| *** spatel has joined #openstack-ansible | 00:24 | |
| *** spatel has quit IRC | 00:29 | |
| *** cshen has joined #openstack-ansible | 01:29 | |
| *** cshen has quit IRC | 01:33 | |
| *** NewJorg has quit IRC | 01:42 | |
| *** NewJorg has joined #openstack-ansible | 01:44 | |
| *** noonedeadpunk has quit IRC | 01:47 | |
| *** gary_perkins_ has quit IRC | 01:47 | |
| *** gary_perkins has joined #openstack-ansible | 01:48 | |
| *** ChiTo has quit IRC | 02:03 | |
| *** spatel has joined #openstack-ansible | 02:21 | |
| *** cshen has joined #openstack-ansible | 03:29 | |
| *** cshen has quit IRC | 03:34 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-ansible | 04:33 | |
| *** spatel has quit IRC | 04:37 | |
| *** cshen has joined #openstack-ansible | 05:29 | |
| *** cshen has quit IRC | 05:34 | |
| *** cshen has joined #openstack-ansible | 06:00 | |
| *** noonedeadpunk has joined #openstack-ansible | 06:01 | |
| *** miloa has joined #openstack-ansible | 06:03 | |
| *** cshen has quit IRC | 06:04 | |
| *** d34dh0r53 has joined #openstack-ansible | 06:17 | |
| *** d34dh0r53 has quit IRC | 06:17 | |
| *** janno has quit IRC | 06:51 | |
| *** janno has joined #openstack-ansible | 06:51 | |
| *** MickyMan77 has quit IRC | 07:07 | |
| *** shyamb has joined #openstack-ansible | 07:13 | |
| *** cshen has joined #openstack-ansible | 07:18 | |
| *** andrewbonney has joined #openstack-ansible | 07:28 | |
| *** tosky has joined #openstack-ansible | 07:35 | |
| *** shyam89 has joined #openstack-ansible | 07:41 | |
| *** jbadiapa has joined #openstack-ansible | 07:43 | |
| *** shyamb has quit IRC | 07:43 | |
| *** tosky has quit IRC | 08:05 | |
| *** tosky has joined #openstack-ansible | 08:10 | |
| *** shyamb has joined #openstack-ansible | 08:11 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_senlin master: Initial commit to os_senlin https://review.opendev.org/749365 | 08:12 |
|---|---|---|
| *** cshen has quit IRC | 08:13 | |
| *** tosky has quit IRC | 08:14 | |
| *** shyam89 has quit IRC | 08:14 | |
| *** tosky has joined #openstack-ansible | 08:14 | |
| *** cshen has joined #openstack-ansible | 08:15 | |
| *** cshen has quit IRC | 08:18 | |
| *** cshen has joined #openstack-ansible | 08:20 | |
| *** SecOpsNinja has joined #openstack-ansible | 09:28 | |
| *** tosky has quit IRC | 09:39 | |
| *** tosky has joined #openstack-ansible | 09:55 | |
| *** shyamb has quit IRC | 10:13 | |
| *** shyamb has joined #openstack-ansible | 10:22 | |
| *** shyam89 has joined #openstack-ansible | 10:33 | |
| *** shyamb has quit IRC | 10:35 | |
| *** shyam89 has quit IRC | 10:47 | |
| *** shyamb has joined #openstack-ansible | 10:50 | |
| *** cshen has quit IRC | 10:52 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_nova master: Bump libvirt version to prevent compute failure https://review.opendev.org/750320 | 10:57 |
| noonedeadpunk | jrosser: ^ this nasty thing made nova-compute fail for centos 7.... | 10:58 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_nova stable/ussuri: Bump libvirt version to prevent compute failure https://review.opendev.org/750321 | 10:58 |
| *** dave-mccowan has joined #openstack-ansible | 10:59 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_swift stable/ussuri: Delegate gnocchi retrievement task to setup host https://review.opendev.org/750169 | 10:59 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_glance stable/ussuri: Fix native service path https://review.opendev.org/750139 | 11:00 |
| *** cshen has joined #openstack-ansible | 11:03 | |
| jrosser | noonedeadpunk: you missing a link - what is [1] ? | 11:05 |
| noonedeadpunk | I am:( | 11:06 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_nova master: Bump libvirt version to prevent compute failure https://review.opendev.org/750320 | 11:06 |
| noonedeadpunk | http://paste.openstack.org/show/797571/ | 11:06 |
| *** dave-mccowan has quit IRC | 11:12 | |
| *** mgariepy has quit IRC | 11:15 | |
| *** dave-mccowan has joined #openstack-ansible | 11:15 | |
