d_code | in my mucking around of all the ways to run the builds, I’ve found myself where a container goofed because of my error, but I had to do the cleanup-containers to get past it | 02:41 |
*** sdake has quit IRC | 02:43 | |
*** achanda has joined #kolla | 02:45 | |
*** salv-orlando has joined #kolla | 02:46 | |
*** dims has joined #kolla | 02:46 | |
*** dims has quit IRC | 02:50 | |
*** neilus has joined #kolla | 02:52 | |
*** salv-orlando has quit IRC | 02:54 | |
*** Marga_ has quit IRC | 02:55 | |
*** Marga_ has joined #kolla | 02:58 | |
*** neilus has quit IRC | 02:59 | |
*** jtriley has quit IRC | 03:01 | |
*** Marga_ has quit IRC | 03:02 | |
*** Marga_ has joined #kolla | 03:06 | |
*** sdake has joined #kolla | 03:08 | |
sdake | d_code agreed | 03:09 |
sdake | i just rebuilt after a total cleanup and toolbox is returning a 500 server error | 03:09 |
*** d_code has quit IRC | 03:10 | |
*** Marga_ has quit IRC | 03:11 | |
sdake | it is definately failing because it expects shm to be mounted and it is not | 03:11 |
*** asalkeld has quit IRC | 03:11 | |
*** aginwala has quit IRC | 03:12 | |
*** cfarquhar has joined #kolla | 03:14 | |
*** cfarquhar has quit IRC | 03:14 | |
*** cfarquhar has joined #kolla | 03:14 | |
*** cfarquhar has quit IRC | 03:14 | |
*** aginwala has joined #kolla | 03:17 | |
*** d_code has joined #kolla | 03:17 | |
d_code | :/ | 03:19 |
*** cfarquhar has joined #kolla | 03:21 | |
*** cfarquhar has quit IRC | 03:21 | |
*** cfarquhar has joined #kolla | 03:21 | |
*** achanda has quit IRC | 03:21 | |
*** achanda has joined #kolla | 03:22 | |
*** aginwala has quit IRC | 03:24 | |
*** cfarquhar has quit IRC | 03:25 | |
*** sdake has quit IRC | 03:37 | |
*** achanda has quit IRC | 03:48 | |
*** d_code has quit IRC | 03:55 | |
*** Marga_ has joined #kolla | 03:59 | |
*** Marga_ has quit IRC | 04:00 | |
*** Marga_ has joined #kolla | 04:01 | |
*** salv-orlando has joined #kolla | 04:03 | |
*** salv-orlando has quit IRC | 04:13 | |
*** salv-orlando has joined #kolla | 04:17 | |
*** tzn has joined #kolla | 04:18 | |
*** salv-orlando has quit IRC | 04:22 | |
*** tzn has quit IRC | 04:22 | |
*** haplo37 has quit IRC | 04:24 | |
*** haplo37 has joined #kolla | 04:25 | |
*** sdake has joined #kolla | 04:30 | |
*** sdake_ has joined #kolla | 04:34 | |
*** sdake has quit IRC | 04:35 | |
*** neilus has joined #kolla | 04:40 | |
*** achanda has joined #kolla | 04:44 | |
openstackgerrit | Steven Dake proposed openstack/kolla: Test of the Ubuntu gate on master https://review.openstack.org/279852 | 04:45 |
*** neilus has quit IRC | 04:45 | |
*** haplo37 has quit IRC | 05:17 | |
*** mandre has quit IRC | 05:20 | |
*** salv-orlando has joined #kolla | 05:21 | |
*** mandre has joined #kolla | 05:22 | |
*** salv-orlando has quit IRC | 05:31 | |
*** sdake_ has quit IRC | 05:32 | |
*** Marga_ has quit IRC | 06:04 | |
*** neilus has joined #kolla | 06:29 | |
*** neilus has quit IRC | 06:34 | |
*** Marga_ has joined #kolla | 06:36 | |
*** neophy has joined #kolla | 07:01 | |
*** jasonsb has quit IRC | 07:46 | |
openstackgerrit | Michal Rostecki proposed openstack/kolla: Added Elasticearch and its deployment. https://review.openstack.org/267714 | 07:53 |
