*** EricGonczer_ has quit IRC | 00:04 | |
*** Marga_ has joined #openstack-containers | 00:14 | |
*** suro-patz1 has quit IRC | 00:15 | |
*** adrian_otto has quit IRC | 00:29 | |
openstackgerrit | Merged openstack/magnum: Update pod manifest parsing according to v1beta3 https://review.openstack.org/178079 | 00:32 |
---|---|---|
*** Tango has quit IRC | 00:46 | |
*** suro-patz has joined #openstack-containers | 01:02 | |
*** suro-patz1 has joined #openstack-containers | 01:05 | |
*** suro-patz has quit IRC | 01:08 | |
*** fangfenghua has joined #openstack-containers | 01:12 | |
*** Marga_ has quit IRC | 01:37 | |
*** Marga_ has joined #openstack-containers | 01:37 | |
*** Marga_ has quit IRC | 01:42 | |
*** EricGonczer_ has joined #openstack-containers | 01:43 | |
*** erkules_ has joined #openstack-containers | 01:45 | |
*** erkules has quit IRC | 01:48 | |
openstackgerrit | Andrew Melton proposed openstack/magnum: Make Docker client timeout configurable https://review.openstack.org/179257 | 01:48 |
*** vilobhmm1 has quit IRC | 01:48 | |
*** EricGonczer_ has quit IRC | 02:03 | |
*** kebray has joined #openstack-containers | 02:16 | |
*** unicell has quit IRC | 02:33 | |
*** kebray has quit IRC | 02:35 | |
openstackgerrit | Digambar proposed openstack/magnum: Remove cluster_coe from magnum conf. https://review.openstack.org/179323 | 02:42 |
*** EricGonczer_ has joined #openstack-containers | 03:04 | |
*** fangfeng_ has joined #openstack-containers | 03:05 | |
*** EricGonczer_ has quit IRC | 03:09 | |
*** achanda has quit IRC | 03:10 | |
*** unicell has joined #openstack-containers | 03:11 | |
*** unicell has quit IRC | 03:15 | |
*** unicell has joined #openstack-containers | 03:16 | |
*** kebray has joined #openstack-containers | 03:17 | |
*** suro-patz1 has quit IRC | 03:19 | |
*** fangfenghua has quit IRC | 03:23 | |
*** kebray has quit IRC | 03:29 | |
*** dims has quit IRC | 03:38 | |
*** vilobhmm1 has joined #openstack-containers | 03:39 | |
*** chuenlye has joined #openstack-containers | 03:55 | |
*** achanda has joined #openstack-containers | 04:10 | |
*** achanda has quit IRC | 04:15 | |
*** adrian_otto has joined #openstack-containers | 04:25 | |
*** achanda has joined #openstack-containers | 04:30 | |
*** dims has joined #openstack-containers | 04:39 | |
*** dims has quit IRC | 04:45 | |
*** vilobhmm1 has quit IRC | 04:47 | |
*** chuenlye has quit IRC | 04:51 | |
*** suro-patz has joined #openstack-containers | 05:47 | |
*** suro-patz has quit IRC | 05:48 | |
*** chuenlye has joined #openstack-containers | 05:52 | |
*** chuenlye has quit IRC | 05:56 | |
*** suro-patz has joined #openstack-containers | 05:57 | |
*** chuenlye has joined #openstack-containers | 05:58 | |
*** chandankumar has joined #openstack-containers | 06:36 | |
*** zul has joined #openstack-containers | 06:44 | |
*** suro-patz has quit IRC | 06:56 | |
*** achanda has quit IRC | 06:58 | |
*** achanda has joined #openstack-containers | 07:08 | |
*** achanda has quit IRC | 07:23 | |
*** zul has quit IRC | 07:25 | |
*** zul has joined #openstack-containers | 07:25 | |
*** oro has joined #openstack-containers | 07:34 | |
*** zul has quit IRC | 08:02 | |
*** zul has joined #openstack-containers | 08:09 | |
*** zul has quit IRC | 08:31 | |
*** zul has joined #openstack-containers | 09:07 | |
*** adrian_otto has quit IRC | 09:34 | |
*** sdake_ has joined #openstack-containers | 09:55 | |
*** sdake_ has quit IRC | 09:58 | |
*** sdake has quit IRC | 09:59 | |
*** chandankumar has quit IRC | 10:09 | |
*** zul has