| *** stuartgr has quit IRC | 11:31 | |
| *** stuartgr has joined #openstack-ansible | 11:32 | |
| *** shyamb has quit IRC | 11:43 | |
| *** shyamb has joined #openstack-ansible | 11:45 | |
| *** MickyMan77 has joined #openstack-ansible | 11:54 | |
| *** mathlin has joined #openstack-ansible | 11:57 | |
| *** mgariepy has joined #openstack-ansible | 12:05 | |
| openstackgerrit | Jay Jahns proposed openstack/openstack-ansible-os_neutron stable/ussuri: Add Initial NSX Integration https://review.opendev.org/750328 | 12:08 |
| openstackgerrit | Jay Jahns proposed openstack/openstack-ansible-os_nova stable/ussuri: Set Bridge Information for NSX Integration https://review.opendev.org/750330 | 12:08 |
| *** shyamb has quit IRC | 12:11 | |
| *** rh-jelabarre has joined #openstack-ansible | 12:12 | |
| *** mathlin has quit IRC | 12:12 | |
| *** redrobot has joined #openstack-ansible | 12:17 | |
| *** cshen has quit IRC | 12:18 | |
| *** cshen has joined #openstack-ansible | 13:03 | |
| MickyMan77 | I have problem to get the HAProxy to work, I'm not able to ping the VIP address. When I check the net stat I can see the vip address on all controller nodes. | 13:11 |
| MickyMan77 | [root@controller01 ~]# netstat -pan | grep -I 443tcp 0 0 10.26.13.254:443 0.0.0.0:* LISTEN 143691/haproxytcp 0 0 10.26.11.254:443 0.0.0.0:* LISTEN 143691/haproxy | 13:11 |
| MickyMan77 | --- | 13:11 |
| MickyMan77 | [root@controller01-for ~]# netstat -pan | grep -i 443 | 13:11 |
| MickyMan77 | tcp 0 0 10.26.13.254:443 0.0.0.0:* LISTEN 143691/haproxy | 13:12 |
| MickyMan77 | tcp 0 0 10.26.11.254:443 0.0.0.0:* LISTEN 143691/haproxy | 13:12 |
| MickyMan77 | --- | 13:13 |
| MickyMan77 | arp -n | grep 254 | 13:13 |
| MickyMan77 | 10.26.13.254 (incomplete) br-mgmt | 13:13 |
| miloa | @MikyMan77 If the vip address is on all controllers nodes it means (I think) that keepalived cannot check if the other nodes are up and as they are saw as down it bring up the vip on each node. Perhaps check if each node can ping each other on its main address. | 13:38 |
| MickyMan77 | the nodes can ping the main ip address on all nodes and also ping all container ip addresses via the br-mgmt Nic. | 13:47 |
| MickyMan77 | but not the HAProxy vip adress. | 13:47 |
| *** d34dh0r53 has joined #openstack-ansible | 13:56 | |
| *** cshen has quit IRC | 13:59 | |
| miloa | Did you check your log to see if there is any messages of keepalived ? | 13:59 |
| MickyMan77 | hmm, | 14:03 |
| MickyMan77 | tail keepalived-notifications.log | 14:03 |
| MickyMan77 | 2020-09-08 11:19:30 Trying to restart haproxy to get out of faulty state | 14:03 |
| *** MickyMan77 has quit IRC | 14:24 | |
| *** cshen has joined #openstack-ansible | 14:28 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_keystone master: Define condition for the first play host one time https://review.opendev.org/749649 | 14:33 |
| *** cshen has quit IRC | 14:33 | |
| *** cshen has joined #openstack-ansible | 14:33 | |
| *** spatel has joined #openstack-ansible | 14:36 | |
| *** spatel has quit IRC | 14:44 | |
| *** spatel has joined #openstack-ansible | 15:03 | |
| SecOpsNinja | jrosser, spatel, today i was able to put lets encrypt cert in my openstack haproxy and now magnum stoped complaining about CA cert!! What i did was creating a br-osa-dmz in infra host with static ip that corresponded to external_lb_vip_address. Now, because i want to use that network for floating ips, i added the same network to our compute host (with no ip) and configured the flat provider_ | 15:14 |
| SecOpsNinja | network to use host_bind_override: "br-osa-dmz". This was working fine, until i added this network in openstack as a external provider and create a router that connect to it. Atm the ip that was associated to the br-osa-dmz, switched to brq14a7b7c8-6e and now i can't access the openstackl public endpoint. i supose that this isn't a problem and i did something wrong but i can't understand wha | 15:14 |