*** achanda has quit IRC | 07:56 | |
*** salv-orlando has joined #kolla | 08:09 | |
*** athomas has quit IRC | 08:15 | |
*** athomas has joined #kolla | 08:15 | |
*** Jeffrey4l has joined #kolla | 08:17 | |
*** neilus has joined #kolla | 08:17 | |
*** neilus has quit IRC | 08:22 | |
*** achanda has joined #kolla | 08:25 | |
*** neophy has quit IRC | 08:28 | |
*** klint has joined #kolla | 08:47 | |
*** achanda has quit IRC | 08:47 | |
*** achanda has joined #kolla | 08:49 | |
*** salv-orlando has quit IRC | 08:56 | |
*** achanda has quit IRC | 09:04 | |
*** klint has quit IRC | 09:13 | |
*** harmw has quit IRC | 09:17 | |
*** harmw has joined #kolla | 09:30 | |
*** salv-orlando has joined #kolla | 09:57 | |
*** salv-orlando has quit IRC | 10:08 | |
*** iceyao has joined #kolla | 10:54 | |
*** clayton has quit IRC | 11:03 | |
*** clayton has joined #kolla | 11:04 | |
*** athomas has quit IRC | 11:04 | |
*** AJaeger has left #kolla | 11:06 | |
*** salv-orlando has joined #kolla | 11:15 | |
*** dwalsh has joined #kolla | 11:17 | |
*** dwalsh has quit IRC | 11:23 | |
*** salv-orlando has quit IRC | 11:25 | |
*** limamauricio has joined #kolla | 11:28 | |
*** fthiagogv has joined #kolla | 11:28 | |
*** dwalsh has joined #kolla | 11:29 | |
*** dwalsh has quit IRC | 11:33 | |
*** mikelk has joined #kolla | 11:35 | |
*** dwalsh has joined #kolla | 11:39 | |
limamauricio | Hello, I'm ran kolla-ansible deploy and now I have this error http://pastebin.com/UNxs2mac | 11:43 |
*** dwalsh has quit IRC | 11:43 | |
limamauricio | This error is totally new for me, someone help me? | 11:44 |
*** neilus has joined #kolla | 12:10 | |
*** dwalsh has joined #kolla | 12:20 | |
fthiagogv | This error occuring with me too. Seeing keystone container logs is shown this msg: "AH00558: httpd: Could not reliably determine the server's fully qualified domain name, using 10.102.20.155. Set the 'ServerName' directive globally to suppress this message" | 12:46 |
fthiagogv | Verifying roles for keystone, there is not anything about apache2.conf, such as official documentation. Is this a bug or is not needed this config? | 12:46 |
*** mikelk has quit IRC | 12:49 | |
*** dwalsh has quit IRC | 12:53 | |
fthiagogv | ls | 13:05 |
fthiagogv | ops | 13:05 |
*** dave-mccowan has joined #kolla | 13:22 | |
*** d_code has joined #kolla | 13:28 | |
*** dims has joined #kolla | 13:30 | |
*** liyi has quit IRC | 13:52 | |
*** salv-orlando has joined #kolla | 13:59 | |
*** salv-orlando has quit IRC | 14:12 | |
*** sdake has joined #kolla | 14:36 | |
fthiagogv | Hi, I found this error: stdout: An exception occurred during task execution. To see the full traceback, use -vvv. The error was: TypeError: Value of unknown type: <class 'keystoneauth1.exceptions.http.Unauthorized'>, The request you have made requires authentication. (HTTP 401) (Request-ID: req-365e110d-8357-4528-8020-d48d4bb2c47f). Do someone have any ideia? | 14:43 |
fthiagogv | This occur when i am trying deploy. | 14:44 |
fthiagogv | Also, this is the TASK: [keystone | Creating the admin project, user and role] | 14:45 |
sdake | samyaple you around this fine morning | 14:52 |
sdake | fthiagogv yes we are aware of it | 14:53 |
sdake | fthiagogv it affects both centos nd ubuntu | 14:53 |
sdake | it is either a recent regression in kolla or a change in upstream that we have not adapted to | 14:53 |
sdake | recent as in less then 1 week. | 14:53 |
*** neilus has quit IRC | 14:59 | |
*** fthiagogv has quit IRC | 15:02 | |
*** dims has quit IRC | 15:03 | |
-openstackstatus- NOTICE: Gerrit is offline for filesystem repair | 15:06 | |