quit IRC | 10:44 | |
*** chuenlye has quit IRC | 10:56 | |
*** chuenlye has joined #openstack-containers | 10:56 | |
*** chuenlye has quit IRC | 10:56 | |
*** kbyrne has joined #openstack-containers | 11:10 | |
*** fangfeng_ has quit IRC | 11:21 | |
*** fangfenghua has joined #openstack-containers | 11:44 | |
*** EricGonczer_ has joined #openstack-containers | 12:03 | |
*** ashishb has joined #openstack-containers | 12:10 | |
*** jsotoca has joined #openstack-containers | 12:19 | |
*** zul has joined #openstack-containers | 12:30 | |
*** zul has quit IRC | 12:35 | |
*** EricGonczer_ has quit IRC | 12:45 | |
*** dboik_ has quit IRC | 12:59 | |
*** madhuri_ has joined #openstack-containers | 13:10 | |
*** rlrevell has joined #openstack-containers | 13:11 | |
*** unicell1 has joined #openstack-containers | 13:14 | |
*** rpothier has joined #openstack-containers | 13:16 | |
*** unicell has quit IRC | 13:16 | |
*** dboik has joined #openstack-containers | 13:18 | |
*** dboik has quit IRC | 13:18 | |
*** ashishb has quit IRC | 13:18 | |
*** dboik has joined #openstack-containers | 13:19 | |
*** sdake has joined #openstack-containers | 13:24 | |
*** sdake has quit IRC | 13:29 | |
*** dims has joined #openstack-containers | 13:30 | |
*** ashishb has joined #openstack-containers | 13:34 | |
*** openstackgerrit has quit IRC | 13:51 | |
*** openstackgerrit has joined #openstack-containers | 13:51 | |
*** dboik_ has joined #openstack-containers | 13:59 | |
*** dboik has quit IRC | 14:03 | |
*** madhuri_ has quit IRC | 14:03 | |
*** oro has quit IRC | 14:12 | |
*** oro has joined #openstack-containers | 14:15 | |
*** madhuri_ has joined #openstack-containers | 14:19 | |
*** oro has quit IRC | 14:31 | |
*** oro has joined #openstack-containers | 14:33 | |
*** achanda has joined #openstack-containers | 14:37 | |
*** madhuri_ has quit IRC | 14:38 | |
*** achanda has quit IRC | 14:43 | |
*** EricGonczer_ has joined #openstack-containers | 14:48 | |
*** adrian_otto has joined #openstack-containers | 14:52 | |
*** dims is now known as dimsum__ | 14:58 | |
*** ashishb has quit IRC | 14:59 | |
*** sdake has joined #openstack-containers | 15:01 | |
*** dboik_ has quit IRC | 15:05 | |
*** dboik has joined #openstack-containers | 15:05 | |
*** EricGonc_ has joined #openstack-containers | 15:19 | |
*** Tango has joined #openstack-containers | 15:19 | |
*** EricGonczer_ has quit IRC | 15:23 | |
*** ashishb has joined #openstack-containers | 15:23 | |
Tango | Hi everyone, is there instruction to rebuild the fedora-21-atomic with the latest Kubernetes service binaries? I need to pick up a recent bug fix in Kubernetes. | 15:30 |
*** dimsum__ has quit IRC | 15:31 | |
*** dimsum__ has joined #openstack-containers | 15:31 | |
*** ashishb has quit IRC | 15:34 | |
tcammann | Hello, I was looking at doing some API request validation in magnum; do we think names for bays, baymodels and containers should be mandatory? | 15:35 |
*** fangfenghua_ has joined #openstack-containers | 15:48 | |
*** prad has joined #openstack-containers | 15:50 | |
Tango | tcammann: What's the common practice for naming resources in other services like Neutron? | 15:54 |
dimsum__ | tcammann: there's some guidelines for naming https://github.com/openstack/api-wg/tree/master/guidelines though it does not say if it should be mandatory or now | 15:57 |
dimsum__ | s/now/not/ | 15:57 |
tcammann | Thanks dimsum__, Tango. | 16:03 |
tcammann | From my experience, if it has a name field its usually mandatory, e.g. nova instances. I'm sure there are exceptions though. | 16:05 |
