| *** Slower has quit IRC | 00:12 | |
| *** achanda has quit IRC | 00:13 | |
| *** EricGonczer_ has joined #openstack-containers | 00:26 | |
| *** sdake_ has quit IRC | 00:30 | |
| *** EricGonczer_ has quit IRC | 00:49 | |
| openstackgerrit | Merged stackforge/magnum: Change Service name to magnum https://review.openstack.org/148059 | 01:07 |
|---|---|---|
| *** dimsum__ has joined #openstack-containers | 01:33 | |
| openstackgerrit | Merged stackforge/magnum: Port parsable error middleware from Ironic https://review.openstack.org/148069 | 01:35 |
| *** dimsum__ has quit IRC | 01:38 | |
| openstackgerrit | Merged stackforge/magnum: Port NoExceptionTracebackHook from Ironic https://review.openstack.org/148070 | 01:41 |
| *** EricGonczer_ has joined #openstack-containers | 01:53 | |
| *** Slower has joined #openstack-containers | 02:04 | |
| *** erkules_ has joined #openstack-containers | 02:36 | |
| *** erkules has quit IRC | 02:38 | |
| *** dimsum__ has joined #openstack-containers | 02:38 | |
| *** dimsum__ has quit IRC | 02:43 | |
| *** sdake has quit IRC | 02:43 | |
| openstackgerrit | Steven Dake proposed stackforge/magnum: Add note about heat finishing the job before continuing https://review.openstack.org/148063 | 02:52 |
| openstackgerrit | Steven Dake proposed stackforge/magnum: Log error and exit if templates are not installed https://review.openstack.org/148067 | 02:57 |
| *** EricGonczer_ has quit IRC | 02:57 | |
| *** sdake has joined #openstack-containers | 02:57 | |
| sdake | yar | 02:57 |
| *** kebray has joined #openstack-containers | 03:07 | |
| sdake | all the error checking in magnum needs love | 03:36 |
| sdake | I can crash it 5 different ways in the last 10 minutes | 03:36 |
| sdake | we need to have sync between conductor and heat- that owuld help | 03:36 |
| jay-lau-513 | sdake can you help file bugs for all of your issues so that we can help | 03:42 |
| jay-lau-513 | sdake yes, we need better error handling | 03:42 |
| *** fandi has quit IRC | 04:54 | |
| *** fandi has joined #openstack-containers | 04:56 | |
| *** kebray has quit IRC | 05:04 | |
| *** hongbin has quit IRC | 05:44 | |
| *** LinstatSDR has quit IRC | 05:44 | |
| *** dimsum__ has joined #openstack-containers | 07:17 | |
| *** Marga_ has joined #openstack-containers | 07:19 | |
| *** LinstatSDR has joined #openstack-containers | 07:20 | |
| *** dimsum__ has quit IRC | 07:22 | |
| *** Marga_ has quit IRC | 07:31 | |
| *** LinstatSDR has quit IRC | 07:31 | |
| *** LinstatSDR has joined #openstack-containers | 07:31 | |
| *** LinstatSDR has quit IRC | 07:55 | |
| *** achanda has joined #openstack-containers | 08:22 | |
| *** LinstatSDR has joined #openstack-containers | 08:46 | |
| *** LinstatSDR has quit IRC | 08:49 | |
| *** oro has joined #openstack-containers | 09:12 | |
| *** achanda has quit IRC | 10:09 | |
| *** dimsum__ has joined #openstack-containers | 10:22 | |
| *** dimsum__ has quit IRC | 10:27 | |
| *** erkules_ is now known as erkules | 10:44 | |
| *** gtt116_ has joined #openstack-containers | 11:21 | |
| *** gtt116__ has quit IRC | 11:24 | |
| *** xxj has quit IRC | 11:26 | |
| *** oro has quit IRC | 11:28 | |
| *** xxj has joined #openstack-containers | 11:31 | |
| *** oro has joined #openstack-containers | 12:00 | |
| *** dimsum__ has joined #openstack-containers | 12:25 | |
| *** Marga_ has joined #openstack-containers | 12:49 | |
| *** oro has quit IRC | 13:02 | |
| *** Marga_ has quit IRC | 13:08 | |
| *** Marga_ has joined #openstack-containers | 13:14 | |