| SecOpsNinja | t... | 15:14 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_keystone master: Define condition for the first play host one time https://review.opendev.org/749649 | 15:20 |
| *** cshen has quit IRC | 15:32 | |
| noonedeadpunk | let's skip our meeting today if there's no objections? | 16:00 |
| *** cshen has joined #openstack-ansible | 16:01 | |
| spatel | works for me | 16:02 |
| spatel | noonedeadpunk: look like very close on this patch https://review.opendev.org/#/c/749365/ | 16:03 |
| spatel | jrosser: did your senlin stuff working after fixing our last auth_url patch? | 16:04 |
| *** cshen has quit IRC | 16:06 | |
| *** mgariepy has quit IRC | 16:09 | |
| *** miloa has quit IRC | 16:11 | |
| noonedeadpunk | spatel: I posted patch to senlin https://review.opendev.org/#/c/749874/ which should cover found issue | 16:25 |
| noonedeadpunk | but yeah, it's supposed to work with self-signed ssls | 16:25 |
| spatel | noonedeadpunk: that is cool! | 16:26 |
| spatel | self-sign SSL isn't fun :) | 16:27 |
| noonedeadpunk | it is for CI envs especially) | 16:27 |
| spatel | noonedeadpunk: why i am not seeing any centos-7 builds here - https://review.opendev.org/#/c/749365/ | 16:28 |
| spatel | it was there last week | 16:28 |
| noonedeadpunk | because we've dropped all centos 7 jobs for master | 16:28 |
| noonedeadpunk | as we've planned to drop it for V | 16:29 |
| spatel | V ? | 16:30 |
| noonedeadpunk | Victoria - next openstack release | 16:30 |
| spatel | ah! so we are not going to continue c7 anymore right? | 16:31 |
| noonedeadpunk | yep, we didn't. There're so many nasty workarounds for centos7 to make it work with py3... | 16:31 |
| noonedeadpunk | And distro packages has been dropped for train, so no distro path even for ussuri | 16:31 |
| noonedeadpunk | and ceph is barely supported as well... | 16:32 |
| spatel | +1 | 16:32 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add centos-8 support https://review.opendev.org/739646 | 16:32 |
| spatel | so i am running pike on my production with c7 in that case i can't upgrade to latest until i upgrade c8 right? | 16:33 |
| noonedeadpunk | so you're supposed to update to ussuri, then resetup to centos 8 and you can move forward | 16:33 |
| spatel | got it | 16:39 |
| *** renich has joined #openstack-ansible | 17:07 | |
| *** mgariepy has joined #openstack-ansible | 17:10 | |
| *** renich has quit IRC | 17:29 | |
| *** gyee has joined #openstack-ansible | 17:33 | |
| *** mgariepy has quit IRC | 17:37 | |
| *** mgariepy has joined #openstack-ansible | 17:38 | |
| *** carlosmss has joined #openstack-ansible | 17:38 | |
| carlosmss | Hi guys, someone seen this error with RabbitMQ, I've searched about this issue but no success. My scenario is a cluster AMPQ deployed with Openstack-Ansible.: AMQP server on xxx.xx.xx.xx:5672 is unreachable. WARNING oslo.messaging._drivers.impl_rabbit [-] Unexpected error during heartbeat thread processing, retrying...: ConnectionForced: Too many heartbeats missed | 17:44 |
| *** MickyMan77 has joined #openstack-ansible | 17:51 | |
| MickyMan77 | I'm not able to get the keepalived to work with network bridge, it will not add the VIP ip address. If I change the keepalived config so I use ethX interface instead of br-mgmt it works perfect. | 18:01 |
| MickyMan77 | I'm using centos 8 and the br-mgmt is port of a bond. | 18:02 |
| MickyMan77 | *part of a bond | 18:02 |
| noonedeadpunk | bridge can't be bond port actually... bond can be one of the interfaces on the bridge | 18:03 |
| noonedeadpunk | and keepalive adds ip address o nthe interface when it has another ip from the same network iirc (may be wrong) | 18:04 |
| noonedeadpunk | I guess I didn't test keepalived with centos 8 on multinode setup though.... | 18:05 |
| noonedeadpunk | spatel: did you have any issues in your lab with several centos8 controllers and keepalived? | 18:05 |
| spatel | noonedeadpunk: no | 18:06 |
| spatel | i am running 3 node controller and haven't seen any issue so far | 18:06 |
| *** MickyMan77 has quit IRC | 18:07 | |
