*** yangyapeng has quit IRC | 00:02 | |
*** slagle has quit IRC | 00:20 | |
*** spsurya has joined #openstack-kolla | 00:21 | |
*** sbezverk has joined #openstack-kolla | 00:26 | |
*** manheim has quit IRC | 00:32 | |
*** jascott1 has joined #openstack-kolla | 00:32 | |
*** jascott1 has quit IRC | 00:36 | |
*** StephenWang1991 has joined #openstack-kolla | 00:39 | |
*** chas has joined #openstack-kolla | 00:41 | |
*** StephenWang1991 has quit IRC | 00:44 | |
spsurya | morning all | 00:45 |
---|---|---|
*** chas has quit IRC | 00:46 | |
*** cuongnv has joined #openstack-kolla | 00:47 | |
*** hieulq has joined #openstack-kolla | 00:47 | |
*** tovin07 has joined #openstack-kolla | 00:47 | |
*** trozet has quit IRC | 00:50 | |
sdake | sup spsurya | 00:56 |
openstackgerrit | Steven Dake proposed openstack/kolla-ansible master: DNM: Test for a deploy of kolla-kubernetes https://review.openstack.org/459137 | 00:57 |
spsurya | sdake: wanted some info on your patch https://review.openstack.org/#/c/452854 | 00:59 |
spsurya | same for this one too https://review.openstack.org/#/c/452855/ | 00:59 |
spsurya | one small item too https://review.openstack.org/#/c/457870/ | 01:00 |
sdake | spsurya ok - will get to it in a moment - eating shorlty | 01:04 |
spsurya | sdake: np | 01:05 |
openstackgerrit | Steven Dake proposed openstack/kolla-kubernetes master: DNM: Test out insecure setting https://review.openstack.org/459138 | 01:05 |
*** yangyapeng has joined #openstack-kolla | 01:11 | |
spsurya | Jeffrey4l: sup https://review.openstack.org/#/c/455866 | 01:12 |
*** duonghq has joined #openstack-kolla | 01:22 | |
duonghq | morning guys | 01:22 |
Jeffrey4l | spsurya, when not use https://docs.openstack.org/developer/kolla-ansible/vagrant-dev-env.html ? | 01:23 |
Jeffrey4l | i guess we should avoid to use github.com. | 01:23 |
spsurya | Jeffrey4l: yes we should, but that moment link was not working | 01:24 |
spsurya | I will the update the | 01:24 |
spsurya | same | 01:24 |
Jeffrey4l | the original link is wrong. try mine. | 01:24 |
*** zhubingbing has joined #openstack-kolla | 01:24 | |
openstackgerrit | Surya Prakash (spsurya) proposed openstack/kolla-ansible master: Development Environment With Vagrant link not working https://review.openstack.org/455866 | 01:26 |
spsurya | Jeffrey4l: corrected | 01:26 |
Jeffrey4l | thanks. +2ed | 01:28 |
spsurya | Jeffrey4l: thanks | 01:29 |
*** hrw has quit IRC | 01:30 | |
*** hrw has joined #openstack-kolla | 01:32 | |
*** iceyao has joined #openstack-kolla | 01:36 | |
*** Seyong has joined #openstack-kolla | 01:36 | |
openstackgerrit | Merged openstack/kolla-ansible master: Development Environment With Vagrant link not working https://review.openstack.org/455866 | 01:39 |
*** zhenguo has joined #openstack-kolla | 01:48 | |
*** Pavo has quit IRC | 01:49 | |
spsurya | sdake not getting running $kubectl get pods --all-namespaces | grep dns | 01:55 |
spsurya | https://www.irccloud.com/pastebin/gzG0ohU8/ | 01:55 |
*** Pavo has joined #openstack-kolla | 01:57 | |
*** schwicht has quit IRC | 02:01 | |
*** Pavo has quit IRC | 02:03 | |
*** jascott1 has joined #openstack-kolla | 02:05 | |
spsurya | Apr 24 01:42:25 localhost journal: E0424 01:42:25.325310 1 leaderelection.go:229] error retrieving resource lock kube-system/kube-controller-manager: Get https://192.168.121.93:6443/api/v1/namespaces/kube-system/endpoints/kube-controller-manager: dial tcp 192.168.121.93:6443: getsockopt: connection refused | 02:10 |
spsurya | these are thge message log i can see the error | 02:11 |
*** jascott1 has quit IRC | 02:11 | |
spsurya | sbezverk: what can be the reason of failing this ? | 02:12 |
*** zhuzeyu has joined #openstack-kolla | 02:31 | |
*** StephenWang1991 has joined #openstack-kolla | 02:38 | |
*** chas has joined #openstack-kolla | 02:42 | |
*** daidv has joined #openstack-kolla | 02:42 | |
*** rwsu has joined #openstack-kolla | 02:45 | |
sbezverk | spsurya: netstat -tunlp, check if 6443 port is opened | 02:46 |
*** chas has quit IRC | 02:46 | |
sbezverk | if it is not then api-server is not running, if it is then make sure you copied /etc/kubernetes/admin.conf to $HOME/.kube/conifg | 02:47 |
zhuzeyu | hi, sbezverk | 02:50 |
zhuzeyu | i have a question want to ask | 02:50 |
spsurya | https://www.irccloud.com/pastebin/pnn7XYEr/ | 02:51 |
zhuzeyu | when i deploy kolla-k8s, how can i use local registry to pull images, I use the methods same as kolla-ansible used, but it seem that doesn't work | 02:52 |
zhuzeyu | thanks | 02:52 |
daidv | Morning, can anyone review for this backport: https://review.openstack.org/#/c/456513/ | 02:56 |
sbezverk | spsurya: port is opened which is good | 02:58 |
sbezverk | then copy file I mentioned before | 02:58 |
sbezverk | chmod +r $HOME/.kube/conifg | 02:58 |
spsurya | sbezverk: yes | 02:58 |
sbezverk | and try again | 02:59 |
spsurya | sbezverk: i see | 02:59 |
sbezverk | zhuzeyu: you need to change your cloud yaml to point to your local registery | 02:59 |
sbezverk | I am off for now, ttyl | 03:01 |
zhuzeyu | thanks, let me have a try | 03:02 |
*** salv-orlando has joined #openstack-kolla | 03:07 | |
*** salv-orlando has quit IRC | 03:12 | |
*** ssurana has joined #openstack-kolla | 03:14 | |
openstackgerrit | zhubingbing proposed openstack/kolla master: Fix tempest dockerfile https://review.openstack.org/459152 | 03:18 |
*** KImj has joined #openstack-kolla | 03:20 | |
*** KImj has quit IRC | 03:25 | |
*** MarginHu has joined #openstack-kolla | 03:26 | |
*** tovin07_ has joined #openstack-kolla | 03:27 | |
*** janonymous has joined #openstack-kolla | 03:29 | |
openstackgerrit | shaofeng cheng proposed openstack/kolla-ansible master: Add OS_REGION_NAME to admin-openrc.sh.j2 https://review.openstack.org/459155 | 03:32 |
*** StephenWang1991 has quit IRC | 03:33 | |
*** goldyfruit has quit IRC | 03:36 | |
*** dave-mccowan has quit IRC | 03:37 | |
*** gkadam has joined #openstack-kolla | 03:46 | |
*** thomasem_ has quit IRC | 04:04 | |
*** salv-orlando has joined #openstack-kolla | 04:08 | |
*** jascott1 has joined #openstack-kolla | 04:08 | |
*** salv-orlando has quit IRC | 04:13 | |
*** jascott1 has quit IRC | 04:13 | |
*** myoung has joined #openstack-kolla | 04:19 | |
*** iceyao has quit IRC | 04:32 | |
*** jayahn has joined #openstack-kolla | 04:36 | |
*** tervancovan has joined #openstack-kolla | 04:38 | |
tervancovan | Hi all | 04:38 |
tervancovan | I found that reason why verification of k8s didn't working in my company | 04:40 |
tervancovan | port 53 for these ips - >8.8.8.8, 8.8.4.4 was blocked | 04:40 |
*** StephenWang1991 has joined #openstack-kolla | 04:41 | |
tervancovan | sdake so "nslookup kubernetes" did not run | 04:42 |
*** chas has joined #openstack-kolla | 04:43 | |
openstackgerrit | hosup choi proposed openstack/kolla-kubernetes master: Update deployment-guide for ubuntu user https://review.openstack.org/459163 | 04:44 |
*** chas has quit IRC | 04:47 | |
*** salv-orlando has joined #openstack-kolla | 04:57 | |
*** thomasem_ has joined #openstack-kolla | 05:00 | |
*** eanylin has quit IRC | 05:02 | |
*** skramaja has joined #openstack-kolla | 05:16 | |
*** Guest31033 has joined #openstack-kolla | 05:19 | |
*** StephenW_ has joined #openstack-kolla | 05:29 | |
*** StephenWang1991 has quit IRC | 05:29 | |
*** daidv has quit IRC | 05:32 | |
*** iceyao has joined #openstack-kolla | 05:46 | |
*** iceyao has quit IRC | 05:50 | |
*** MarginHu has quit IRC | 06:00 | |
*** MarginHu has joined #openstack-kolla | 06:00 | |
*** StephenW_ has quit IRC | 06:06 | |
*** StephenWang1991 has joined #openstack-kolla | 06:07 | |
*** jianyi has joined #openstack-kolla | 06:15 | |
*** thomasem_ has quit IRC | 06:16 | |
*** thomasem_ has joined #openstack-kolla | 06:20 | |
*** jayahn has quit IRC | 06:21 | |
*** leseb has quit IRC | 06:21 | |
*** leseb has joined #openstack-kolla | 06:28 | |
*** kolla-slack1 has joined #openstack-kolla | 06:36 | |
*** yee379 has quit IRC | 06:36 | |
*** kolla-slack has quit IRC | 06:36 | |
*** yee379 has joined #openstack-kolla | 06:37 | |
*** shardy has joined #openstack-kolla | 06:42 | |
*** chas has joined #openstack-kolla | 06:43 | |
zhuzeyu | hi | 06:47 |
tervancovan | hi | 06:48 |