| *** Marga_ has quit IRC | 13:36 | |
| *** Marga_ has joined #openstack-containers | 13:43 | |
| *** diga has joined #openstack-containers | 14:03 | |
| *** oro has joined #openstack-containers | 14:06 | |
| *** Marga_ has quit IRC | 14:10 | |
| *** dimsum__ has quit IRC | 14:14 | |
| *** EricGonczer_ has joined #openstack-containers | 14:39 | |
| *** fandi has quit IRC | 14:41 | |
| *** EricGonczer_ has quit IRC | 14:51 | |
| *** dimsum__ has joined #openstack-containers | 15:15 | |
| *** dimsum__ has quit IRC | 15:20 | |
| *** fandi has joined #openstack-containers | 15:21 | |
| *** dimsum__ has joined #openstack-containers | 15:26 | |
| *** EricGonczer_ has joined #openstack-containers | 15:33 | |
| sdake | jay-lau-513 I was trying to reproduce them to file them into wseparae bugs but got lost on that task | 15:41 |
| *** dimsum__ has quit IRC | 15:51 | |
| openstackgerrit | Steven Dake proposed stackforge/magnum: Add note about heat finishing the job before continuing https://review.openstack.org/148063 | 16:14 |
| openstackgerrit | Steven Dake proposed stackforge/magnum: Log error and exit if templates are not installed https://review.openstack.org/148067 | 16:17 |
| openstackgerrit | Jay Lau proposed stackforge/magnum: Add more tests for test_objects.py https://review.openstack.org/148113 | 16:17 |
| sdake | morning | 16:19 |
| *** hongbin has joined #openstack-containers | 16:19 | |
| hongbin | ping sdake | 16:20 |
| *** oro has quit IRC | 16:20 | |
| openstackgerrit | Steven Dake proposed stackforge/magnum: Log error and exit if templates are not installed https://review.openstack.org/148067 | 16:22 |
| sdake | shoot hognbin | 16:26 |
| sdake | hongbin | 16:26 |
| hongbin | hey | 16:26 |
| hongbin | two things about the quick start | 16:27 |
| hongbin | 1. I think we'd better to clain the OS/version we tested | 16:27 |
| hongbin | E.g. Ubuntu 14.04 | 16:27 |
| sdake | Fedora 21 has been tested, and yaunying claimed he tested ubuntu 14.04 | 16:27 |
| sdake | so sure, add areview for those two oses | 16:27 |
| *** dimsum__ has joined #openstack-containers | 16:28 | |
| hongbin | k | 16:28 |
| hongbin | 2. For the bay create | 16:28 |
| sdake | you mean the host os to run magnum on right? | 16:28 |
| sdake | the only micro os supported at this point is fedora at omic | 16:28 |
| hongbin | It provision 3 m1.medium nodes. Is that correct? | 16:28 |
| sdake | if you set nodecount to 2, yo uwill get 3 m1.medium nodes | 16:28 |
| sdake | I usually set nodecount to 1 | 16:28 |
| hongbin | The thing is users usually don't have that much resources. | 16:29 |
| sdake | good point | 16:29 |
| sdake | maybe m1.small then | 16:29 |
| hongbin | That would be great | 16:29 |
| sdake | m1.small is too small for my normal use, but should fit in a 8gb footprint | 16:29 |
| sdake | I have 16gb footprint and have 64gb on order | 16:29 |
| hongbin | k | 16:30 |
| sdake | so your going to do those two reviews? | 16:30 |
| sdake | how much ram do you ahve? | 16:30 |
| sdake | perhaps we should put a tested on fedora 21, ubutnux with 8gb ram or something similar | 16:30 |
| hongbin | I usually run devstack in a m1.large | 16:30 |
| sdake | oh you run devstack in a vm | 16:31 |
| hongbin | yes | 16:31 |
| sdake | I see, I run devstack on bare metal | 16:31 |
| hongbin | you have resources :) | 16:31 |
| hongbin | Yes, I can test it in a vm. m1.small, node-count is 2. Correct? | 16:32 |
| sdake | that should work | 16:32 |
| hongbin | k | 16:32 |
| sdake | that should create 3 2gb nodes | 16:32 |