| spatel | This is my keepalived and haven't seen any issue so far - http://paste.openstack.org/show/797603/ | 18:10 |
| spatel | in my keepalived its using br-mgmt interface and no issue so far. | 18:11 |
| noonedeadpunk | thanks for the info:) | 18:12 |
| noonedeadpunk | track_script looks weird though | 18:13 |
| *** mgariepy has quit IRC | 18:14 | |
| spatel | noonedeadpunk: yes, not sure why and how its working then | 18:25 |
| noonedeadpunk | have you tested failover? | 18:25 |
| noonedeadpunk | but it's more about check scripts.... | 18:25 |
| spatel | let me test failover and see | 18:26 |
| noonedeadpunk | It looks like instead of list of scripts you have only single one... | 18:27 |
| spatel | hmm | 18:28 |
| *** mgariepy has joined #openstack-ansible | 18:28 | |
| noonedeadpunk | but actually by default it has to be list, and you have content as list as well.... | 18:29 |
| noonedeadpunk | dict_keys..... hm | 18:29 |
| noonedeadpunk | ah damn it | 18:29 |
| noonedeadpunk | spatel: can you try set `keepalived_scripts.keys() | list` for track_scripts in openstack_ansible/inventory/group_vars/haproxy/keepalived.yml and re-run playbook? | 18:30 |
| spatel | let me check | 18:31 |
| spatel | This is what i have track_scripts: "{{ keepalived_scripts.keys() }}" | 18:32 |
| noonedeadpunk | yeah, just add `| list` filter | 18:33 |
| spatel | you want me to put "| list" ? | 18:33 |
| spatel | ok doing it | 18:33 |
| spatel | which playbook you want me to run? | 18:34 |
| noonedeadpunk | haproxy-install | 18:37 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Fix keys() method output https://review.opendev.org/750473 | 18:37 |
| spatel | running... | 18:37 |
| spatel | noonedeadpunk: neat - http://paste.openstack.org/show/797604/ | 18:39 |
| noonedeadpunk | yeah, nice | 18:41 |
| noonedeadpunk | thanks so much for testing - saving me tons of time! | 18:42 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_manila master: Add centos-8 support https://review.opendev.org/739646 | 18:42 |
| noonedeadpunk | https://review.opendev.org/#/c/750473/ is going to cover that | 18:42 |
| spatel | +1 | 18:43 |
| spatel | noonedeadpunk: fyi, keepalived failover is working | 18:50 |
| *** viks____ has quit IRC | 18:58 | |
| *** d34dh0r53 has quit IRC | 19:08 | |
| *** d34dh0r53 has joined #openstack-ansible | 19:10 | |
| *** spatel has quit IRC | 19:18 | |
| *** SecOpsNinja has left #openstack-ansible | 19:21 | |
| *** cshen has joined #openstack-ansible | 19:28 | |
| *** cshen has quit IRC | 19:32 | |
| *** MickyMan77 has joined #openstack-ansible | 20:01 | |
| MickyMan77 | @noonedeadpunk the bridge is config like this. | 20:05 |
| MickyMan77 | bond0 have this config | 20:05 |
| MickyMan77 | BRIDGE=br-mgmt | 20:05 |
| MickyMan77 | the nic have this, | 20:06 |
| MickyMan77 | SLAVE=yesMASTER=bond_mgt0 | 20:06 |
| MickyMan77 | and the br-mgmt is working perfect. | 20:06 |
| openstackgerrit | Merged openstack/openstack-ansible master: Fix neutron-server default serial https://review.opendev.org/746546 | 20:06 |
| MickyMan77 | but I'm not able to add the VIP IP address via keepalived. | 20:07 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Define condition for the first play host one time https://review.opendev.org/749649 | 20:17 |
| jrosser | MickyMan77: keepalived uses multicast between the nodes to decide which node has the VIP | 20:28 |
| jrosser | if you use tcpdump on the different interfaces you should be able to see that from all the nodes | 20:28 |
| noonedeadpunk | unless it's blocked or not allowed, yeah | 20:28 |
| jrosser | if that traffic does not make it then things will be broken | 20:29 |
| MickyMan77 | the logs say this. | 20:31 |
| MickyMan77 | VRRP_Group(haproxy): Syncing instances to FAULT state | 20:31 |
| MickyMan77 | (external) entering FAULT state | 20:31 |
| MickyMan77 | (internal) entering FAULT state | 20:31 |
| MickyMan77 | (external) removing VIPs. | 20:32 |
| MickyMan77 | (internal) removing VIPs. | 20:32 |
| jrosser | ok but it will be necessary to debug one level deeper | 20:34 |
