*** waxfire has quit IRC | 00:00 | |
*** waxfire has joined #openstack-ansible | 00:00 | |
*** jeh has joined #openstack-ansible | 00:12 | |
*** cheng1 has quit IRC | 00:39 | |
*** cheng1 has joined #openstack-ansible | 00:40 | |
*** jeh has quit IRC | 01:07 | |
*** jeh has joined #openstack-ansible | 01:18 | |
*** jeh has quit IRC | 01:43 | |
*** jeh has joined #openstack-ansible | 01:43 | |
*** jeh has quit IRC | 01:53 | |
*** markvoelker has joined #openstack-ansible | 02:14 | |
*** markvoelker has quit IRC | 02:19 | |
*** mmethot has joined #openstack-ansible | 02:24 | |
*** mmethot_ has quit IRC | 02:25 | |
*** markvoelker has joined #openstack-ansible | 04:05 | |
*** markvoelker has quit IRC | 04:09 | |
*** evrardjp has quit IRC | 04:33 | |
*** evrardjp has joined #openstack-ansible | 04:33 | |
*** markvoelker has joined #openstack-ansible | 06:06 | |
*** jcath has joined #openstack-ansible | 06:09 | |
*** markvoelker has quit IRC | 06:10 | |
*** stingrayza has joined #openstack-ansible | 06:21 | |
jcath | jrosser: a little confused about the br-xxx and bonding in a multi-nodes deployment. should I have to create the bonding and bridges manually before running the setup-hosts.yml ? or the playbooks will create them automatically? thanks | 06:24 |
---|---|---|
*** also_stingrayza has quit IRC | 06:25 | |
*** ianychoi has joined #openstack-ansible | 06:30 | |
noonedeadpunk | generally we don't configure networking, unless that role has been run https://opendev.org/openstack/openstack-ansible/src/branch/master/tests/roles/bootstrap-host | 08:21 |
noonedeadpunk | It's generally triggered with bootstrap-aio.sh script | 08:21 |
noonedeadpunk | I never look into mnaio, so not sure if this role is supposed to be run there... | 08:22 |
-openstackstatus- NOTICE: The opendev specific CentoOS and openSUSE mirror disappeared and thus CentOS and openSUSE jobs are all broken. | 08:35 | |
*** jcath has quit IRC | 08:41 | |
*** jawad_axd has joined #openstack-ansible | 09:07 | |
*** arkan has joined #openstack-ansible | 09:12 | |
*** jcath has joined #openstack-ansible | 09:13 | |
arkan | Hi guys, I'm getting in cinder-volumes-container this error cinder.exception.ClusterNotFound: Cluster {'name': 'ceph@rbd'} could not be found, my config storage is https://pastebin.com/QtgPaRDS and the full error in container is https://pastebin.com/CmQij1rN | 09:25 |
arkan | I will try to change volume_backend_name: rbd in openstack_user_config.yml to volume_backend_name: ceph, as I can see from this page (although I'm not sure) https://docs.ceph.com/docs/mimic/rbd/rbd-openstack/#configuring-cinder | 09:28 |
*** jcath has quit IRC | 09:29 | |
*** markvoelker has joined #openstack-ansible | 09:35 | |
*** markvoelker has quit IRC | 09:44 | |
arkan | humm, It worked, the solution is https://pastebin.com/sDaLStgA | 09:48 |
arkan | now the /etc/cinder/cinder.conf has ini group [ceph] instead of [rbd] and volume_backend_name: ceph instead of volume_backend_name: rbd | 09:49 |
*** also_stingrayza has joined #openstack-ansible | 10:20 | |
*** stingrayza has quit IRC | 10:23 | |
*** tosky has joined #openstack-ansible | 10:59 | |
jrosser | arkan: the backend name in cinder.conf is simply that, just a name. it being [ceph] as you have set it is purely co-incidence that you are using ceph | 11:01 |
arkan | jrosser: then what solve the issue is volume_backend_name, I changed it from 'rbd' to 'ceph' | 11:02 |
arkan | now I can see that the vms can create volumes without errors | 11:03 |
arkan | I mean I can launch vms with volumes wihtout erros | 11:03 |
arkan | before the modification that I've done I had problems, and I paste it above | 11:04 |