| sdake | m1.small is a bit too small for running kolla for example | 16:32 |
| hongbin | sounds good. | 16:32 |
| hongbin | en | 16:32 |
| sdake | but it will run one pod like the xamples show | 16:33 |
| hongbin | k | 16:33 |
| hongbin | I will go through the guide with m1.small to validate the steps first. | 16:35 |
| openstackgerrit | Steven Dake proposed stackforge/magnum: Remove redundant information https://review.openstack.org/148115 | 16:35 |
| hongbin | And let you know the result. Then we can discuss further. | 16:35 |
| sdake | our quickstart guide is looking pretty good | 16:35 |
| sdake | you may have to change em1 on non-fedora systems | 16:35 |
| sdake | i'm not sure if that will even work from inside a vm | 16:36 |
| sdake | I havdren't tried it | 16:36 |
| sdake | vm on vm is evil :) | 16:36 |
| hongbin | em1? | 16:36 |
| sdake | interface in devstack setup | 16:36 |
| hongbin | I agree. QEMU is too slow | 16:37 |
| *** oro has joined #openstack-containers | 16:38 | |
| *** hongbin has quit IRC | 16:39 | |
| sdake | jay-lau-513 are you around | 16:42 |
| *** EricGonczer_ has quit IRC | 16:54 | |
| *** hongbin has joined #openstack-containers | 17:02 | |
| *** diga has quit IRC | 17:04 | |
| *** dimsum__ has quit IRC | 17:06 | |
| *** dimsum__ has joined #openstack-containers | 17:20 | |
| *** oro has quit IRC | 17:34 | |
| *** Marga_ has joined #openstack-containers | 17:35 | |
| openstackgerrit | Steven Dake proposed stackforge/magnum: Magnum was automatically deleting stacks that failed to orchestrate https://review.openstack.org/148118 | 17:39 |
| *** oro has joined #openstack-containers | 17:47 | |
| openstackgerrit | Steven Dake proposed stackforge/magnum: Remove automatic deletion of failed stacks https://review.openstack.org/148118 | 17:47 |
| *** EricGonczer_ has joined #openstack-containers | 17:50 | |
| *** EricGonczer_ has quit IRC | 18:09 | |
| *** dimsum__ has quit IRC | 18:20 | |
| *** oro has quit IRC | 18:22 | |
| *** oro has joined #openstack-containers | 18:35 | |
| openstackgerrit | Steven Dake proposed stackforge/magnum: Delete bay independent of presence of heat stack https://review.openstack.org/148122 | 18:35 |
| *** ashishb has joined #openstack-containers | 18:40 | |
| *** ashishb has joined #openstack-containers | 18:41 | |
| *** fandi has quit IRC | 18:42 | |
| *** dimsum__ has joined #openstack-containers | 19:20 | |
| *** dimsum__ has quit IRC | 19:25 | |
| *** ashishb has quit IRC | 19:30 | |
| *** achanda has joined #openstack-containers | 19:37 | |
| hongbin | sdake: The localrc doesn't work for me | 20:04 |
| hongbin | Here is the error | 20:05 |
| hongbin | 2015-01-18 19:47:29.130 | + cd /opt/stack/heat 2015-01-18 19:47:29.130 | + git checkout stable/juno 2015-01-18 19:47:29.132 | error: pathspec 'stable/juno' did not match any file(s) known to git. | 20:05 |
| hongbin | I run devstack on Ubuntu14.04 | 20:05 |
| hongbin | K. The cause is that devstack use the --depth option to clone | 20:23 |
| hongbin | 2015-01-18 20:19:56.075 | + timeout -s SIGINT 0 git clone --depth 1 git://git.openstack.org/openstack/heat.git /opt/stack/heat | 20:23 |
| *** subscope_ has joined #openstack-containers | 20:32 | |
| *** oro has quit IRC | 20:34 | |
| *** achanda has quit IRC | 20:37 | |
| *** oro has joined #openstack-containers | 20:51 | |
| *** oro has quit IRC | 21:13 | |
| *** oro has joined #openstack-containers | 21:29 | |
| *** subscope_ has quit IRC | 22:23 | |