tcammann | The fields I was going to apply this to were the name fields for Bays, BayModels and Containers. | 16:06 |
tcammann | imho they should have mandatory names | 16:06 |
dimsum__ | tcammann: +1 from me | 16:07 |
openstackgerrit | Tom Cammann proposed openstack/magnum: Add string length validation to names in APIs https://review.openstack.org/179416 | 16:09 |
*** Marga_ has joined #openstack-containers | 16:09 | |
tcammann | feel free to disagree in the review :) ^^ | 16:09 |
openstackgerrit | Tom Cammann proposed openstack/magnum: Add string length validation to names in APIs https://review.openstack.org/179416 | 16:15 |
*** unicell1 has quit IRC | 16:16 | |
*** Marga_ has quit IRC | 16:22 | |
*** Marga_ has joined #openstack-containers | 16:22 | |
*** unicell has joined #openstack-containers | 16:41 | |
tcammann | adrian_otto: irt your comment on the patch above - Bay models are going to be 'pets' which I think | 16:49 |
tcammann | deserve real names, perhaps containers and bays are more like 'cattle' and don't deserve a real name. | 16:50 |
tcammann | In nova all instances have to have a name and I was following that model. | 16:50 |
*** vilobhmm1 has joined #openstack-containers | 16:59 | |
*** vilobhmm1 has quit IRC | 16:59 | |
*** vilobhmm1 has joined #openstack-containers | 17:00 | |
*** chandankumar has joined #openstack-containers | 17:02 | |
*** suro-patz has joined #openstack-containers | 17:05 | |
openstackgerrit | Tom Cammann proposed openstack/magnum: Add string length validation to names in APIs https://review.openstack.org/179416 | 17:06 |
tcammann | adrian_otto: But you are right its beyond the scope of the bug, I have removed that mandory check. | 17:06 |
tcammann | s/mandory/mandatory/ | 17:06 |
*** Marga_ has quit IRC | 17:07 | |
*** achanda has joined #openstack-containers | 17:07 | |
*** Marga_ has joined #openstack-containers | 17:07 | |
adrian_otto | thanks tcammann | 17:11 |
adrian_otto | I appreciate your explanation. I'll vote on your revision now. | 17:12 |
*** Marga_ has quit IRC | 17:13 | |
*** Marga_ has joined #openstack-containers | 17:13 | |
openstackgerrit | Digambar proposed openstack/magnum: Remove cluster_coe from magnum conf https://review.openstack.org/179426 | 17:14 |
*** kebray has joined #openstack-containers | 17:16 | |
*** kebray has quit IRC | 17:16 | |
Tango | adrian_otto: Hi Adrian, did you build the fedora images in https://fedorapeople.org/groups/magnum/ ? | 17:17 |
adrian_otto | Tango, I think sdake made them initially, and apmelton may have adapted them to some extent, why? | 17:18 |
Tango | I guess the kube* binaries are built into the image, right? | 17:19 |
adrian_otto | yes, I think so | 17:19 |
Tango | adrian_otto: I found a bug in Kubernetes code in handling the load balancer, which has been fixed. So I would like to build a new image with the latest code to pick up the fix | 17:20 |
*** subscope has joined #openstack-containers | 17:21 | |
Tango | adrian_otto: I have been copying the binaries over manually, but it's getting cumbersome to fiddle with systemctl | 17:21 |
adrian_otto | copying them to what, you dev environment? | 17:22 |
adrian_otto | s/you/your/ | 17:22 |
Tango | Copying the new kubernetes binaries to the bay VM's, like kube-apiserver, kube-controller-manager | 17:23 |
Tango | then restarting the services | 17:24 |
adrian_otto | for upgrading, or setting up fresh? | 17:24 |
Tango | The new image would be for setting up fresh bay | 17:24 |