*** ChanServ changes topic to "Gerrit is offline for filesystem repair" | 15:07 | |
*** dims has joined #kolla | 15:10 | |
sdake | https://bugs.launchpad.net/kolla/+bug/1545292 | 15:14 |
openstack | Launchpad bug 1545292 in kolla "regression in last week causes keystone to return unauthorized" [Critical,Confirmed] - Assigned to Steven Dake (sdake) | 15:14 |
sdake | fth^^ | 15:14 |
*** salv-orlando has joined #kolla | 15:20 | |
*** salv-orlando has quit IRC | 15:25 | |
*** daneyon has joined #kolla | 15:30 | |
*** Jeffrey4l has quit IRC | 15:31 | |
*** achanda has joined #kolla | 15:38 | |
*** iceyao has quit IRC | 15:41 | |
*** achanda has quit IRC | 15:46 | |
*** neilus has joined #kolla | 15:49 | |
*** neilus has quit IRC | 15:54 | |
*** stvnoyes has quit IRC | 16:08 | |
*** stvnoyes has joined #kolla | 16:09 | |
*** absubram has joined #kolla | 16:15 | |
*** absubram_ has joined #kolla | 16:17 | |
SamYaple | sdake: for a minute | 16:18 |
sdake | ya ubuntu is busted centos is busted | 16:18 |
SamYaple | master? | 16:18 |
sdake | master is busted | 16:18 |
sdake | yes master | 16:18 |
sdake | https://bugs.launchpad.net/kolla/+bug/1545292 | 16:19 |
openstack | Launchpad bug 1545292 in kolla "regression in last week causes keystone to return unauthorized" [Critical,Confirmed] - Assigned to Steven Dake (sdake) | 16:19 |
*** absubram has quit IRC | 16:19 | |
*** absubram_ is now known as absubram | 16:19 | |
SamYaple | im not sure any _regression_ happened | 16:19 |
sdake | i'm going to submit a test patch with the last week's worth of stuff reverted - that should tell us if its keystone our our patches | 16:20 |
SamYaple | I dont think that codes been touched | 16:20 |
sdake | ya if you read in the bug it could be a keystone change | 16:20 |
sdake | i wasn't sure | 16:20 |
sdake | its a regressoin in keystone or kolla | 16:20 |
sdake | or new feature we have to ddal with | 16:20 |
sdake | anyway just getting around to getting my day started i'll look into it | 16:20 |
sdake | enjoy | 16:20 |
SamYaple | Im testing ubuntu now | 16:21 |
sdake | hard to do anything with gerrit offline | 16:21 |
SamYaple | it was working as of last night | 16:21 |
SamYaple | yea gerrit has been foobar for a while i think | 16:21 |
SamYaple | i tried it yesterday too :/ | 16:21 |
sdake | they upgraded friday | 16:21 |
SamYaple | ah | 16:21 |
sdake | so SamYaple check out the bug log | 16:21 |
SamYaple | this explains it | 16:21 |
sdake | its got a console log | 16:21 |
sdake | where I changed nothing but loc file | 16:21 |
sdake | that fails with unauthorized | 16:22 |
sdake | it looks like a type error | 16:22 |
sdake | which doesn't even make sense to me | 16:22 |
SamYaple | ayoung: did you guys mess with token based auth in the past few days? | 16:23 |
sdake | pretty sure ayoung doesn't ever work on the weekends ;) | 16:24 |
SamYaple | not sure this is urgent so its fine | 16:24 |
sdake | everyone complaining | 16:24 |
sdake | so its urgent | 16:24 |
sdake | we have had 6-8 people in lasst 3 days try to deploy | 16:25 |
sdake | keep hitting this problem | 16:25 |
sdake | these are new operators or devs | 16:25 |
SamYaple | i built and deployed last night | 16:25 |
sdake | i dont know what to tell you | 16:25 |
sdake | its been happening for several days from what I have seen on irc | 16:25 |
SamYaple | sdake: https://github.com/openstack/keystone/commit/5286b4a297b5a94895a311a9e564aa87cb54dbfd | 16:26 |
SamYaple | this is a change i asked for back in liberty | 16:26 |
SamYaple | happy its here now | 16:26 |