*** chas has quit IRC | 06:48 | |
zhuzeyu | i have a question want to ask, does the version of kubernetes-entrypoint iamges only use 4.0.0 | 06:49 |
*** salv-orlando has quit IRC | 06:50 | |
zhuzeyu | i modify the images version in cloud.yaml , other service can pull images with 4.0.1 in local registry, but I find the in log that kubernetes-entrypoint still use 4.0.0 | 06:51 |
masber | good afternoon all, horizon ubuntu binaries images from docker hub looks like doesnt have css and images? | 06:52 |
zhuzeyu | but there is no 4.0.0 kubernetes-entrypoint in my local registry, so it get eror | 06:52 |
masber | I can get to the login page but image and format is missing | 06:52 |
*** iceyao has joined #openstack-kolla | 06:53 | |
*** athomas has joined #openstack-kolla | 06:54 | |
masber | same issue after login | 06:55 |
masber | lots of http 404 errors form the website | 06:55 |
*** iceyao has quit IRC | 06:57 | |
*** satyar has joined #openstack-kolla | 06:58 | |
*** numans has joined #openstack-kolla | 07:00 | |
openstackgerrit | Numan Siddique proposed openstack/kolla master: Support openvswitch-ovn docker images https://review.openstack.org/459179 | 07:00 |
openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible stable/ocata: Fix outdated InfluxDB configuration https://review.openstack.org/456491 | 07:02 |
*** chas has joined #openstack-kolla | 07:07 | |
*** manheim has joined #openstack-kolla | 07:20 | |
*** magicboiz has joined #openstack-kolla | 07:21 | |
*** manheim has quit IRC | 07:21 | |
*** manheim has joined #openstack-kolla | 07:22 | |
*** pcaruana has joined #openstack-kolla | 07:24 | |
*** ssurana has quit IRC | 07:26 | |
*** lpetrut has joined #openstack-kolla | 07:26 | |
hrw | morning | 07:26 |
*** ssurana has joined #openstack-kolla | 07:27 | |
*** shardy has quit IRC | 07:28 | |
*** matrohon_ has joined #openstack-kolla | 07:31 | |
*** ssurana has quit IRC | 07:31 | |
manheim | hello | 07:33 |
*** jaosorior has joined #openstack-kolla | 07:33 | |
*** rmart04 has joined #openstack-kolla | 07:34 | |
tervancovan | Can I use just one NIC (eth) to test of kolla-k8s? | 07:36 |
tervancovan | I have a just one nic device (eth0) | 07:36 |
hrw | my patch of a day is https://review.openstack.org/#/c/430940 ;D | 07:37 |
tervancovan | so Can I set "neutron_external_interface=eth0" like this? | 07:37 |
*** shardy has joined #openstack-kolla | 07:40 | |
*** manheim has quit IRC | 07:42 | |
*** lpetrut has quit IRC | 07:43 | |
*** Serlex has joined #openstack-kolla | 07:44 | |
*** iceyao has joined #openstack-kolla | 07:48 | |
*** iceyao has quit IRC | 07:52 | |
*** blallau has left #openstack-kolla | 07:55 | |
*** blallau has joined #openstack-kolla | 07:55 | |
*** salv-orlando has joined #openstack-kolla | 07:58 | |
*** iceyao has joined #openstack-kolla | 08:01 | |
*** gfidente has joined #openstack-kolla | 08:07 | |
*** manheim has joined #openstack-kolla | 08:10 | |
*** lpetrut has joined #openstack-kolla | 08:12 | |
Serlex | Morning mgkwill | 08:12 |
*** dixiaoli has joined #openstack-kolla | 08:16 | |
*** oanson has quit IRC | 08:23 | |
*** oanson has joined #openstack-kolla | 08:24 | |
openstackgerrit | Bertrand Lallau proposed openstack/kolla-ansible master: Tempest: remove all oslo.policy related stuff https://review.openstack.org/459190 | 08:24 |
*** ssurana has joined #openstack-kolla | 08:29 | |
*** magicboiz has quit IRC | 08:30 | |
*** pbourke has quit IRC | 08:31 | |
*** pbourke has joined #openstack-kolla | 08:31 | |
zhuzeyu | tervancovan, need two nic device in doc | 08:33 |
*** ssurana has quit IRC | 08:35 | |
*** zhubingbing has quit IRC | 08:42 | |
*** magicboiz has joined #openstack-kolla | 08:42 | |
*** zhubingbing has joined #openstack-kolla | 08:44 | |
openstackgerrit | Dou Rui Yuan proposed openstack/kolla-ansible master: Use cryptography instead of pycrypto https://review.openstack.org/459195 | 08:47 |
tervancovan | zhuzeyu thank you for reminding it but I have only one device right now so I am looking for work around right now... | 08:54 |
*** mgoddard has joined #openstack-kolla | 08:58 | |
zhuzeyu | why not add virtual network interface to test | 09:15 |
*** zhubingbing has quit IRC | 09:16 | |
*** zhubingbing_ has joined #openstack-kolla | 09:18 | |
*** skramaja has quit IRC | 09:19 | |
*** skramaja has joined #openstack-kolla | 09:19 | |
*** rmart04_ has joined #openstack-kolla | 09:23 | |
*** lpetrut has quit IRC | 09:24 | |
openstackgerrit | Merged openstack/kolla-ansible stable/ocata: Fix outdated InfluxDB configuration https://review.openstack.org/456491 | 09:24 |
*** rmart04 has quit IRC | 09:24 | |
*** rmart04_ is now known as rmart04 | 09:24 | |
*** MarginHu has quit IRC | 09:24 | |
*** MarginHu has joined #openstack-kolla | 09:24 | |
openstackgerrit | Merged openstack/kolla-ansible stable/ocata: neutron_vpnaas_agent: kolla_docker parameter 'volumes' is wrong https://review.openstack.org/457879 | 09:25 |
openstackgerrit | zhuangzejian proposed openstack/kolla-ansible master: Project service has none description in deployment https://review.openstack.org/459211 | 09:26 |
*** MarginHu has quit IRC | 09:28 | |
*** rmart04 has quit IRC | 09:29 | |
*** rmart04 has joined #openstack-kolla | 09:30 | |
openstackgerrit | Dou Rui Yuan proposed openstack/kolla-ansible master: Use cryptography instead of pycrypto https://review.openstack.org/459195 | 09:31 |
*** zhuzeyu has quit IRC | 09:38 | |
*** zhuzeyu has joined #openstack-kolla | 09:39 | |
*** StephenWang1991 has quit IRC | 09:39 | |
*** StephenWang1991 has joined #openstack-kolla | 09:40 | |
*** zhubingbing_ has quit IRC | 09:41 | |
*** sambetts|afk is now known as sambetts | 09:47 | |
*** jmccarthy has joined #openstack-kolla | 09:47 | |
*** salv-orl_ has joined #openstack-kolla | 09:49 | |
*** satyar has quit IRC | 09:52 | |
*** salv-orlando has quit IRC | 09:52 | |
*** masber has quit IRC | 09:54 | |
*** dixiaoli_ has joined #openstack-kolla | 09:54 | |
*** dixiaoli has quit IRC | 09:56 | |
spsurya | tervancovan: you can create dummy one | 09:58 |
spsurya | with ip link | 09:58 |
tervancovan | spsurya thanks | 09:58 |
tervancovan | spsurya could you explain or share link of document for it? | 09:59 |
spsurya | tervancovan: sure | 09:59 |
*** zhubingbing has joined #openstack-kolla | 10:04 | |
*** iceyao has quit IRC | 10:07 | |
*** skramaja_ has joined #openstack-kolla | 10:09 | |
*** skramaja has quit IRC | 10:09 | |
*** cuongnv has quit IRC | 10:10 | |
*** egonzalez has joined #openstack-kolla | 10:11 | |
*** jascott1 has joined #openstack-kolla | 10:11 | |
openstackgerrit | hosup choi proposed openstack/kolla-kubernetes master: Seperate kolla and kolla-ansible search paths Add method to find kolla-ansible path Fix variable name from kolla_dir to kolla_ansible_dir https://review.openstack.org/459228 | 10:12 |
*** jascott1 has quit IRC | 10:17 | |
*** tovin07_ has quit IRC | 10:18 | |
jmccarthy | Have any "4.0.1" images been pushed or ? I tried a deploy but I think it complained about missing toolbox (I was able to deploy "4.0.0") | 10:18 |
egonzalez | jmccarthy no, images aren't pushed yet | 10:20 |
jmccarthy | Ah ok just checking it wasn't just me :) | 10:22 |
*** yangyapeng has quit IRC | 10:22 | |
sdake | spsurya the reason dns fails to enter 3/3 is because the CNI is not yet operational | 10:22 |
sdake | morning folks | 10:24 |
tervancovan | morning sdake | 10:24 |
sdake | morning tervancovan | 10:24 |
sdake | jmccarthy i dont know if 4.0.1 images have been pushed or not | 10:25 |
spsurya | tervancovan: $ sudo ip link add eth1 type dummy | true | 10:26 |
spsurya | sudo ip link set eth1 up | 10:26 |
spsurya | sdake: morning | 10:26 |
*** jianyi has quit IRC | 10:26 | |
tervancovan | Thaaaanks spsurya | 10:27 |
*** StephenWang1991 has quit IRC | 10:28 | |
*** iceyao has joined #openstack-kolla | 10:30 | |
*** ssurana has joined #openstack-kolla | 10:31 | |
spsurya | tervancovan: np | 10:34 |
spsurya | sdake: what can be the possible solution | 10:35 |
spsurya | ? | 10:35 |
sdake | spsurya run kubectl get pods -n kube-system -> paste | 10:35 |
*** ssurana has quit IRC | 10:36 | |
*** oanson has quit IRC | 10:37 | |
spsurya | not running atm | 10:37 |