*** kebray has joined #openstack-containers | 17:25 | |
adrian_otto | could you elaborate a bit more, so I can get a clear picture… maybe I could offer a bit more help? | 17:26 |
Tango | Sure, this is in reference to the blueprint for creating an external load balancer with a Kubernetes service | 17:27 |
Tango | The way this works is that when Kubernetes detects the flag in the service yaml, it will contact Neutron to create all the resources for the load balancer, like pool, member, VIP, ... | 17:28 |
Tango | When I was prototyping this to implement our blueprint, I found a bug in the Kubernetes code: the API call to Neutron was incorrect | 17:28 |
Tango | I reported the bug, which has been fixed | 17:29 |
adrian_otto | oh, that's terrific! | 17:29 |
adrian_otto | so then you needed to update the k8s binaries to take advantage of the fix? | 17:29 |
Tango | This fix gets picked up in the latest Kubernetes version 0.16.0 | 17:29 |
Tango | right. I have been working around this by creating a bay first with the current fedora-21-atomic-2 image, which has Kubernetes version 0.11.0 | 17:30 |
Tango | Then manually replace the Kubernetes services with the latest version | 17:31 |
Tango | So now to run end to end without this manual update, I would like to build a new fedora image to get the fix | 17:32 |
adrian_otto | right. | 17:33 |
adrian_otto | I do not know our process for updating those images yet. There was something about this on the ML recently IIRC. | 17:35 |
Tango | oh ok, I can take a look there. | 17:36 |
adrian_otto | I'm sure that sdake knows the update procedure, so we can start with him. | 17:38 |
*** suro-patz has quit IRC | 17:39 | |
*** suro-patz has joined #openstack-containers | 17:39 | |
Tango | I will try to ping sdake also. While at it, maybe I will document the procedure, if it's not already written up | 17:39 |
sdake | we have 0.15 in master tango | 17:40 |
sdake | tango what is the minimum version you need? | 17:42 |
sdake | tango if you want t obuild master, you will need to create an rpm | 17:43 |
sdake | version 0.16 is the latest | 17:43 |
Tango | sdake: The fix was just commited Thursday last week. The new version 0.16 will pick it up and is coming out, I saw a pre-release the day before | 17:43 |
sdake | 0.16 is out | 17:43 |
sdake | http://koji.fedoraproject.org/koji/buildinfo?buildID=632390 | 17:44 |
Tango | sdake: ok great. | 17:44 |
sdake | i can walk you through the image creation process | 17:44 |
sdake | but i dont have time atm to actually create the images | 17:44 |
Tango | sure, you can give some quick tips and I will try to build | 17:45 |
sdake | lets go step by step its a bit involved | 17:45 |
Tango | Did you use diskimage builder? | 17:45 |
sdake | first make a fedora account on accounts.fedoraproject.org i think | 17:45 |
sdake | there is this thing called copr | 17:46 |
sdake | it lets you rebuild a src rpm into a buidlroot based on any distro of fedora | 17:46 |
sdake | for example, we will build against f21 using the f23 source rpm here https://kojipkgs.fedoraproject.org//packages/kubernetes/0.16.0/2.fc23/src/kubernetes-0.16.0-2.fc23.src.rpm | 17:46 |
sdake | its a simple paste field, very simple to build with copr | 17:47 |
sdake | ping me when you have a copr repo done | 17:47 |
Tango | ok, creating an account now | 17:47 |
sdake | you may have to sign a cla i don't know what the current process is | 17:49 |
*** adrian_otto has quit IRC | 17:50 | |
sdake | tango hopefully you can add this to the documentation when you get done | 17:51 |