| jrosser | to look at the traffic on each node and see if it is as we should expect | 20:34 |
| jrosser | also, as this is centos i think it would be a good idea to check the bridge forwarding, becasue long long ago we had to fix this for Centos AIO builds https://github.com/openstack/ansible-role-systemd_networkd/commit/242b3c3fb4a3f8ebeba5c24cf8c7a510cba14143 | 20:34 |
| jrosser | and also a snaity check, this is actual servers and not vmwaare or other virtualisation? | 20:36 |
| MickyMan77 | this is ibm servers, x3650 and x3850 | 20:38 |
| *** andrewbonney has quit IRC | 20:39 | |
| jrosser | ok, thats fine | 20:40 |
| jrosser | OSA does not set up the host networking, and in the past there has been issues with centos bridges not forwarding by default, so that is worth checking | 20:40 |
| noonedeadpunk | fwiw I have the follwoing config of interfaces for ubuntu with vip 172.20.0.9 http://paste.openstack.org/show/797606/ | 20:41 |
| jrosser | then tcpdump on the bridge and eth interfaces to see if you can find the vrrp traffic would be the next step | 20:41 |
| jrosser | "If a configured script returns a non-zero exit code f times in succession, Keepalived changes the state of the VRRP instance or group to FAULT" | 20:45 |
| jrosser | noonedeadpunk: is it this? https://review.opendev.org/750473 | 20:46 |
| noonedeadpunk | um, it's just fixing check scripts for keepalived. it didn't fail for spatel though.... | 20:48 |
| noonedeadpunk | I mean without it | 20:48 |
| noonedeadpunk | so not sure it it will help here | 20:48 |
| jrosser | the track scripts also check for ping of an upstream thing | 20:49 |
| jrosser | so i guess that being non-contactable is going to cause a transition to FAULT state | 20:50 |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_cinder master: Use cinder_service_setup_host for qos and types creation https://review.opendev.org/750491 | 20:50 |
| noonedeadpunk | btw that is pretty nasty bug I think ^ | 20:50 |
| noonedeadpunk | and worth pending time to implement these options in openstack collection.... | 20:50 |
| noonedeadpunk | *spending | 20:50 |
| *** cshen has joined #openstack-ansible | 20:56 | |
| *** spatel has joined #openstack-ansible | 20:56 | |
| *** cshen has quit IRC | 21:00 | |
| openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible-os_cinder master: Set correct permissions for rootwrap.d https://review.opendev.org/750493 | 21:05 |
| noonedeadpunk | mgariepy: jamesdenton seems ovn is failing now for some reason ;( https://review.opendev.org/#/c/733017 | 21:08 |
| jamesdenton | htmm | 21:09 |
| noonedeadpunk | on tempest networking https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_ec6/733017/23/check/openstack-ansible-deploy-aio_ovn_lxc-ubuntu-focal/ec6b241/logs/openstack/aio1_utility_container-e55222e2/utility/stestr_results.html | 21:09 |
| jamesdenton | cool, wanted to make sure i was on the right track there. | 21:10 |
| noonedeadpunk | but I think it started failing after my rebase.... | 21:11 |
| noonedeadpunk | so might be we merged smth breaking...... | 21:11 |
| noonedeadpunk | (we shouldn't though) | 21:11 |
| jamesdenton | lemme spin one up and see | 21:12 |
| MickyMan77 | @jrosser will this found the vrrp traffic ?, # tcpdump -vvv -n -i any host 224.0.0.18 | 21:18 |
| jrosser | i just tried this tcpdump -i br-mgmt vrrp | 21:19 |
| jrosser | and i see a bunch of 20:54:39.142988 IP 10.11.128.21 > vrrp.mcast.net: VRRPv2, Advertisement, vrid 111, prio 100, authtype simple, intvl 1s, length 20 | 21:21 |
| jrosser | which is the ip of the other haproxy (i have two in this particular case) | 21:21 |
| MickyMan77 | completely empty, tcpdump does not show anything. (I also have restated the keepalived service) | 21:21 |
| * jrosser late here, enough for today | 21:21 | |
| spatel | MickyMan77: what is the problem with keepalived? | 21:29 |
| spatel | failover issue? | 21:29 |
| MickyMan77 | @spatel it will not add VIP address on the br-mgmt | 21:30 |