spsurya | willl paste soon | 10:37 |
*** oanson has joined #openstack-kolla | 10:37 | |
*** felixtsang has joined #openstack-kolla | 10:38 | |
felixtsang | hi every one, newly join the chat . i am studying kolla deployment. A question : Can kolla deploy HA controller nodes? | 10:39 |
jmccarthy | Ok, I'll have a go at building them instead maybe in a while, for me I seem to be having trouble since openvswitch was broken out into it's own role (out from under neutron) | 10:39 |
spsurya | sdake: | 10:41 |
spsurya | https://www.irccloud.com/pastebin/5vBvqXRH/ | 10:41 |
spsurya | just run. I will run this again. | 10:42 |
spsurya | https://www.irccloud.com/pastebin/s18yoEu1/ | 10:42 |
spsurya | all ran except dns one | 10:42 |
egonzalez | jmccarthy ovs split should not have any behaviour in kolla images, images were already independent, but ansible role was merged into neutron | 10:45 |
egonzalez | felixtsang yep, just add more nodes into [control] group at inventory file | 10:45 |
jmccarthy | Sorry I don't mean to confuse images/ansible - I mean get both by pip install kolla-ansible==<version> | 10:46 |
felixtsang | ++egonzalez | 10:46 |
jmccarthy | I think it's the other way ? openvswitch is no longer under neutron now | 10:46 |
egonzalez | jmccarthy yep, was merged and splited out now | 10:47 |
jmccarthy | Which is why I wanted to try 4.0.1, thinking that they are split in that, and see if it worked :) | 10:48 |
egonzalez | jmccarthy nope, in 4.0.1 (stable) is not splited, only in current master branch | 10:49 |
jmccarthy | Oh right - ok 4.0.1 wouldn't help me then :) | 10:49 |
sdake | spsurya was this not working for you previously? | 10:51 |
spsurya | sdake: it was working for me | 10:51 |
sdake | and then stopped working? | 10:51 |
spsurya | but since two days it is not | 10:51 |
spsurya | IIRC | 10:51 |
*** felixtsang has quit IRC | 10:51 | |
sdake | i haven' tused the guide in about two days | 10:51 |
sdake | so - it could be that canal images have changed in some subtle way | 10:52 |
sdake | or it could be an environmental problem on your end | 10:52 |
spsurya | I am runing the same i used to | 10:52 |
*** iceyao has quit IRC | 10:54 | |
spsurya | sdake: ttyl | 10:56 |
sdake | spsurya night - i'll try to look at that problem today | 10:56 |
sdake | spsurya i just woke up - so getting warmped up | 10:56 |
openstackgerrit | Merged openstack/kolla-ansible stable/ocata: Mount /lib/modules folder for cinder-volume container https://review.openstack.org/458430 | 11:03 |
*** dave-mccowan has joined #openstack-kolla | 11:09 | |
*** zhugaoxiao has quit IRC | 11:13 | |
hrw | sdake: can you look at https://review.openstack.org/430940 one? (main non-x86 support one) | 11:13 |
*** salv-orl_ has quit IRC | 11:15 | |
*** satyar has joined #openstack-kolla | 11:16 | |
*** zhurong has joined #openstack-kolla | 11:17 | |
*** zhugaoxiao has joined #openstack-kolla | 11:17 | |
*** janki has joined #openstack-kolla | 11:21 | |
*** rhallisey has joined #openstack-kolla | 11:28 | |
*** oanson has quit IRC | 11:28 | |
openstackgerrit | Merged openstack/kolla-ansible master: Tempest: remove all oslo.policy related stuff https://review.openstack.org/459190 | 11:29 |
*** oanson has joined #openstack-kolla | 11:29 | |
*** iniazi_ has joined #openstack-kolla | 11:29 | |
egonzalez | blallau ping | 11:30 |
*** iniazi has quit IRC | 11:30 | |
blallau | yep | 11:32 |
*** zhugaoxiao has quit IRC | 11:34 | |
*** zhugaoxiao has joined #openstack-kolla | 11:35 | |
*** lpetrut has joined #openstack-kolla | 11:38 | |
*** Pavo has joined #openstack-kolla | 11:39 | |
spsurya | sdake: just came from office | 11:42 |
openstackgerrit | Merged openstack/kolla-ansible master: Fix swift-ring-builder path for container https://review.openstack.org/459054 | 11:44 |
*** dixiaoli_ has quit IRC | 11:48 | |
*** lpetrut_ has joined #openstack-kolla | 11:48 | |
openstackgerrit | Zeyu Zhu proposed openstack/kolla-kubernetes master: Modify the image_tag to 4.0.0 in all_values.yaml https://review.openstack.org/459250 | 11:48 |
openstackgerrit | Merged openstack/kolla-ansible master: Add kolla-ansible to operating-kolla.rst document https://review.openstack.org/456110 | 11:50 |
*** janki has quit IRC | 11:53 | |
*** jchhatbar has joined #openstack-kolla | 11:53 | |
*** rwallner has joined #openstack-kolla | 11:59 | |
*** iceyao has joined #openstack-kolla | 11:59 | |
*** slagle has joined #openstack-kolla | 12:00 | |
*** salv-orlando has joined #openstack-kolla | 12:00 | |
*** lpetrut_ has quit IRC | 12:01 | |
*** lpetrut_ has joined #openstack-kolla | 12:03 | |
*** lpetrut_ has quit IRC | 12:03 | |
*** lpetrut_ has joined #openstack-kolla | 12:03 | |
*** athomas has quit IRC | 12:06 | |
openstackgerrit | Marcin Juszkiewicz proposed openstack/kolla master: ceph: take care of ceph-fuse package availability https://review.openstack.org/440634 | 12:06 |
*** athomas has joined #openstack-kolla | 12:12 | |
*** jascott1 has joined #openstack-kolla | 12:13 | |
*** jchhatbar is now known as janki | 12:16 | |
*** jascott1 has quit IRC | 12:18 | |
*** lpetrut_ has quit IRC | 12:19 | |
*** lpetrut_ has joined #openstack-kolla | 12:19 | |
*** haplo37_ has quit IRC | 12:23 | |
openstackgerrit | Pavel Gluschak (scsnow) proposed openstack/kolla-ansible master: kolla_toolbox: allow docker api version to be specified https://review.openstack.org/459269 | 12:28 |
*** haplo37_ has joined #openstack-kolla | 12:34 | |
*** ssurana has joined #openstack-kolla | 12:34 | |
*** rwellum has joined #openstack-kolla | 12:35 | |
openstackgerrit | hosup choi proposed openstack/kolla-kubernetes master: Seperate kolla and kolla-ansible search paths Add method to find kolla-ansible path Add dictionary variable for ymls Add array variable for roles Fix variable name from kolla_dir to kolla_ansible_dir https://review.openstack.org/459228 | 12:36 |
openstackgerrit | jimmygc proposed openstack/kolla-ansible master: Add vmware DVS support to kolla-ansible https://review.openstack.org/459270 | 12:38 |
*** ssurana has quit IRC | 12:38 | |
*** zhurong has quit IRC | 12:39 | |
openstackgerrit | hosup choi proposed openstack/kolla-kubernetes master: Update deployment-guide for ubuntu user Add title for installing ubunut package Change-Id: Ib67a90ac2b0504e9fe6b635b71612b87631507d1 https://review.openstack.org/459163 | 12:39 |
openstackgerrit | jimmygc proposed openstack/kolla master: Support VMware DVS https://review.openstack.org/459273 | 12:45 |
openstackgerrit | jimmygc proposed openstack/kolla-ansible master: Add vmware DVS support to kolla-ansible https://review.openstack.org/459270 | 12:46 |
*** salv-orlando has quit IRC | 12:47 | |
openstackgerrit | hosup choi proposed openstack/kolla-kubernetes master: Seperate kolla and kolla-ansible search paths https://review.openstack.org/459228 | 12:48 |
openstackgerrit | Eduardo Gonzalez proposed openstack/kolla-ansible master: Add cli options to cleanup-images https://review.openstack.org/431503 | 12:53 |
*** manheim has quit IRC | 12:53 | |
openstackgerrit | Merged openstack/kolla-ansible master: Fix the remove mariadb volume failed https://review.openstack.org/447359 | 12:53 |
*** lpetrut_ has quit IRC | 12:54 | |
*** lpetrut has quit IRC | 12:57 | |
*** goldyfruit has joined #openstack-kolla | 12:58 | |
*** kencjohnston has quit IRC | 13:02 | |
*** iceyao has left #openstack-kolla | 13:02 | |
*** yangyapeng has joined #openstack-kolla | 13:05 | |
*** gkadam has quit IRC | 13:07 | |
*** schwicht has joined #openstack-kolla | 13:11 | |
*** trozet has joined #openstack-kolla | 13:12 | |
*** Guest31033 has quit IRC | 13:22 | |
*** lpetrut has joined #openstack-kolla | 13:28 | |
*** jtriley has joined #openstack-kolla | 13:31 | |
*** srwilkers has joined #openstack-kolla | 13:33 | |
Serlex | mgkwill: it was inventory issue, ignore :) | 13:33 |
*** n0isyn0ise has left #openstack-kolla | 13:34 | |
*** lpetrut has quit IRC | 13:34 | |
*** eanylin has joined #openstack-kolla | 13:46 | |
*** manheim has joined #openstack-kolla | 13:47 | |
*** tovin07_ has joined #openstack-kolla | 13:52 | |
*** manheim has quit IRC | 13:53 | |
*** manheim has joined #openstack-kolla | 13:54 | |
*** tovin07_ has left #openstack-kolla | 13:56 | |
*** manheim_ has joined #openstack-kolla | 14:01 | |
*** manheim has quit IRC | 14:04 | |
*** apuimedo has joined #openstack-kolla | 14:15 | |
apuimedo | inc0: sdake: We need to push some images to dockerhub and I was wondering if you use openstack infra for the pushes to https://hub.docker.com/u/kolla/ | 14:16 |