sdake | please add me as coauthor on the commit since I figured all htis bs out :) | 17:51 |
Tango | sdake: Will do, this is very useful. | 17:52 |
Tango | sdake: You will be the primary author, I only test and write it down :) | 17:52 |
sdake | you have to used Co-Authored-By line in the commit for me to get stacklaytics credit | 17:53 |
sdake | and for whatever inane reason people look at that for bonuses etc | 17:53 |
Tango | ok, that's good to know | 17:53 |
dimsum__ | sdake: functional test patch is ready when you get a chance - https://review.openstack.org/#/c/173537/ - it has that weird "CREATE_COMPLETE" i mentioned to you yesterday (bay's status) | 17:55 |
sdake | i was out yesterday | 17:56 |
sdake | whatis the gate runtime? | 17:56 |
sdake | 44min? | 17:56 |
dimsum__ | average, yes | 17:56 |
sdake | why is the large keystone timeout required | 17:59 |
sdake | just curious - doesn't block review | 17:59 |
dimsum__ | sdake: i'll tune it down after watching the runs for a few days. i was not sure what to start with, hence the large timeout. the CI job timeout is 2 hours too, so i have to tune a few places down after we get some data. | 18:01 |
sdake | i dont like the bay_object usage in the test cases | 18:01 |
sdake | isn't there a way to get that from the rest api? | 18:01 |
*** adrian_otto has joined #openstack-containers | 18:03 | |
sdake | ideally we shouldn't import from the magnum namespace right/ | 18:04 |
sdake | dimsum__ | 18:04 |
dimsum__ | yes, i can fix that | 18:05 |
sdake | cool i'll mark it as -1 then and add a comment | 18:05 |
sdake | if there is no othe way to do it, then it is what it is | 18:05 |
sdake | but it sounds like we can use the rest interface for that | 18:05 |
dimsum__ | i'll just use regular strings | 18:05 |
dimsum__ | and drop the import | 18:06 |
*** fangfenghua has quit IRC | 18:06 | |
*** fangfenghua_ is now known as fangfenghua | 18:06 | |
sdake | enjoy | 18:08 |
*** sdake_ has joined #openstack-containers | 18:09 | |
openstackgerrit | Davanum Srinivas (dims) proposed openstack/magnum: Functional tests for listing resources and templates https://review.openstack.org/173537 | 18:10 |
dimsum__ | thanks sdake. done ^^^ | 18:10 |
*** sdake has quit IRC | 18:13 | |
*** sdake has joined #openstack-containers | 18:15 | |
*** sdake_ has quit IRC | 18:19 | |
*** kebray has quit IRC | 18:22 | |
*** Marga_ has quit IRC | 18:31 | |
*** Marga_ has joined #openstack-containers | 18:31 | |
*** Marga_ has quit IRC | 18:38 | |
*** Marga_ has joined #openstack-containers | 18:38 | |
*** daneyon has joined #openstack-containers | 18:48 | |
*** daneyon has quit IRC | 18:49 | |
*** daneyon has joined #openstack-containers | 18:49 | |
*** achanda has quit IRC | 19:01 | |
sdake | tango any luck? | 19:02 |
*** sdake_ has joined #openstack-containers | 19:04 | |
*** subscope has quit IRC | 19:04 | |
Tango | sdake: reading up on copr | 19:05 |
*** sdake__ has joined #openstack-containers | 19:05 | |
*** sdake has quit IRC | 19:05 | |
*** daneyon_ has joined #openstack-containers | 19:08 | |
*** Marga_ has quit IRC | 19:09 | |
*** Marga_ has joined #openstack-containers | 19:09 | |
*** sdake_ has quit IRC | 19:09 | |
*** daneyon has quit IRC | 19:11 | |
*** ashishb has joined #openstack-containers | 19:13 | |
Tango | In the page to create a project, I guess I would select fedora-21-x86_64 under Chroots? | 19:18 |
Tango | How about the epel-6, epel-7 ? | 19:19 |