SamYaple | ill have a patch forth coming | 16:27 |
sdake | thanks | 16:27 |
sdake | deprecate is not the same hting as remove though ;-) | 16:27 |
SamYaple | if you read the patch you should understand it | 16:27 |
SamYaple | its no longer a default middleware | 16:27 |
sdake | yup just saw the title real quick | 16:27 |
SamYaple | so if we _add_ it back this all works | 16:27 |
SamYaple | so it is deprecated, not removed | 16:28 |
SamYaple | but default deploys break if they rely on it | 16:28 |
SamYaple | since its not default anymore | 16:28 |
sdake | right | 16:28 |
sdake | ok please take over the bug | 16:28 |
SamYaple | alrighty | 16:28 |
sdake | i'd like the assignees to match up with the committers | 16:29 |
SamYaple | cant do much without gerrit | 16:29 |
SamYaple | so maybe later tonight | 16:29 |
sdake | i hear ya | 16:29 |
sdake | ya gerrit will probably be down for awhile | 16:29 |
sdake | they did a mssive upgrade friday which took 3 hours longer then their window | 16:29 |
sdake | i suspect they may revert that upgrade ;-) | 16:29 |
sdake | its happened before | 16:30 |
SamYaple | the keystone or gerrit one? | 16:30 |
sdake | gerrit | 16:30 |
sdake | so master has been broken for approximately 10 days? | 16:31 |
SamYaple | yea maybe | 16:31 |
SamYaple | nooooo | 16:31 |
sdake | you must not be rebuilding properly | 16:31 |
SamYaple | absolutely not | 16:31 |
sdake | the patch was committed 10 days ago? | 16:31 |
sdake | or only merged recently | 16:31 |
SamYaple | the patch was submitted to _gerrit_ 10 days ago | 16:31 |
sdake | i can't tell shit without gerrit | 16:31 |
SamYaple | merged yesterday | 16:31 |
sdake | nice | 16:31 |
sdake | so like those 5+ people i told you about | 16:31 |
sdake | they have been on our asses for a fix for the last couple days ;) | 16:32 |
sdake | tells you we are loved :) | 16:32 |
SamYaple | not possible but ok | 16:32 |
sdake | well whenever the commit went in | 16:32 |
SamYaple | worst case its been 36 hours with this issue | 16:32 |
sdake | until today | 16:32 |
SamYaple | looks like its been less than 24, but again with gerrit... | 16:32 |
sdake | yes, what i'm getting at is thats good | 16:32 |
sdake | 24 hours with so many people trying to deploy | 16:32 |
sdake | irc thur/fri was off the hook busy | 16:33 |
SamYaple | yea i read the scrollback | 16:33 |
sdake | sometimes i read the logs but my scroll buffer isn't long enough to capture my sleep time | 16:33 |
*** pbourke has joined #kolla | 16:34 | |
sdake | hey pbourke | 16:34 |
pbourke | hey there sdake | 16:34 |
pbourke | second last day of the trip | 16:34 |
sdake | there si another problem with centos alteast | 16:34 |
sdake | after I fixed teh busted toolbox container | 16:34 |
sdake | docker (with overlay storage driver) was unable to umount the shm file for the container | 16:35 |
sdake | which causes container start to fail | 16:35 |
sdake | which causes ansible to go byebye | 16:35 |
sdake | shouldn't we delete containers before starting them ? | 16:35 |
sdake | and what the hell is a sandbox id | 16:35 |
SamYaple | kolla-mesos? | 16:38 |
sdake | ansible | 16:38 |
sdake | moment trying to get a log out | 16:38 |
sdake | of my vm | 16:38 |
sdake | rather not trying doing | 16:38 |
sdake | but my cmoputer just went to vm sleep - out of batteries | 16:38 |
SamYaple | sorry im not sure what you are asking | 16:39 |
sdake | hang tight | 16:39 |