dmellado | egonzalez: ^^ ;) | 14:17 |
egonzalez | apuimedo for now is manual work, but automating pushes is wip atm | 14:18 |
apuimedo | egonzalez: define manual | 14:18 |
dmellado | egonzalez: oh, I thought that was automated already | 14:18 |
dmellado | to you have some documentation on it? | 14:19 |
egonzalez | build local and push to dockerhub by someone | 14:19 |
dmellado | s/t/d | 14:19 |
apuimedo | as in somebody builds them on their laptop/server and then massively pushes? With a script | 14:19 |
apuimedo | ok, fair enough | 14:19 |
inc0 | it's nto automated | 14:19 |
inc0 | but that's top of my list | 14:19 |
dmellado | is there any spec or plan to have that automated? | 14:19 |
dmellado | inc0: ? | 14:19 |
apuimedo | inc0: egonzalez: what's the current work on automating it? | 14:19 |
hrw | can someone (other than egonzalez) check https://review.openstack.org/#/c/434810/ patch? it handles nova-libvirt on ppc64le | 14:20 |
inc0 | we had a lot of discussions regarging tag naming and stuff like tha | 14:20 |
inc0 | t | 14:20 |
inc0 | I'll work on script today | 14:20 |
openstackgerrit | Merged openstack/kolla master: handle rtslib(-fb) package names and dependencies https://review.openstack.org/450805 | 14:20 |
dmellado | inc0: I'll keep an eye on it, thanks! | 14:21 |
apuimedo | inc0: and you'll run it on the infra, or externally for now? | 14:21 |
inc0 | I'll try to get it to infra asap | 14:22 |
dmellado | so the image building is done just in local laptop, got it | 14:22 |
inc0 | yeah | 14:23 |
inc0 | we want to run daily builds | 14:24 |
apuimedo | inc0: we wanted to do per commit | 14:26 |
apuimedo | since our build burden is lower | 14:26 |
inc0 | well per commit is a little overkill imho | 14:27 |
hrw | rtslib merged! yay! :D | 14:28 |
*** yangyape_ has joined #openstack-kolla | 14:28 | |
inc0 | do you think we need per-commit | 14:28 |
inc0 | ? | 14:28 |
inc0 | (it's also going to be a bit of logistical problem) | 14:28 |
inc0 | daily - I'm going to use prebuild images from latest commit | 14:29 |
inc0 | we upload built images per commit in form of tarballed registry | 14:29 |
inc0 | to tarballs.openstack.org | 14:29 |
*** yangyapeng has quit IRC | 14:30 | |
apuimedo | inc0: nah... We don't really need it | 14:31 |
apuimedo | it's because we were using dockerhub automated builds | 14:31 |
*** magicboiz has quit IRC | 14:31 | |
Reepicheep | out of curiosity.. and forgive me I haven't looked but are there any smoke / spec tests beyond docker build status for the images? | 14:33 |
*** magicboiz has joined #openstack-kolla | 14:33 | |
inc0 | Reepicheep: we also run deploy gates on merhe | 14:33 |
inc0 | on-voting but we don't merge if they fail | 14:33 |
*** srwilkers has quit IRC | 14:33 | |
Reepicheep | ok.. I see so that is pre-commit | 14:34 |
inc0 | so it works like this, you commit- gates run, you merge-images are uploaded, daily we take latest uploaded images and will push it to dockerhub | 14:35 |
*** lpetrut has joined #openstack-kolla | 14:35 | |
openstackgerrit | Merged openstack/kolla master: Fix tempest dockerfile https://review.openstack.org/459152 | 14:36 |
Reepicheep | thanks, That makes sense. I would agree per-commit pushs to docker hub would seem a little over kill but daily would be nice | 14:36 |
*** eaguilar has joined #openstack-kolla | 14:37 | |
*** srwilkers has joined #openstack-kolla | 14:38 | |
*** magicboiz has quit IRC | 14:38 | |
hrw | down to 4 patches in queue | 14:43 |
hrw | https://review.openstack.org/430940 adds non-x86 support. tested with debian, centos, ubuntu on aarch64 (150-200 images per combination), ppc64le (good but worse that aarch64 due to lack of packages in centos) and x86-64 (to know it works still). | 14:47 |
hrw | https://review.openstack.org/434810 takes care of nova-libvirt dependencies on ppc64le (no Xen, no DMI) | 14:48 |
hrw | https://review.openstack.org/440634 takes care of 'ceph-fuse' package availability (which is poor) | 14:48 |
hrw | and https://review.openstack.org/444832 (mariadb repos under CentOS) is waiting for recheck results | 14:49 |
hrw | if someone has a time then please take a look | 14:49 |
*** japestinho has quit IRC | 14:50 | |
manheim_ | i can see that ansible 2.3 is in apt repos, did you have any issues with it? | 14:52 |
manheim_ | maybe something in TASK [mariadb : Creating haproxy mysql user] | 14:52 |
openstackgerrit | Pavel Gluschak (scsnow) proposed openstack/kolla-ansible master: kolla_toolbox: allow docker api version to be specified https://review.openstack.org/459269 | 14:55 |
*** manheim_ has quit IRC | 14:57 | |
*** lpetrut has quit IRC | 14:57 | |
*** magicboiz has joined #openstack-kolla | 14:57 | |
*** adrian_otto has joined #openstack-kolla | 15:01 | |
*** salv-orlando has joined #openstack-kolla | 15:02 | |
*** manheim has joined #openstack-kolla | 15:05 | |
*** eaguilar_ has joined #openstack-kolla | 15:06 | |
egonzalez | hrw what you mean at comment in https://review.openstack.org/#/c/431503/6 ? | 15:09 |
*** eaguilar has quit IRC | 15:09 | |
hrw | egonzalez: you run 'docker images' for each argument | 15:09 |
hrw | egonzalez: if I run that script with more than one option then you run 'docker images' per option | 15:10 |
egonzalez | hrw is only run once, since each argument cannot be used with others | 15:10 |
hrw | egonzalez: "cleanup-images --all --dangling" has sense still | 15:11 |
hrw | egonzalez: --all will remove all Kolla ones, --dangling will remove orphans and base distro images stay | 15:12 |
*** skramaja_ has quit IRC | 15:12 | |
hrw | sure, it will remove just the second ones which will be another script bug | 15:13 |
*** janki has quit IRC | 15:13 | |
openstackgerrit | Merged openstack/kolla stable/newton: Fix neutron-vpnaas for redhat derived bases https://review.openstack.org/449952 | 15:13 |
*** felixtsang has joined #openstack-kolla | 15:14 | |
*** ssurana has joined #openstack-kolla | 15:15 | |
hrw | egonzalez: is my answer good? | 15:16 |
egonzalez | hrw, --all remove also kolla dangling images (not all dangling for non kolla images) | 15:17 |
hrw | I thought that --dangling is more for untagged ones like this: | 15:19 |
hrw | <none> <none> 4ad1f9535b94 2 weeks ago 522.3 MB | 15:19 |
*** dixiaoli has joined #openstack-kolla | 15:19 | |
hrw | I use docker-clean usually | 15:19 |
*** ssurana has quit IRC | 15:19 | |
egonzalez | hrw yep, filter dangling=true output all dangling images (kolla and not kolla) | 15:20 |
egonzalez | hrw --all removes all images with label=kolla_version (this include orphaned images, since also have the label) | 15:21 |
hrw | egonzalez: so --all --dangling still have sense? | 15:21 |
*** apuimedo is now known as apuimedo|away | 15:21 | |
hrw | on my aarch64/ssd machine 'docker images|wc -l' with 1184 images takes 21.69s | 15:21 |
*** dixiaoli has quit IRC | 15:22 | |
hrw | my x86-64/ssd does same in 3.12s (1097 images) | 15:22 |
hrw | in my 'lets see how many images we already built' I have 6 iterations over 'docker images' - moved to 'ask once, store in file' after 3rd run of script | 15:24 |
*** dixiaoli has joined #openstack-kolla | 15:24 | |
*** eaguilar_ has quit IRC | 15:26 | |
*** dixiaoli has quit IRC | 15:28 | |
*** david-lyle has joined #openstack-kolla | 15:29 | |
*** mgoddard has quit IRC | 15:30 | |
*** trozet has quit IRC | 15:31 | |
*** krtaylor has quit IRC | 15:34 | |
*** eaguilar has joined #openstack-kolla | 15:34 | |
kfox1111 | morning. | 15:37 |
hrw | hi kfox1111 | 15:40 |
*** zhurong has joined #openstack-kolla | 15:42 | |
*** apuimedo|away is now known as apuimedo | 15:44 | |
*** felixtsang has quit IRC | 15:45 | |
*** salv-orl_ has joined #openstack-kolla | 15:49 | |
*** felixtsang has joined #openstack-kolla | 15:50 | |
kfox1111 | hrw: hi. :) | 15:51 |