*** Marga_ has quit IRC | 19:22 | |
*** Marga_ has joined #openstack-containers | 19:23 | |
*** Marga_ has quit IRC | 19:30 | |
*** Marga_ has joined #openstack-containers | 19:30 | |
*** chandankumar has quit IRC | 19:33 | |
*** ashishb has quit IRC | 19:38 | |
*** daneyon_ has quit IRC | 19:41 | |
*** daneyon has joined #openstack-containers | 19:41 | |
*** achanda has joined #openstack-containers | 19:59 | |
*** erkules_ is now known as erkules | 20:07 | |
*** erkules has joined #openstack-containers | 20:07 | |
*** julim has quit IRC | 20:09 | |
*** suro-patz has quit IRC | 20:10 | |
*** suro-patz has joined #openstack-containers | 20:11 | |
*** suro-patz has quit IRC | 20:11 | |
*** suro-patz has joined #openstack-containers | 20:12 | |
*** fangfenghua has quit IRC | 20:13 | |
*** sdake__ is now known as sdake | 20:27 | |
sdake | tango just select fedora21 | 20:27 |
Tango | sdake: ok, I did that, and added the src.rpm for Kubernetes 0.16 | 20:28 |
sdake | after a bit it should build | 20:28 |
sdake | what is your fedora id | 20:29 |
sdake | tango? | 20:29 |
Tango | yep, it's tango. The build is pending now | 20:29 |
sdake | your build is pending | 20:29 |
sdake | yup | 20:29 |
sdake | ok next step get a machine with docker on it | 20:29 |
Tango | ok I have that ready | 20:30 |
*** suro-patz1 has joined #openstack-containers | 20:30 | |
sdake | git clone https://github.com/jasonbrooks/byo-atomic.git | 20:31 |
sdake | i'll spare you trying to dechipher the instructions and give them to you | 20:31 |
sdake | # docker build --rm -t $USER/atomicrepo byo-atomic/. | 20:32 |
sdake | # docker run --privileged -d -p 80:80 --name atomicrepo $USER/atomicrepo | 20:32 |
sdake | # docker exec -it atomicrepo bash | 20:32 |
sdake | so the first one should be docker build --rm -t tango/atomicrepo ./ | 20:32 |
sdake | if your in the byo-atomic directory | 20:32 |
*** suro-patz has quit IRC | 20:32 | |
sdake | also make sure httpd is not running on your system | 20:33 |
sdake | on your host | 20:33 |
sdake | after the docker exec, yo uwill be inside the container | 20:33 |
sdake | cd fedora-atomic | 20:33 |
sdake | git checkout f21 | 20:33 |
Tango | I don't need to wait for the fedora build for these steps? | 20:36 |
Tango | Doing the docker build step now | 20:44 |
*** rlrevell has quit IRC | 21:01 | |
*** chuenlye has joined #openstack-containers | 21:03 | |
*** suro-patz1 has quit IRC | 21:08 | |
sdake | tango no | 21:09 |
sdake | you don't need to wait | 21:09 |
Tango | sdake: The docker build step finishes. But I am getting an error on the docker run step | 21:10 |
sdake | which error | 21:10 |
Tango | FATA[0004] Error response from daemon: Cannot start container 6600eba2f43d5ed3f5399cf46375e451aecec1576cb202a084d529aba0bba742: Error starting userland proxy: listen tcp 0.0.0.0:80: bind: address already in use | 21:10 |
Tango | not sure what's on port 80 | 21:11 |
Tango | I don't think httpd is running | 21:11 |
*** suro-patz has joined #openstack-containers | 21:14 | |
sdake | it is clearly running ;) | 21:19 |
*** achanda has quit IRC | 21:19 | |
sdake | you can use netstat to find the process | 21:19 |
sdake | i don't recall the syntax, try asking in #kolla or ask larsks | 21:19 |
sdake | he pointed it out a few days ago | 21:19 |
*** achanda has joined #openstack-containers | 21:20 | |
sdake | 115 sudo netstat -antp | grep 5000 | 21:20 |
sdake | change 5000 to 80 | 21:20 |
sdake | tango ^ | 21:21 |
Tango | seems like apache2 is taking up port 80, trying to stop it | 21:22 |
*** oro has quit IRC | 21:23 | |
*** suro-patz has quit IRC | 21:31 | |
*** EricGonczer_ has joined #openstack-containers | 21:31 | |
Tango | I took down apache2 to free up port 80 | 21:32 |