SamYaple | so we are back to passing username/password into keystone contaienr to bootstrap | 16:41 |
SamYaple | im going to submit a "this makes it work" patch with the proper non-deprecated fix afterward | 16:41 |
sdake | http://paste.fedoraproject.org/322272/53820351/ | 16:47 |
sdake | take a look at line 95 SamYaple | 16:48 |
sdake | it just implodes from that point | 16:48 |
sdake | this is early on in the boot process, starting kolla-toolbox | 16:48 |
*** absubram has quit IRC | 16:48 | |
SamYaple | sounds selinux-ish | 16:49 |
sdake | disabled | 16:50 |
sdake | then shm can't umount | 16:50 |
sdake | presumably becauase the container never started | 16:50 |
sdake | presumeably because of that sanbox id | 16:50 |
sdake | the gate doesn't fail in this way iirc | 16:51 |
sdake | i've read through the gate scripts | 16:51 |
sdake | but our userbase is running into this problem too | 16:51 |
ayoung | sdake, any chance it is a Trust related issue? | 16:55 |
sdake | ayoung you do work on weekends i guess :) | 16:55 |
sdake | ayoung I think SamYaple found the commit that did it | 16:55 |
ayoung | sdake, I leave my chat up on weekends anyway | 16:55 |
sdake | admin_token_something or another | 16:55 |
ayoung | link>? | 16:55 |
sdake | recent change, so should be easy to fix | 16:55 |
SamYaple | ayoung: it was the deprecation of the admin token | 16:56 |
sdake | https://github.com/openstack/keystone/commit/5286b4a297b5a94895a311a9e564aa87cb54dbfd | 16:56 |
SamYaple | https://github.com/openstack/keystone/commit/5286b4a297b5a94895a311a9e564aa87cb54dbfd | 16:56 |
SamYaple | ayoung: already have a patch for kolla, but gerrit is down | 16:56 |
ayoung | Still? | 16:56 |
sdake | ayoung we are good | 16:56 |
sdake | ayoung no it went down about 2 hours ago | 16:56 |
SamYaple | ayoung: i guess. i cant review or push | 16:56 |
sdake | gerrit had a filesystem corruption | 16:57 |
sdake | so who knows when it will be back ;-) | 16:57 |
SamYaple | it ran out of inodes | 16:57 |
ayoung | sdake, if you can fix it in Kolla that is better than reverting in Keystone. BTW is this about initializeing Keystone? | 16:57 |
SamYaple | ayoung: yea | 16:57 |
SamYaple | its the keystone-manage bootstrap command | 16:57 |
sdake | ayoung ya we will fix in kolla | 16:57 |
SamYaple | we can use that | 16:57 |
ayoung | SamYaple, AWESOME! | 16:57 |
ayoung | You are going to make morgainfainberg very happy with that | 16:57 |
SamYaple | if you recall i brought that up to you guys with mordred around the shade time | 16:57 |
SamYaple | what that was getting started | 16:57 |
SamYaple | happy to see it land | 16:58 |
ayoung | admin-token was always a scary concept | 16:58 |
sdake | ayoung we dont ask upstreams to revert patches unless the patch is wrong ;) | 16:58 |
SamYaple | ayoung: indeed | 16:58 |
ayoung | sdake, unlike devstack that does that to us all the time.... | 16:58 |
sdake | changing the default middleware is a little wierd - wouldn't expect taht in a deprecation, but whatever works | 16:58 |
ayoung | "but people are still using that feature you deprecated back in Cactus..." | 16:58 |
sdake | we follow the deprecation policies of our upstream | 16:59 |
SamYaple | onward and upward is my motto | 16:59 |
sdake | if you deprecate on a 6 month cycle, we will deal with it | 16:59 |
SamYaple | yea sdake this first patch is just going to add admin_token_auth back to the middleware | 17:00 |