| spatel | in last 3 years i never troubleshoot keepalived, because it just work without any issue. i am running 4 openstack cluster and non has any issue | 21:30 |
| MickyMan77 | what os do you run on ? | 21:30 |
| spatel | centos7 and centos8 | 21:31 |
| MickyMan77 | hmm, using centos8 | 21:31 |
| MickyMan77 | *i'm using | 21:31 |
| spatel | last week i build 3 controller node lab and didn't see any issue | 21:31 |
| spatel | i am assuming something is missing in config side. | 21:32 |
| MickyMan77 | can you give me a copy of your, openstack_user_config.yml and user_variables.yml | 21:33 |
| spatel | why don't you post your openstack_user_config.yml and user_variable.yml file | 21:33 |
| MickyMan77 | where can I post it ? | 21:34 |
| spatel | http://paste.openstack.org/show/797612/ | 21:34 |
| spatel | http://paste.openstack.org is your friend to share snippet | 21:35 |
| MickyMan77 | thx | 21:35 |
| spatel | http://paste.openstack.org/show/797613/ | 21:36 |
| spatel | i have posted my both file.. just verify your config snippet and you should be good. | 21:36 |
| *** jbadiapa has quit IRC | 21:41 | |
| *** d34dh0r53 has quit IRC | 21:42 | |
| *** d34dh0r53 has joined #openstack-ansible | 21:42 | |
| MickyMan77 | @spatel how have you. config the bridge, br-host br-mgmt | 21:45 |
| spatel | In production i have bond0 with LACP | 21:46 |
| spatel | create br-mgmt and attached with bond0.XX vlan id | 21:47 |
| spatel | MickyMan77: something like this - http://paste.openstack.org/show/797614/ | 21:48 |
| spatel | if you are using systemd-networkd style network in that case configuration will be little different. | 21:50 |
| MickyMan77 | NM_CONTROLLED=no I use NetworkManager, is that ok ? | 21:50 |
| spatel | I hate NetworkManager :( | 21:51 |
| spatel | you have to show your network config otherwise its hard to answer whatever question you are asking | 21:53 |
| spatel | systemd-networks vs legacy way | 21:54 |
| MickyMan77 | network settings, http://paste.openstack.org/show/797615/ | 21:57 |
| spatel | MickyMan77: looks good | 22:08 |
| spatel | if you able to ping other end then you should be good | 22:09 |
| spatel | i am not seeing any issue here | 22:09 |
| spatel | does ifcfg-mgt0 and ifcfg-mgt1 coming from same switch? | 22:10 |
| MickyMan77 | yes | 22:10 |
| MickyMan77 | they are connected to a Cisco 6500 | 22:10 |
| spatel | ok | 22:10 |
| MickyMan77 | and I can ping all nodes and container | 22:10 |
| spatel | can you ping other controller node? | 22:10 |
| MickyMan77 | the only thing that not work is the vip adress. | 22:11 |
| spatel | is br-mgmt is your external endpoint? | 22:11 |
| spatel | in my case i have br-host for external and br-mgmt for internal vip | 22:11 |
| spatel | br-mgmt isn't routable to outside world (its purely for openstack control plane) | 22:12 |
| spatel | not sure how config will look like when we have only br-mgmt and not br-host | 22:13 |
| MickyMan77 | our openstack installation is only for internal use. | 22:13 |
| spatel | Same here, we are running in private datacenter | 22:14 |
| MickyMan77 | and we will only use the br-mgmt for now.. | 22:14 |
| spatel | may be that is your problem | 22:14 |
| spatel | I use br-host and br-mgmt to all my deployment and haven't seen any issue so far.. | 22:15 |
| -spatel- haproxy_keepalived_external_interface: br-host | 22:16 | |
| -spatel- haproxy_keepalived_internal_interface: br-mgmt | 22:16 | |
| MickyMan77 | I will check that tomorrow. | 22:16 |
| MickyMan77 | need to sleep now.. | 22:16 |
| spatel | see this is what i have, so in your case it will be br-mgmt on both external/internal | 22:16 |
| spatel | MickyMan77: have a good night :) | 22:16 |
| spatel | i gotta go too | 22:16 |
| MickyMan77 | thx for the help | 22:17 |
| *** spatel has quit IRC | 22:18 | |
| *** tosky has quit IRC | 22:55 | |
| *** cshen has joined #openstack-ansible | 22:56 | |
| *** cshen has quit IRC | 23:01 | |
| *** carlosmss has quit IRC | 23:05 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!