*** rmart04 has quit IRC | 15:51 | |
*** salv-orlando has quit IRC | 15:52 | |
*** zhurong has quit IRC | 15:57 | |
*** blallau has quit IRC | 16:01 | |
openstackgerrit | Bertrand Lallau proposed openstack/kolla-kubernetes master: deployment-guide.rst enhancement and small fixes https://review.openstack.org/459392 | 16:02 |
*** zhurong has joined #openstack-kolla | 16:05 | |
*** zhurong has quit IRC | 16:07 | |
*** mdnadeem_home has joined #openstack-kolla | 16:08 | |
*** mgoddard has joined #openstack-kolla | 16:11 | |
*** manheim has quit IRC | 16:13 | |
*** manheim has joined #openstack-kolla | 16:14 | |
*** Serlex has quit IRC | 16:19 | |
*** manheim has quit IRC | 16:20 | |
*** jmccarthy has left #openstack-kolla | 16:21 | |
*** salv-orl_ has quit IRC | 16:26 | |
*** matrohon_ has quit IRC | 16:29 | |
*** jascott1 has joined #openstack-kolla | 16:31 | |
*** krtaylor has joined #openstack-kolla | 16:31 | |
*** rmart04 has joined #openstack-kolla | 16:33 | |
openstackgerrit | Merged openstack/kolla master: mariadb: handle lack of external repos on non x86 https://review.openstack.org/444832 | 16:34 |
hrw | YAY! | 16:35 |
*** rmart04 has quit IRC | 16:38 | |
*** srwilkers has quit IRC | 16:39 | |
*** athomas has quit IRC | 16:40 | |
*** athomas has joined #openstack-kolla | 16:50 | |
*** egonzalez has quit IRC | 16:51 | |
*** felixtsang has quit IRC | 16:59 | |
*** felixtsang has joined #openstack-kolla | 16:59 | |
*** jaosorior is now known as jaosorior_away | 16:59 | |
*** adrian_otto1 has joined #openstack-kolla | 17:06 | |
*** zhugaoxiao has quit IRC | 17:06 | |
*** zhugaoxiao has joined #openstack-kolla | 17:07 | |
*** adrian_otto has quit IRC | 17:09 | |
*** jemcevoy has joined #openstack-kolla | 17:12 | |
*** srwilkers has joined #openstack-kolla | 17:14 | |
*** imcsk8 has quit IRC | 17:23 | |
*** imcsk8 has joined #openstack-kolla | 17:23 | |
*** ssurana has joined #openstack-kolla | 17:23 | |
*** adrian_otto1 has quit IRC | 17:23 | |
*** srwilkers has quit IRC | 17:38 | |
*** salv-orlando has joined #openstack-kolla | 17:38 | |
*** gfidente is now known as gfidente|afk | 17:38 | |
*** mdnadeem_home has quit IRC | 17:41 | |
*** tonanhngo has joined #openstack-kolla | 17:45 | |
*** felixtsang has quit IRC | 17:48 | |
*** jascott1 has quit IRC | 17:51 | |
*** srwilkers has joined #openstack-kolla | 17:54 | |
*** chas has quit IRC | 17:56 | |
*** felixtsang has joined #openstack-kolla | 17:58 | |
*** manheim has joined #openstack-kolla | 18:16 | |
inc0 | sleepy Monday I see | 18:27 |
*** manheim has quit IRC | 18:29 | |
*** manheim has joined #openstack-kolla | 18:29 | |
kfox1111 | indeed | 18:36 |
*** felixtsang has quit IRC | 18:38 | |
*** felixtsang has joined #openstack-kolla | 18:39 | |
kfox1111 | interesting.... https://review.openstack.org/#/c/456760/9/ansible/tune/roles/logging/templates/logging.conf.j2 | 18:40 |
*** manheim has quit IRC | 18:41 | |
*** manheim has joined #openstack-kolla | 18:41 | |
*** felixtsang has quit IRC | 18:43 | |
*** salv-orlando has quit IRC | 18:45 | |
*** mgoddard has quit IRC | 18:48 | |
inc0 | never used this project | 18:48 |
*** manheim has quit IRC | 18:59 | |
*** srwilkers has quit IRC | 18:59 | |
*** schwicht has quit IRC | 18:59 | |
*** jascott1 has joined #openstack-kolla | 19:08 | |
openstackgerrit | Bertrand Lallau proposed openstack/kolla-kubernetes master: deployment-guide.rst enhancement and small fixes https://review.openstack.org/459392 | 19:09 |
*** schwicht has joined #openstack-kolla | 19:13 | |
*** satyar has quit IRC | 19:18 | |
inc0 | centos people, for kolla-ansible "Aborting, target uses selinux but python bindings (libselinux-python) aren't installed!" <- how to handle this error | 19:19 |
inc0 | ? | 19:19 |
inc0 | sdake sbezverk ^ | 19:21 |
sbezverk | inc0: never seen this error before.. | 19:21 |
inc0 | thats kolla-ansible tho, but I expect it to be something non-ansible specifi | 19:22 |
inc0 | c | 19:22 |
*** dciabrin has quit IRC | 19:23 | |
*** srwilkers has joined #openstack-kolla | 19:23 | |
sbezverk | inc0: if selinux get completely disabled, do you still see it? | 19:23 |
inc0 | disable selinux...that's it probably | 19:23 |
*** dciabrin has joined #openstack-kolla | 19:24 | |
*** schwicht has quit IRC | 19:24 | |
sbezverk | inc0: setenforce 0 and try | 19:24 |
inc0 | y, it was enforcing | 19:24 |
*** Pavo has quit IRC | 19:26 | |
*** jmccarthy has joined #openstack-kolla | 19:27 | |
*** schwicht has joined #openstack-kolla | 19:27 | |
jmccarthy | In the all-in-one inventory, are additional sections for openvswitch -db, and -vswitchd needed ? Just wondering, if sub components are not listed, then they all go according to the main openvswitch definition is it ? | 19:29 |
jmccarthy | (or mulitnode for that matter) | 19:30 |
sean-k-mooney | inc0: was that form a nova-compute log | 19:30 |
inc0 | sean-k-mooney: deployment | 19:30 |
inc0 | it fails on creating file on localhost by mariadb role | 19:31 |
sean-k-mooney | oh this wasnt spawning a vm | 19:31 |
sean-k-mooney | ah i see | 19:31 |
inc0 | jmccarthy: there is little difference between aio and multinode | 19:31 |
inc0 | ehh I'm new to centos :( | 19:32 |
jmccarthy | Ok, what I mean is openvswitch has, network, compute, and manilla-share - so the openvswitch_vswitchd and openvswitch_db for example will end up on all of them or ? | 19:33 |
*** shardy has quit IRC | 19:33 | |
sean-k-mooney | i would guess there has been a change to hte mysql binary that nolonger pulls in that dependecy | 19:33 |
inc0 | hmmm...good point | 19:33 |
sean-k-mooney | jmccarthy: it should | 19:34 |
inc0 | there is no need for 2 groups I think | 19:34 |
*** shardy has joined #openstack-kolla | 19:34 | |
sean-k-mooney | openvswitch is required to be deployed on nodes in the network, compute, and manilla-share groups | 19:34 |
sean-k-mooney | i think its only going to do it once though even if the host is listed in all of the above | 19:35 |
jmccarthy | sean, inc0 - ok I don't know it well enough offhand, I seem to be having issue with openvswitch since it is listed separately from neutron now - didn't know if it was related to that | 19:35 |
*** salv-orlando has joined #openstack-kolla | 19:35 | |
sean-k-mooney | jmccarthy: it should be enabled by default so you should not have need to change anything if you are using the default inventory file | 19:36 |
inc0 | this task fails https://github.com/openstack/kolla-ansible/blob/5a4cb8a56fa5b5c37874d21f992d3ce77a11e49f/ansible/roles/mariadb/tasks/lookup_cluster.yml#L8 | 19:36 |
jmccarthy | Ok, I'll have to try it again and see where I'm going wrong :) | 19:37 |
sean-k-mooney | inc0: this is selinux on the host blocking you then as this is a local action | 19:37 |
sean-k-mooney | we used to have an entry in the quickstart a release or 2 ago telling you to disable selinux on the host | 19:38 |
sean-k-mooney | have you priviouly had mariadb installed on the host? | 19:38 |
inc0 | dmsimard for the rescue;) https://dmsimard.com/2016/01/08/selinux-python-virtualenv-chroot-and-ansible-dont-play-nice/ thanks man! | 19:39 |
*** shardy has quit IRC | 19:39 | |
*** felixtsang has joined #openstack-kolla | 19:40 | |
sean-k-mooney | inc0: ah yes ansible_python_interpreter i have used that before | 19:41 |
inc0 | helped. | 19:41 |
sean-k-mooney | so why is this running in a python virtual enviornment anyway? | 19:42 |
sean-k-mooney | is this just a querk of your dev env? | 19:43 |
*** iniazi_ has quit IRC | 19:46 | |
*** iniazi has joined #openstack-kolla | 19:46 | |
*** schwicht has quit IRC | 19:48 | |
*** chas has joined #openstack-kolla | 19:56 | |
*** matrohon_ has joined #openstack-kolla | 19:56 | |
dmsimard | oh hai | 19:58 |
dmsimard | You can set ansible_python_interpreter at the task level if need be | 19:59 |
dmsimard | i.e, https://github.com/rdo-infra/ci-config/blob/master/jenkins/jobs/scripts/weirdo-create-vm.sh#L99-L104 | 19:59 |
* dmsimard goes back to his cave | 19:59 | |