SamYaple | im going to do a follow up with the keystone-manage bootstrap commands to make it proper | 17:00 |
sdake | understood | 17:01 |
sdake | wish I could find my damn password for freenode | 17:02 |
sdake | so i can go harass the docker folks :) | 17:02 |
sdake | SamYaple which version of docker-py are you running | 17:03 |
*** pbourke has quit IRC | 17:04 | |
sdake | https://github.com/docker/docker/issues/19934 | 17:05 |
sdake | SamYaple fwiw I wouldn't waste a bunch of time on user namespaces, as it doesn't work on OLE/CENTOS/etc https://github.com/docker/docker/issues/20193 | 17:13 |
d_code | sdake: so…we know the fix for the centos-source bug but can’t commit until Gerrit is back up, right? Can I try the patch locally? | 17:27 |
d_code | for the record, I’ve been in IRC since Tuesday trying to get this working ;-) | 17:27 |
d_code | I know there was a midcycle conference… I’m just convinced that Kolla is the answer to my problems (as soon as I get some selinux policies written) :-) | 17:28 |
sdake | d_code you had a different problem | 17:28 |
d_code | oh…thought this was the auth problem | 17:28 |
sdake | d_code you have a problem I am debugging now | 17:29 |
sdake | the auth problem ahppened when you deployed from source | 17:29 |
d_code | right | 17:29 |
sdake | when you deploy from binary, you get the dhcp doesn't s tart problem | 17:29 |
d_code | the dhcp problem happened for binary | 17:29 |
sdake | you didn't do soure build until last day or two | 17:29 |
d_code | right | 17:29 |
sdake | i htink friday night | 17:29 |
sdake | so that will be fixed once gerrit is online and you should be good to go from a source-build perspective | 17:29 |
sdake | SamYaple can you fpaste your patch please? | 17:29 |
sdake | so d_code can get on with his work :) | 17:30 |
sdake | i am chasing an entirely different problem, kolla_toolbox doesn't work | 17:30 |
sdake | d_code you will definately need the patch i posted in gerrit last night | 17:30 |
sdake | oh no that patch only affects binary | 17:31 |
sdake | ya with SamYaple's keystone change, you should be good to go | 17:31 |
sdake | d_code when working with master, upstreams naturally deprecate functionality | 17:32 |
sdake | d_code when they do, we fix it, but it takes a couple days turnaround time to figure out what changed and how to fix it | 17:32 |
d_code | yeah…no problem…I get it | 17:32 |
sdake | a proper deprecation shouldn't change existing functionality | 17:32 |
sdake | the deprecation in keystone changed existing functionality | 17:32 |
sdake | centos binary, not sure - still working on it | 17:33 |
sdake | do you ahve a preference? | 17:33 |
sdake | or just want to getit deployed | 17:33 |
d_code | long term, I prefer binary….but for now I just want to get it deployed so that I can start working with the images | 17:33 |
sdake | ok | 17:33 |
sdake | well I'll try to get you binary in short order, but I am in the middle of a remodel and probably going to be spsending most of the day hanging closets | 17:34 |
d_code | if we can figure out the binary by April…I’ll be good :) | 17:34 |
sdake | ya i'd say by next wed it will all be in repo | 17:35 |
sdake | SamYaple you still about - re schedule on stable | 17:36 |
*** ChanServ changes topic to "Kolla IRC meetings on Wednesday 16:30 UTC (Day light savings doesn't apply) - Agenda @ https://wiki.openstack.org/wiki/Meetings/Kolla - IRC channel is *LOGGED* @ http://eavesdrop.openstack.org/irclogs/%23kolla/" | 17:37 | |
-openstackstatus- NOTICE: Gerrit is back up | 17:37 | |