| openstackgerrit | Hongbin Lu proposed stackforge/magnum: Make the quickstart guide to work https://review.openstack.org/148133 | 22:36 |
| sdake | hongbin odd | 22:49 |
| hongbin | It seems it is due to a latest change in devstack | 22:51 |
| hongbin | I have proposed a fix to devstack https://review.openstack.org/#/c/148130/ | 22:51 |
| sdake | interesting | 22:54 |
| sdake | I have a bit older version of devstack generally once I get devstack working I dont change it ;-) | 22:54 |
| sdake | were you able to start a bay and pod etc? | 22:54 |
| hongbin | Bay is fine. | 22:57 |
| hongbin | Pod no | 22:57 |
| hongbin | I am figuring it out | 22:58 |
| sdake | can you ssh into your vm? | 22:59 |
| sdake | I will help you debug a bit if you liek | 22:59 |
| sdake | you may find this help ful sudo iptables --table nat -A POSTROUTING -o em1 -j MASQUERADE | 23:00 |
| sdake | where em1 is your network interface | 23:00 |
| *** dimsum__ has joined #openstack-containers | 23:01 | |
| sdake | first step is to see if you can ssh into your vm | 23:01 |
| sdake | next step is to see if you can get data into your vm from outside your network | 23:01 |
| sdake | if those two things fail, I'm not sure why pod creation wouldn't work | 23:01 |
| sdake | wfm on master | 23:01 |
| hongbin | 1. I can ssh to the VM | 23:04 |
| hongbin | 2. Inside the VM, it seems I cannot get data from outside. I tried to ping google.ca. It didn't work. | 23:05 |
| sdake | try ping 8.8.8.8 | 23:05 |
| sdake | (ips rather then dns) | 23:05 |
| sdake | http://docs.openstack.org/developer/devstack/guides/neutron.html | 23:06 |
| hongbin | In the master VM, no. I cannot ping it | 23:06 |
| sdake | read the part about firewalls (I added that) :) | 23:07 |
| sdake | ok, if you cannot ping at all then your masquarade rule is not set | 23:07 |
| sdake | recommend setting it above - it will forward packets all through em1 | 23:07 |
| sdake | (or whatever your initial interface is named) | 23:07 |
| hongbin | k. I will try that. | 23:07 |
| hongbin | thx | 23:07 |
| sdake | did you have devstack working previously? | 23:08 |
| *** dimsum__ has quit IRC | 23:08 | |
| hongbin | you mean if I have worked with devstack, or have a working devstack already? | 23:08 |
| sdake | working devstack setup with neutron | 23:08 |
| hongbin | yes, no in other branch | 23:09 |
| hongbin | this is the first time I stack on master | 23:09 |
| hongbin | s/no/but/ | 23:09 |
| *** yuanying has joined #openstack-containers | 23:14 | |
| *** dimsum__ has joined #openstack-containers | 23:16 | |
| sdake | yuanying around? | 23:17 |
| yuanying | morning, sdake | 23:17 |
| sdake | morning fine sir | 23:17 |
| sdake | woudl you mind tackling the review queue this morning? | 23:17 |
| sdake | I think we are all set for milestone #1 ;-) | 23:17 |
| yuanying | Now I'm reviewing. | 23:17 |
| sdake | I can't find anything seriously wrong at this point | 23:17 |
| sdake | I added a slew of blueprints for milestone #2 as well | 23:18 |
| sdake | if you have any to suggest feel free to write em down ;) | 23:18 |
| yuanying | ok | 23:18 |
| yuanying | Now, I'm reviewing this patch. https://review.openstack.org/#/c/148118 | 23:19 |
| yuanying | I want to remain faliled stack because I want to why that stack is failed. | 23:19 |
| yuanying | If stack is failed, I can't find it. | 23:20 |
| sdake | the code as is deletes the stack | 23:20 |
| jay-lau-513 | sdake just come here | 23:20 |
| sdake | see the bug report for more details | 23:20 |
| sdake | jay-lau-513 if you have more blueprints add em in pls | 23:21 |