*** chas has quit IRC | 20:01 | |
*** sambetts is now known as sambetts|afk | 20:01 | |
*** pcaruana has quit IRC | 20:01 | |
*** harlowja has quit IRC | 20:03 | |
*** schwicht has joined #openstack-kolla | 20:07 | |
*** MasterOfBugs has joined #openstack-kolla | 20:08 | |
*** manheim has joined #openstack-kolla | 20:09 | |
*** tonanhngo_ has joined #openstack-kolla | 20:12 | |
*** manheim has quit IRC | 20:13 | |
*** tonanhngo has quit IRC | 20:14 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: Docker Logging Support https://review.openstack.org/458649 | 20:16 |
*** tonanhngo has joined #openstack-kolla | 20:20 | |
*** tonanhngo_ has quit IRC | 20:22 | |
inc0 | jemcevoy: you had issues with ovs restarting due to guru meditation? | 20:23 |
inc0 | or sth like that right? | 20:23 |
inc0 | kfox1111: around? | 20:26 |
*** jascott1 has quit IRC | 20:26 | |
*** jascott1 has joined #openstack-kolla | 20:26 | |
inc0 | my ovs container restart with this "Guru meditation now registers SIGUSR1 and SIGUSR2 by default for backward compatibility. SIGUSR1 will no longer be registered in a future release, so please use SIGUSR2 to generate reports." | 20:27 |
inc0 | seen that anyone? | 20:27 |
*** rhallisey has quit IRC | 20:29 | |
*** eaguilar_ has joined #openstack-kolla | 20:37 | |
*** eaguilar_ has quit IRC | 20:37 | |
*** rwallner has quit IRC | 20:38 | |
*** eaguilar has quit IRC | 20:38 | |
*** rwallner has joined #openstack-kolla | 20:38 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: Docker Logging Support https://review.openstack.org/458649 | 20:43 |
*** srwilkers has quit IRC | 20:44 | |
*** srwilkers has joined #openstack-kolla | 20:45 | |
kfox1111 | inc0: ping. | 20:49 |
inc0 | hey | 20:49 |
kfox1111 | whats up? | 20:49 |
inc0 | I'm trying to debug strange issue, but I think I'm onto sth | 20:49 |
inc0 | and it might be k-ans related | 20:49 |
kfox1111 | k. whats up? | 20:50 |
kfox1111 | (or not, which ever the case may be. :) | 20:50 |
*** rwallner has quit IRC | 20:51 | |
*** krtaylor has quit IRC | 20:51 | |
*** rwallner has joined #openstack-kolla | 20:51 | |
inc0 | for some reason my ovs doesn't seem to be deployed | 20:52 |
kfox1111 | ok. | 20:53 |
*** dixiaoli has joined #openstack-kolla | 20:58 | |
*** matrohon_ has quit IRC | 20:58 | |
*** harlowja has joined #openstack-kolla | 20:59 | |
*** adrian_otto has joined #openstack-kolla | 21:02 | |
jascott1 | kfox1111 I figured out the cause of my mysql problem, it couldnt bind port because apparently I am using wrong ip somewhere | 21:03 |
kfox1111 | hm. | 21:03 |
* inc0 facepalm | 21:03 | |
inc0 | I had master in my kolla-ansible | 21:03 |
jascott1 | can anyone take a look at this and explain the mapping of k8s interfaces to kolla-k8s ? https://etherpad.openstack.org/p/knet | 21:04 |
kfox1111 | jascott1: I think it binds to 0.0.0.0. so not sure an ip. but having an already allocated mysql port might do it. | 21:04 |
jascott1 | hmm | 21:05 |
jascott1 | so pod and service CIDRs are not used for kolla specific settings correct? | 21:06 |
kfox1111 | jascott1: thats your config? | 21:06 |
jascott1 | no I just took all the network settings out the guide so I could reason about them | 21:06 |
kfox1111 | k. should I tweak in place or add notes? | 21:07 |
kfox1111 | maybe just add I guess... | 21:07 |
jascott1 | please | 21:07 |
jascott1 | either! | 21:07 |
jascott1 | after I understand it maybe I can write up something for others to explain | 21:07 |
kfox1111 | cool. | 21:09 |
*** Pavo has joined #openstack-kolla | 21:11 | |
*** rwallner has quit IRC | 21:14 | |
*** rwallner has joined #openstack-kolla | 21:15 | |
*** vhosakot has joined #openstack-kolla | 21:18 | |
*** adrian_otto has quit IRC | 21:19 | |
*** Pavo has quit IRC | 21:20 | |
*** adrian_otto has joined #openstack-kolla | 21:21 | |
kfox1111 | jascott1: I think thats the traffic type use cases? | 21:21 |
kfox1111 | can you think of any more? | 21:21 |
jascott1 | i cannot | 21:21 |
jascott1 | thanks so much | 21:21 |
kfox1111 | does that kind of help? | 21:22 |
jascott1 | it does | 21:22 |
jascott1 | so in the end can someone get by with 2 nics plus flannel (and calicos) ? | 21:22 |
kfox1111 | hmm..... depends on how yo uare defining nics I guess. | 21:23 |
kfox1111 | flannel and calico's are virtual. | 21:23 |
jascott1 | right | 21:23 |
kfox1111 | you can get away with having just one physical nic if using vlans. | 21:23 |
kfox1111 | (I do that on one cloud) | 21:24 |
jascott1 | so everything kolla related seems to use mgmt interface | 21:24 |
kfox1111 | could probably get away with 1 network, but two's safer. | 21:24 |
jascott1 | well not external | 21:24 |
kfox1111 | and 3's much preffered if you have bmc's imo. | 21:24 |
jascott1 | in globals, api_interface and tunnel are left as 0.0.0.0 ? | 21:24 |
kfox1111 | yeah. its a carryover from when the config was in kolla-ansible. | 21:25 |
jascott1 | oh ah | 21:25 |
kfox1111 | I think we shoudl move those two config options directly into kolla-kube's config now. | 21:25 |
kfox1111 | so the user never has to touch it. | 21:25 |
kfox1111 | its bouod to 0.0.0.0, but the servers are in a pod's ip, so they only really have 1 ip. | 21:25 |
kfox1111 | so doesn't need to worry about it. | 21:26 |
kfox1111 | in fact, if you got a moment, and wanted to make a ps that went through kolla-kubernetes/ansible and s/api_interface_adddress/0.0.0.0/ that would be cool. :) | 21:26 |
openstackgerrit | Rich Wellum proposed openstack/kolla-kubernetes master: Various cleanups and additions https://review.openstack.org/458939 | 21:27 |
*** rwallner has quit IRC | 21:28 | |
jascott1 | ok thanks | 21:28 |
jascott1 | im going to try to get mine working and then can do | 21:28 |
kfox1111 | ok. cool. :) | 21:28 |
*** rwallner has joined #openstack-kolla | 21:28 | |
rwellum | Hey jascott1 and kfox1111 can you take a look at: http://prnt.sc/f0bmuv - see if it makes any sense? :) | 21:29 |
jascott1 | rwellum i cannot tell you if its correct but it does make sense :) | 21:30 |
*** trozet has joined #openstack-kolla | 21:31 | |
kfox1111 | rwellum: vibr0 is usually used to nat traffic out to the internet, but not traffic back. | 21:31 |
*** mgoddard has joined #openstack-kolla | 21:32 | |
kfox1111 | so... are you trying to make it such that vm's running inside the vm get onto the external network and visa versa, | 21:32 |
*** jtriley has quit IRC | 21:32 | |
kfox1111 | or make it such that vm's can reach out to the external netwokr, but cant be contacted back in? | 21:32 |
*** schwicht has quit IRC | 21:34 | |
kfox1111 | I guess another questionn... is 10.240.43.234 network completely private, or is it your 'corperate private network'? | 21:34 |
kfox1111 | (or home use, or whichever?) | 21:34 |
*** manheim has joined #openstack-kolla | 21:35 | |
*** schwicht has joined #openstack-kolla | 21:35 | |
rwellum | In the deployment guide - sdake states that the neutron interface should be ip-less, for neutron to work. In my case, I just wanted two interfaces and this is usually how I achieve it - one is the physical through, and the other off vibr0. These questions are actually exactly why I wanted to draw this out. | 21:35 |
rwellum | 10.240.. is corporate private. | 21:35 |
kfox1111 | not strictly true. but if you want to use add_port, it will break things if it has an ip. | 21:36 |
kfox1111 | do you want fips on your vm's to show up on corperate private? | 21:36 |
kfox1111 | or just be able to reach them without corperate knowing? | 21:37 |
rwellum | It's handy because I can access them everywhere - but not sure it's what they want... Don't think they want openstack traffic running over their internal network... | 21:37 |
rwellum | I guess if this was bare-metal, 'eth0' would be one of the physical ports with an assigned dhcp address, and eth1 - another port but don't use dhcp or assign a static ip? | 21:38 |
kfox1111 | it will happen either way, its just if they can see them or not. | 21:38 |
rwellum | Yeah true. | 21:38 |
*** krtaylor has joined #openstack-kolla | 21:39 | |
kfox1111 | I guess a related question is, can you get a block of ip's on the 10.240.43.xxx network, or is that all just dhcp? | 21:39 |
*** jascott1 has quit IRC | 21:39 | |
*** jascott1 has joined #openstack-kolla | 21:40 | |
openstackgerrit | Marcus Williams proposed openstack/kolla-ansible master: Add OpenDaylight role https://review.openstack.org/416367 | 21:41 |
kfox1111 | so, in your drawing, I would change "physical interfaces" to be "bridges". | 21:41 |