sdake | ya gerrit | 17:37 |
*** neilus has joined #kolla | 17:37 | |
*** mikelk has joined #kolla | 17:39 | |
*** neilus has quit IRC | 17:42 | |
*** Marga_ has quit IRC | 17:43 | |
sdake | d_code thanks for sticking with us, I know its been a bit trying :) | 17:44 |
d_code | meh… I’m no OpenStack expert (yet), but I’m an experienced enough developer to recognize quality engineering and relatively minor bugs (even if they’re temporary show stoppers) | 17:45 |
d_code | I like that Kolla provides a pretty flexible deployment model. Most of the OpenStack deployment tools I’ve played with are pretty opinionated on what goes where | 17:45 |
sdake | thanks for the positive feedback without actually have seen a deployment in action ;) | 17:47 |
sdake | remind me again were you more interested in a liberty release or mitaka release | 17:47 |
sdake | well kolla_toolbox runs *manually* | 17:59 |
sdake | runs with --detached | 18:00 |
sdake | runs with name kolla_toolbox | 18:00 |
sdake | does not run from ansible | 18:02 |
sdake | implies kolla_docker module is wrong in some way | 18:02 |
*** jasonsb has joined #kolla | 18:03 | |
*** dims has quit IRC | 18:30 | |
*** haplo37 has joined #kolla | 18:33 | |
*** haplo37 has quit IRC | 18:35 | |
*** haplo37 has joined #kolla | 18:36 | |
*** sdake has quit IRC | 18:42 | |
*** Marga_ has joined #kolla | 18:43 | |
openstackgerrit | Lingfeng Xiong proposed openstack/kolla: Add backward compatibility for liberty (ubuntu binary) https://review.openstack.org/279892 | 18:58 |
*** pbourke has joined #kolla | 19:00 | |
*** pbourke has quit IRC | 19:01 | |
openstackgerrit | Lingfeng Xiong proposed openstack/kolla: Fix horizon startup failure https://review.openstack.org/279893 | 19:09 |
*** haplo37 has quit IRC | 19:12 | |
*** sdake has joined #kolla | 19:22 | |
*** neilus has joined #kolla | 19:26 | |
*** aginwala has joined #kolla | 19:28 | |
*** neilus has quit IRC | 19:30 | |
*** aginwala has quit IRC | 19:47 | |
*** aginwala has joined #kolla | 19:54 | |
*** neilus has joined #kolla | 19:56 | |
*** d_code has quit IRC | 20:01 | |
*** d_code has joined #kolla | 20:03 | |
*** sdake has quit IRC | 20:16 | |
*** aginwala has quit IRC | 20:18 | |
*** sdake has joined #kolla | 20:38 | |
*** dwalsh has joined #kolla | 20:57 | |
*** d_code has quit IRC | 21:02 | |
*** dims has joined #kolla | 21:04 | |
*** dwalsh has quit IRC | 21:05 | |
*** d_code has joined #kolla | 21:05 | |
*** dwalsh has joined #kolla | 21:06 | |
*** dwalsh has quit IRC | 21:11 | |
*** iNeilus has joined #kolla | 21:24 | |
*** iNeilus has quit IRC | 21:25 | |
*** neilus_ has joined #kolla | 21:25 | |
*** neilus has quit IRC | 21:27 | |
*** neilus has joined #kolla | 21:28 | |
*** neilus_ has quit IRC | 21:28 | |
*** iNeilus has joined #kolla | 21:31 | |
*** neilus has quit IRC | 21:34 | |
*** sdake has quit IRC | 22:08 | |
*** alisonh has quit IRC | 22:14 | |
*** alisonh has joined #kolla | 22:14 | |
*** bmace has quit IRC | 22:22 | |
*** bmace has joined #kolla | 22:23 | |
*** salv-orlando has joined #kolla | 22:29 | |
*** salv-orlando has quit IRC | 22:38 | |
*** sdake has joined #kolla | 22:42 | |
*** mikelk has quit IRC | 22:44 | |
*** jtriley has joined #kolla | 22:53 | |
*** jtriley has quit IRC | 23:09 | |
*** salv-orlando has joined #kolla | 23:52 | |
*** sdake has quit IRC | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!