*** EricGonczer_ has quit IRC | 21:33 | |
Tango | But the docker run step seems to exit immediately. | 21:33 |
Tango | checking with "sudo docker logs atomicrepo" | 21:34 |
Tango | I got: | 21:34 |
Tango | Passing arguments to httpd using apachectl is no longer supported. | 21:34 |
Tango | You can only start/stop/restart httpd using this script. | 21:34 |
Tango | If you want to pass extra arguments to httpd, edit the | 21:34 |
Tango | /etc/sysconfig/httpd config file. | 21:34 |
Tango | AH00558: httpd: Could not reliably determine the server's fully qualified domain name, using 172.17.0.1. Set the 'ServerName' directive globally to suppress this message | 21:34 |
Tango | (2)No such file or directory: AH02291: Cannot access directory '/etc/httpd/logs/' for main error log | 21:34 |
Tango | AH00014: Configuration check failed | 21:34 |
*** EricGonc_ has quit IRC | 21:34 | |
sdake | it doesn't exit immedateily | 21:37 |
sdake | it should take you int othe container in a shell script | 21:37 |
sdake | are you sure it exits immediately? | 21:37 |
Tango | When I try to connect into the container, it says: "Error response from daemon: Container atomicrepo is not running" | 21:38 |
Tango | (in the next step docker exec_ | 21:38 |
Tango | ) | 21:38 |
Tango | Then I check the logs | 21:38 |
sdake | in the byo-atomic directory | 21:38 |
sdake | there is a runa-apche.sh script | 21:38 |
sdake | can you paste it pls | 21:39 |
*** EricGonczer_ has joined #openstack-containers | 21:39 | |
Tango | #!/bin/bash | 21:39 |
Tango | 21:39 | |
Tango | # Make sure we're not confused by old, incompletely-shutdown httpd | 21:39 |
Tango | # context after restarting the container. httpd won't start correctly | 21:39 |
Tango | # if it thinks it is already running. | 21:39 |
Tango | 21:40 | |
Tango | rm -rf /run/httpd/* | 21:40 |
Tango | 21:40 | |
Tango | exec /usr/sbin/apachectl -D FOREGROUND | 21:40 |
sdake | change that last line to | 21:41 |
sdake | systemctl start httpd | 21:41 |
*** EricGonczer_ has quit IRC | 21:42 | |
sdake | rebuild yoru container | 21:42 |
sdake | no that wont work | 21:43 |
sdake | there is no systemd | 21:43 |
sdake | change last line to | 21:44 |
sdake | /usr/sbin/httpd -D FOREGROUND | 21:45 |
sdake | rebuild your container | 21:45 |
sdake | try again | 21:45 |
Tango | hmm, same result | 21:50 |
Tango | Maybe I can remove from Dockerfile the line to execute run-apache.sh, then attach to the container to run it manually | 21:55 |
*** dimsum__ has quit IRC | 21:56 | |
*** dimsum__ has joined #openstack-containers | 21:56 | |
*** dboik has quit IRC | 21:59 | |
sdake | same result? | 22:00 |
sdake | it said the stuff about apachectl is deprecated? | 22:00 |
sdake | there were a whoel bunch of errors in there, lets fix them one at a time | 22:01 |
sdake | tango yo ucould try running manually and then bg it | 22:01 |
Tango | something is strange, I modified the Dockerfile but it seems to have no effect | 22:01 |
sdake | why no effect? | 22:02 |
Tango | I took out the line to execute run-apache.sh, but the container still exits with the same error message | 22:02 |
Tango | scratching my head | 22:03 |
sdake | try docker rm | 22:06 |
sdake | sudo docker ps -a to find the container | 22:07 |
sdake | you are probably restarting the old container | 22:07 |
Tango | yep, I did rm and ps to make sure it's not there any more | 22:07 |
Tango | looks like it could be something in the stack/atomicrepo image | 22:09 |