rwellum | it's just dhcp | 21:42 |
rwellum | ok good point! | 21:42 |
*** jascott1 has quit IRC | 21:42 | |
kfox1111 | then on the right side of bridges, on the top, have eno1(10.240.43.234) | 21:42 |
kfox1111 | and on the right side of vibr0, have a nat box that points up to the eno1? | 21:42 |
kfox1111 | well... technically the ip's probably on br1 and shoudl point there... | 21:43 |
kfox1111 | ok. dhcp network you don't control.... | 21:43 |
kfox1111 | so your only choice is to hide the openstack behind a nat. | 21:43 |
kfox1111 | the vibr0 should be good enough for that. | 21:43 |
kfox1111 | are there going to be multiple machines in this setup? or this kind of a testing laptop kind of thing? | 21:44 |
rwellum | prntscr.com/f0epg0 | 21:44 |
rwellum | Right now it's just one vm. But I'd like to add multiple VM's if possible - to test out multi-node | 21:45 |
kfox1111 | same or different boxes though? | 21:45 |
*** mgoddard has quit IRC | 21:45 | |
rwellum | Same box I think, | 21:45 |
kfox1111 | k. | 21:45 |
sbezverk | kfox1111: I hit interesting issue.. I had to shut down my cluster for the weekend, when I brought it back up there was no connectivity. I suspect it happened because of flannel leases got expired | 21:45 |
kfox1111 | then I'd skip bringing the br1 bridge into the vm at all. | 21:46 |
kfox1111 | sbezverk: entirely possible. did you shut down evereything, or just some of the controllers? | 21:46 |
sbezverk | so when compute node came up it got new range and some how was incapable to reconverge properly | 21:46 |
sbezverk | kfox1111: everything | 21:46 |
sbezverk | shutdown and powerdown | 21:47 |
rwellum | That's a good point kfox1111 - I'll think about the br1 a bit more. | 21:47 |
kfox1111 | sbezverk: hmm... I would have thought the init container would rerun and reip with the changes. | 21:47 |
sbezverk | kfox1111: I had even some native centos containers I use to test connectivity, even these could not ping outside | 21:47 |
rwellum | sbezverk: when you get a chance can you look at http://prntscr.com/f0epg0 and tell me if it's how or similar to how you set up your VM's and what you do differently? | 21:47 |
kfox1111 | rwellum: yeah. so vm (192.168.122.N) -> vibr0 (192.168.122.1) -> nat -> eno1 (10.240.43.234) | 21:48 |
*** Pavo has joined #openstack-kolla | 21:48 | |
kfox1111 | sbezverk: its entirely possible we somehow didn't handle changing an ip somewhere. we don't test it anywhere. | 21:48 |
*** manheim has quit IRC | 21:49 | |
*** salv-orl_ has joined #openstack-kolla | 21:49 | |
kfox1111 | sbezverk: once the reboot gate is more done, that really should go in there. shutdown the node. change the flannel space, then bring it back up. | 21:49 |
sbezverk | kfox1111: I am positive it not our problem, it is kube/networking problem | 21:49 |
sbezverk | rwellum: there is a mistake in the slide | 21:49 |
sbezverk | for cinder volume | 21:49 |
rwellum | got it kfox1111 - thanks! | 21:49 |
kfox1111 | sbezverk: k8s does handle the flannel range changing during reboot. I have tested that for sure on another system. | 21:49 |
sbezverk | we need to specify IP address of the host where that volume group is hosted | 21:50 |
sbezverk | kfox1111: I tested that too, but whith extended time offline, never before | 21:50 |
kfox1111 | rwellum: I'd aslo tweak the libvirt vibr0 default network to reserve half the addresses. then hand those to neutron. | 21:50 |
kfox1111 | sbezverk: yeah. I did run into that exact case and was able to recover from it. at least on kube 1.4. | 21:51 |
*** rwallner has quit IRC | 21:51 | |
kfox1111 | what I think there is a problem with though, is if flannel is self hosted, the compute node doesn't go down, but flannel dies for a significant amount of time, then relaunches. | 21:51 |
*** rwallner has joined #openstack-kolla | 21:51 | |
kfox1111 | which is maybe possible with an extended controller only situation. | 21:51 |
*** salv-orlando has quit IRC | 21:51 | |
kfox1111 | havent tested that case. | 21:51 |
kfox1111 | I thin kthe node has to be rebooted to recover from that case, as the ip's might stickaround. | 21:52 |
*** manheim has joined #openstack-kolla | 21:52 | |
*** goldyfruit has quit IRC | 21:52 | |
kfox1111 | in your case though, not sure why it would be a problem. we would need to dig into it more. | 21:52 |
kfox1111 | I don't know of anything specifically that would cause a problem. | 21:52 |
*** gfidente|afk has quit IRC | 21:52 | |
sbezverk | kfox1111: possibly, but all these just shows how fragile k8s networking is atm.. | 21:53 |
kfox1111 | hmm.... possibly. | 21:53 |
kfox1111 | one of the things I've been starting to do, | 21:53 |
kfox1111 | is switch flannel to get its ip ranges from k8s. | 21:54 |
sbezverk | I was told to try contiv, some people are saying it is more robust.. I guess after the summit I will give it a try | 21:54 |
kfox1111 | it runs etcd'less then and I think is emune to that issue. | 21:54 |
kfox1111 | If k8s is handing out ranges and registering them with the node object, it shoudl be static for the nodes life. | 21:54 |
sbezverk | kfox1111: right if we can eliminate a posibility of stale state, it should be more robust | 21:54 |
sbezverk | kfox1111: right, this whole idea of switching subnets over time gives me chills | 21:55 |
kfox1111 | sbezverk: this is how that is done: https://github.com/kfox1111/charts/tree/flannel/stable/flannel | 21:56 |
kfox1111 | kubeadm init needs to be tweaked like: https://review.openstack.org/#/c/456964/8/tools/setup_kubernetes.sh | 21:57 |
sbezverk | kfox1111: in your case you will not use etcd? | 21:57 |
kfox1111 | so it knows what range to hand out. | 21:57 |
kfox1111 | yeah. there already is an etcd cluster there, in k8s. | 21:57 |
kfox1111 | it uses that instead of launching a new one. | 21:57 |
*** chas has joined #openstack-kolla | 21:58 | |
sbezverk | kfox1111: how do you inform compute nodes which subnet to use for pods? | 21:58 |
kfox1111 | they ask the apiserver for a range. | 21:58 |
kfox1111 | the extra flag to kubeadm init tels the apiserver which range to use. | 21:58 |
sbezverk | kfox1111: and api server always gives back the same range? | 21:59 |
kfox1111 | flannel just asks the kubelet then for it. | 21:59 |
kfox1111 | that, I don't know for sure. | 21:59 |
sbezverk | to the same compute node? | 21:59 |
kfox1111 | I would expect so though. | 21:59 |
kfox1111 | no reason to reclaim a range when you think the node will come back. | 21:59 |
kfox1111 | as k8s registers nodes. | 21:59 |
sbezverk | kfox1111: I am not sure but I think it is not always the case and potentially subnet could change | 22:00 |
kfox1111 | flannel with direct etcd, etcd has no idea when a flannel will come and go, so has to reclaim. | 22:00 |
kfox1111 | only way to know for sure would be to read code/test/talk to a dev. | 22:01 |
sbezverk | kfox1111: right | 22:01 |
*** chas has quit IRC | 22:02 | |
*** jmccarthy has left #openstack-kolla | 22:02 | |
*** rwallner has quit IRC | 22:04 | |
kfox1111 | infuriating... | 22:04 |
kfox1111 | some super minor change in logging is breaking the ability to put a fip on a vm... | 22:05 |
*** rwallner has joined #openstack-kolla | 22:05 | |
sbezverk | kfox1111: should not be any relation?!? | 22:05 |
sbezverk | kfox1111: unless in the code they relay on some logging and since it is no longer there, breaks | 22:06 |
kfox1111 | yeah. I then went in and turned it back off on neutron/nova/libvirt. | 22:06 |
kfox1111 | still failing... | 22:06 |
kfox1111 | very weird. | 22:06 |
kfox1111 | trying to figure this one out almost a week. :/ | 22:07 |
kfox1111 | gonna just have to bisect the ps.... :/ | 22:08 |
*** Pavo has quit IRC | 22:09 | |
sbezverk | kfox1111: do you see the same issue locally? | 22:09 |
kfox1111 | haven't tested locally yet. | 22:10 |
sbezverk | kfox1111: some issues are easier to debug locally, connected to the containers.. | 22:11 |
kfox1111 | going to strip it down to the bare bones, then will add patch chunks in half at a time. | 22:12 |
kfox1111 | btw... | 22:12 |
kfox1111 | found this yesterday. very cool: | 22:12 |
kfox1111 | https://grafana.com/dashboards/315 | 22:13 |