| sdake | the more we have the better it looks like we actually have a roadmap ;-) | 23:21 |
| jay-lau-513 | sdake you mean the network and scheduler? | 23:21 |
| sdake | those make sense as blueprints | 23:22 |
| sdake | i've been kind of ignoring the container-only use case | 23:22 |
| sdake | and assuming people would just use k8s | 23:22 |
| sdake | but that may be incorrect thinking | 23:22 |
| jay-lau-513 | ok, i see, I will file related bp for review | 23:23 |
| hongbin | sdake: the iptable command works. Now the VM can ping outside | 23:23 |
| sdake | nice hongbin | 23:24 |
| sdake | winning ;) | 23:24 |
| sdake | btw, figuring out the iptables and firewall thing took me literally 150 hours or more | 23:24 |
| sdake | of jerking around with it | 23:24 |
| hongbin | haha | 23:24 |
| hongbin | so I saved 150 hours | 23:24 |
| hongbin | :) | 23:24 |
| hongbin | Now another problem on pod deleting. Maybe a review in queue will solve it | 23:25 |
| hongbin | 2015-01-18 23:22:50.664 4493 DEBUG oslo_concurrency.processutils [-] u'kubectl delete pod mariadb -s http://192.168.1.228:8080' failed. Not Retrying. execute /usr/local/lib/python2.7/dist-packages/oslo_concurrency/processutils.py:249 | 23:25 |
| hongbin | 2015-01-18 23:22:50.665 4493 ERROR magnum.conductor.handlers.common.kube_utils [-] Couldn't delete pod mariadb due to error [Errno 2] No such file or directory | 23:25 |
| sdake | hmm I ddin't notice that | 23:26 |
| sdake | but i'm not sure I tested that or not | 23:26 |
| sdake | we really need functional tests | 23:26 |
| sdake | so we dont hav eto manually test all this stuff constantly | 23:27 |
| hongbin | agree | 23:27 |
| hongbin | so you are sure this is a bug? | 23:27 |
| hongbin | or it is due to my setting? | 23:27 |
| sdake | i don't see how it could be due to your environment | 23:27 |
| sdake | it looks like a bug in kubernetes to me | 23:27 |
| sdake | but I'm not certain | 23:28 |
| sdake | if I could get my pod creation going I could tell you | 23:28 |
| hongbin | k | 23:28 |
| sdake | but my network is busted now | 23:28 |
| hongbin | k | 23:28 |
| hongbin | Just remind me to file a bug report if it is actually a bug | 23:29 |
| hongbin | or maybe I can figure it out later | 23:29 |
| *** dimsum__ has quit IRC | 23:29 | |
| hongbin | anyway. I will leave for while | 23:29 |
| hongbin | thanks for your help | 23:30 |
| sdake | http://paste.fedoraproject.org/171222/21623825 | 23:30 |
| sdake | here is my pod create | 23:30 |
| sdake | it appears busted | 23:30 |
| *** achanda has joined #openstack-containers | 23:30 | |
| *** hongbin has quit IRC | 23:31 | |
| openstackgerrit | Jay Lau proposed stackforge/magnum: Add more tests for test_objects.py https://review.openstack.org/148113 | 23:31 |
| *** oro has quit IRC | 23:32 | |
| sdake | nm, I had old client | 23:33 |
| sdake | going forward, we need to not break the clients | 23:33 |
| sdake | so if there is anything anyone wants to add, now is the time to do it :) | 23:33 |
| sdake | as in backwards compatibility breaking | 23:33 |
| sdake | also the dev guide needs update for the new --manifest flag | 23:33 |
| sdake | hongbin I can confirm you have an environmental problem - pod deletion works for me | 23:35 |
| openstackgerrit | Steven Dake proposed stackforge/magnum: Rename "ackend" to "conductor" https://review.openstack.org/148135 | 23:43 |
| *** yuanying has quit IRC | 23:46 | |
| *** yuanying has joined #openstack-containers | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!