*** EricGonczer_ has joined #openstack-containers | 22:11 | |
*** EricGonczer_ has quit IRC | 22:11 | |
sdake | use rmi and rebuild | 22:14 |
sdake | docker rmi will remove the image | 22:14 |
Tango | ok, let me try that | 22:14 |
*** suro-patz has joined #openstack-containers | 22:16 | |
*** Marga_ has quit IRC | 22:21 | |
*** Marga_ has joined #openstack-containers | 22:22 | |
*** suro-patz1 has joined #openstack-containers | 22:22 | |
*** suro-patz has quit IRC | 22:22 | |
*** suro-patz1 has quit IRC | 22:28 | |
*** Marga_ has quit IRC | 22:31 | |
*** Marga_ has joined #openstack-containers | 22:32 | |
Tango | The build step seems to have problem with the httpd package: Installing : httpd-2.4.10-15.fc21.x86_64 61/63Error unpacking rpm package httpd-2.4.10-15.fc21.x86_64 | 22:33 |
Tango | the rebuilt image behaves the same way | 22:34 |
*** Marga_ has quit IRC | 22:36 | |
sdake | out of disk space? | 22:36 |
sdake | /var/lib/ | 22:36 |
*** Marga_ has joined #openstack-containers | 22:36 | |
Tango | disk space should be OK, only 11% used | 22:37 |
sdake | what about /var? | 22:37 |
*** dboik has joined #openstack-containers | 22:37 | |
sdake | maybe mirror is corrupted | 22:38 |
sdake | try installing httpd from koji | 22:38 |
Tango | maybe I will just start with the plain fedora:21 image and run each step from Dockerfile manually | 22:38 |
*** dboik_ has joined #openstack-containers | 22:38 | |
sdake | khttp://koji.fedoraproject.org/koji/buildinfo?buildID=599665 | 22:38 |
sdake | try installing from that first | 22:39 |
sdake | sudo yum install -y httpd://koli* | 22:39 |
sdake | or whateve rthe link is for httpd | 22:39 |
sdake | https://kojipkgs.fedoraproject.org//packages/httpd/2.4.10/15.fc21/x86_64/httpd-2.4.10-15.fc21.x86_64.rpm | 22:39 |
*** Marga_ has quit IRC | 22:40 | |
*** suro-patz has joined #openstack-containers | 22:40 | |
*** dboik has quit IRC | 22:42 | |
*** Marga_ has joined #openstack-containers | 22:44 | |
*** Marga_ has quit IRC | 22:49 | |
*** Marga_ has joined #openstack-containers | 22:50 | |
*** Marga_ has quit IRC | 22:54 | |
*** prad has quit IRC | 22:57 | |
*** rpothier has quit IRC | 23:04 | |
*** chuenlye has quit IRC | 23:04 | |
*** Tango|2 has joined #openstack-containers | 23:06 | |
*** Marga_ has joined #openstack-containers | 23:08 | |
*** Tango has quit IRC | 23:08 | |
*** suro-patz has quit IRC | 23:12 | |
*** fangfenghua has joined #openstack-containers | 23:13 | |
*** achanda_ has joined #openstack-containers | 23:21 | |
*** chuenlye has joined #openstack-containers | 23:25 | |
*** achanda has quit IRC | 23:25 | |
*** achanda_ has quit IRC | 23:26 | |
*** chuenlye has quit IRC | 23:28 | |
*** chuenlye has joined #openstack-containers | 23:29 | |
*** sdake_ has joined #openstack-containers | 23:36 | |
*** sdake has quit IRC | 23:39 | |
*** sdake has joined #openstack-containers | 23:39 | |
*** sdake_ has quit IRC | 23:42 | |
*** chuenlye has quit IRC | 23:43 | |
*** fangfeng_ has joined #openstack-containers | 23:43 | |
sdake | adrian_otto good news, ported kolla to run kilo | 23:47 |
sdake | seems to work well | 23:47 |
sdake | now for a magnum container and we are good to go ;) | 23:48 |
*** suro-patz has joined #openstack-containers | 23:50 | |
*** suro-patz1 has joined #openstack-containers | 23:51 | |
*** fangfeng_ has quit IRC | 23:52 | |
*** fangfeng_ has joined #openstack-containers | 23:54 | |
*** suro-patz has quit IRC | 23:54 | |
*** fangfeng_ is now known as fangfenghua_ | 23:58 | |
fangfenghua_ | cool | 23:58 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!