kfox1111 | I just launched a prometheus/grafana via helm charts, then pulled that in. :) | 22:13 |
kfox1111 | https://grafana.com/dashboards/917 works really well too. :) | 22:13 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: Docker Logging Support https://review.openstack.org/458649 | 22:15 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: WIP: test1 https://review.openstack.org/459463 | 22:16 |
*** rwallner has quit IRC | 22:17 | |
*** rwallner has joined #openstack-kolla | 22:17 | |
*** trozet has quit IRC | 22:18 | |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: WIP: t2 https://review.openstack.org/459464 | 22:19 |
sbezverk | kfox1111: looks nice.. | 22:20 |
*** trozet has joined #openstack-kolla | 22:20 | |
kfox1111 | yeah. really digging prometheus. | 22:20 |
kfox1111 | they way it works makes it really easy to add extra random data too it, like power metring from random, old, unsupported metered pdu's, or networked thermostats. :) | 22:21 |
kfox1111 | all it takes a lot of times is a simple bash script to glue it together (or whatever else your scripting language of choice is at the time) | 22:24 |
*** rwallner has quit IRC | 22:30 | |
*** rwallner has joined #openstack-kolla | 22:30 | |
rwellum | sbezverk: what's the mistake in the slide? Or did kfox1111 cover it? | 22:31 |
sbezverk | rwellum: for cinder volume the syntax is - 'IP address': 'volume group name' | 22:32 |
sbezverk | on the slide you have - 'interface': 'volume group name' | 22:33 |
sbezverk | which is incorrect | 22:33 |
*** schwicht has quit IRC | 22:36 | |
rwellum | You are correct - thanks! | 22:39 |
*** manheim has quit IRC | 22:39 | |
*** manheim has joined #openstack-kolla | 22:40 | |
*** trozet has quit IRC | 22:42 | |
*** rwallner has quit IRC | 22:43 | |
*** rwallner has joined #openstack-kolla | 22:43 | |
*** goldyfruit has joined #openstack-kolla | 22:46 | |
*** adrian_otto has quit IRC | 22:52 | |
*** rwallner has quit IRC | 22:56 | |
*** rwallner has joined #openstack-kolla | 22:56 | |
rwellum | Otherwise sbezverk - is the picture correct, especially the two NICS I am using? | 22:57 |
sbezverk | I use two nics, in some special cases more, but normally 2 is sufficient | 22:58 |
*** manheim has quit IRC | 23:00 | |
kfox1111 | rwellum: still think you only need the one. | 23:01 |
kfox1111 | hmm.. so t1 passed, and t2 failed... its there somehwere... | 23:02 |
rwellum | That would make it easier, but how - the guide is pretty clear that you need two? | 23:03 |
rwellum | As in neutron needs a dedicated interface, and then you need a mgmt | 23:04 |
inc0 | yay for novaclient | 23:04 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: WIP: test1 https://review.openstack.org/459463 | 23:04 |
inc0 | fresh install of openstack client can't add rules to secgroup | 23:05 |
openstackgerrit | Kevin Fox proposed openstack/kolla-kubernetes master: WIP: test2 https://review.openstack.org/459464 | 23:07 |
kfox1111 | rwellum: depends on connectivity and if you want to run ovs in a container or not. | 23:09 |
kfox1111 | the public network shoudl be bridged to virbr0. | 23:10 |
*** salv-orl_ has quit IRC | 23:10 | |
*** rwallner has quit IRC | 23:10 | |
kfox1111 | that could be done with ovs if its running on the host, or via a linux bridge without. | 23:10 |
*** rwallner has joined #openstack-kolla | 23:11 | |
rwellum | So ovs in a container - need the bridge. | 23:12 |
*** schwicht has joined #openstack-kolla | 23:12 | |
kfox1111 | need an extra bridge. | 23:12 |
*** vhosakot has quit IRC | 23:13 | |
rwellum | Hmm so same considerations bare-metal right? | 23:13 |
kfox1111 | well, not exactly. | 23:14 |
kfox1111 | this is under the constraints that: 1, you have a corperate network, that you don't want (or cant have) the vms directly on, | 23:14 |
*** jascott1 has joined #openstack-kolla | 23:14 | |
kfox1111 | and 2, you will only have 1 physical machine involved. | 23:14 |
kfox1111 | so can be simpler then other setups. | 23:15 |
*** eanylin has quit IRC | 23:16 | |
rwellum | Ok right - that makes sense - I see the point you made earlier about would multiple vm's be on the same server or not. | 23:16 |
jascott1 | rwellum did you see this as well? might be helpful idk https://etherpad.openstack.org/p/knet | 23:16 |
inc0 | ok I'm done with installation of openstack | 23:17 |
jascott1 | woot | 23:17 |
inc0 | took longer than I hoped | 23:17 |
inc0 | tomorrow - vpn server | 23:17 |
rwellum | Hmm jascott1 it's got some good info and some out of date :) | 23:18 |
jascott1 | feel free to add any points | 23:18 |
rwellum | I'll check it's lined up with my slide, thanks for that. | 23:18 |
*** schwicht has quit IRC | 23:19 | |
rwellum | will do | 23:19 |
*** schwicht has joined #openstack-kolla | 23:21 | |
*** hyakuhei has quit IRC | 23:22 | |
jascott1 | rwellum yeah I went thru the guide and took extracted all the references to network stuff so can grok it | 23:22 |
jascott1 | I think some example setups would go a long way | 23:22 |
jascott1 | like diagrams for AIO and also a simple cluster | 23:22 |
kfox1111 | +1. | 23:23 |
kfox1111 | been meaning to work on some reference architecture diagrams. | 23:23 |
kfox1111 | just havent had time. :/ | 23:23 |
jascott1 | maybe between me and rwellum we can take care of it | 23:23 |
rwellum | Well - I like the picture view - because I've made mistake after mistake using sed etc to try the grok. | 23:24 |
kfox1111 | cool. | 23:24 |
*** rwallner has quit IRC | 23:24 | |
*** schwicht has quit IRC | 23:24 | |
rwellum | Agree I'll help jascott1 | 23:24 |
*** rwallner has joined #openstack-kolla | 23:24 | |
jascott1 | awesome | 23:25 |
*** hyakuhei has joined #openstack-kolla | 23:28 | |
*** dave-mccowan has quit IRC | 23:30 | |
*** dixiaoli has quit IRC | 23:35 | |
*** eaguilar has joined #openstack-kolla | 23:36 | |
*** rwallner has quit IRC | 23:37 | |
*** rwallner has joined #openstack-kolla | 23:38 | |
kfox1111 | I don't have too much time before the summit, but do ping me if you would like some review. | 23:39 |
rwellum | I'll be at the summit so no escape kfox1111 :) | 23:40 |
kfox1111 | hehe. fairly warned I be, says I. :) | 23:40 |
*** jrobinson has joined #openstack-kolla | 23:42 | |
*** tonanhngo has quit IRC | 23:43 | |
*** jrobinson has left #openstack-kolla | 23:44 | |
srwilkers | o/ | 23:45 |
jascott1 | what up srwilkers! | 23:46 |
srwilkers | not much dude. how are you? | 23:46 |
jascott1 | good thanks | 23:47 |
jascott1 | yourself? | 23:47 |
srwilkers | not too shabby. just hacking away | 23:49 |
spsurya | morning alll | 23:50 |
jascott1 | yeah im trying to understand this thing called "networking" | 23:50 |
*** iniazi has quit IRC | 23:50 | |
jascott1 | which I used to think I understood in the long long ago | 23:50 |
*** rwallner has quit IRC | 23:50 | |
*** iniazi has joined #openstack-kolla | 23:51 | |
*** rwallner has joined #openstack-kolla | 23:51 | |
rwellum | Overloaded nouns jascott1 - os seems to love that. And naming the same thing differently. | 23:51 |
spsurya | sup jascott1 | 23:51 |
jascott1 | yeah then sprinkle in nomenclature from kubernetes for extra fun | 23:52 |
spsurya | sup rwellum | 23:52 |
jascott1 | hey spsurya! | 23:52 |
rwellum | exactly jascott1 | 23:52 |
rwellum | Good evening spsurya :) | 23:52 |
*** felixtsang has quit IRC | 23:54 | |
*** imcsk8 has quit IRC | 23:54 | |
*** tonanhngo has joined #openstack-kolla | 23:55 | |
jascott1 | rwellum what do you use for diagrams? | 23:56 |
rwellum | Oh no terrible question... | 23:56 |
jascott1 | haha | 23:56 |
rwellum | powerpoint :( | 23:56 |
* jascott1 averts eye contact | 23:57 | |
rwellum | Someone told me about draw.io? | 23:57 |
jascott1 | :) | 23:57 |
rwellum | I know right... Now no-one will talk to me. | 23:57 |
jascott1 | I use plant UML when I can | 23:57 |
jascott1 | but not suitable for everything | 23:58 |
*** chas has joined #openstack-kolla | 23:58 | |
jascott1 | it has a new type 'deployment' that might work for a network diagram | 23:58 |
rwellum | That does look interesting - book marked. | 23:59 |
jascott1 | oh its definitely great for sequence diagrams | 23:59 |
jascott1 | others vary but it rules for those | 23:59 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!