arkan | now I'm investigating why I can not ping my floating ip | 11:05 |
arkan | I can ping from vm1 to vm2 and inverse | 11:05 |
arkan | I can access the internet | 11:05 |
arkan | the communication between the vms through vxlan and it's working | 11:05 |
arkan | floating ip is assigned and it has security groups icmp and ssh (ingress) + default security group added | 11:06 |
arkan | I'm investigating right now | 11:06 |
*** jawad_axd has quit IRC | 11:39 | |
*** jcath has joined #openstack-ansible | 11:41 | |
admin0 | arkan, login to the router namespace .. check if you can ping your instances private ip from there, and also if you can ping your router | 11:45 |
arkan | admin0: ok | 11:45 |
admin0 | if you can ping your private ips from within the namespace, but not the public, then you know its north-side issue and could most probably due to routing or vlan | 11:45 |
jcath | noonedeadpunk: thanks. the bootstrap-host seems to be only trigged with bootstrap-aio ... so, maybe have to create br-mgmt/br-xxxx bridges at every host in a production deployment? | 11:48 |
arkan | root@controller1:~# ip netns exec qrouter-21b624c7-261b-4a88-adf6-8321d64d06c5 ping 8.8.8.8 | 11:59 |
arkan | PING 8.8.8.8 (8.8.8.8) 56(84) bytes of data. | 11:59 |
arkan | 64 bytes from 8.8.8.8: icmp_seq=1 ttl=117 time=14.9 ms | 11:59 |
arkan | admin0: I can ping the router from my vms | 12:00 |
admin0 | ok | 12:00 |
admin0 | can you ping your local instanecs ? | 12:00 |
admin0 | instances* | 12:00 |
arkan | but I can not ping my vm from the router | 12:00 |
arkan | root@controller1:~# ip netns exec qrouter-21b624c7-261b-4a88-adf6-8321d64d06c5 ping 10.0.2.238 | 12:00 |
admin0 | iptables -L -nvx -t nat --line-numbers for a good overview of the iptables setup there | 12:01 |
admin0 | login to the namespace using bash .. and then check what the issue could be .. could be a layer2 issue between vxlan interfaces | 12:01 |
admin0 | of where the namespace is and where your instance it | 12:01 |
admin0 | is* | 12:01 |
arkan | https://pastebin.com/qR9sk57e | 12:03 |
arkan | there are packets for snat and dnat | 12:03 |
arkan | for 192.168.40.213 (my floating ip) | 12:03 |
arkan | to/from my vm 10.0.0.215 | 12:04 |
arkan | 192.168.40.0/24 is under vlan 40 configured that has access to the internet through eno1 (physical interface) | 12:05 |
arkan | my instances are in the other node (compute1) | 12:08 |
arkan | so controller1 communicates with compute1 | 12:09 |
arkan | compute1 has nova + ceph | 12:09 |
arkan | compute1 has nova + ceph (osds) | 12:09 |
arkan | controller1 has infra + ceph mons, mgrs + ceph (osds) | 12:10 |
*** sshnaidm_ has joined #openstack-ansible | 12:34 | |
*** sshnaidm|afk has quit IRC | 12:34 | |
arkan | ok, I added security group (default + icmp_ssh) to the port 10.0.0.1 of the router interface (for internal network), and now I can ping vms from the router netns | 12:58 |
arkan | just I don't know if this is correct, from horizon it created floating IP automatically, but I can see that it's /32 not /24 | 13:00 |
arkan | ip netns exec qrouter-21b624c7-261b-4a88-adf6-8321d64d06c5 ip addr | 13:00 |
arkan | inet 192.168.40.213/32 brd 192.168.40.213 scope global qg-5fbfed04-35 | 13:00 |
arkan | my network should be 192.168.40.0/24 | 13:01 |
admin0 | arkan, maybe try to validate it against this example: https://www.openstackfaq.com/openstack-add-floating-ips/ | 13:02 |
admin0 | its in the subnet definition | 13:02 |
arkan | admin0: thanks I will check | 13:03 |
*** tosky has quit IRC | 13:17 | |
*** tosky has joined #openstack-ansible | 13:17 | |
arkan | ok, I added security group to the provider port for network:router_gateway 192.168.40.8, I can see 4 items with name 'default', I don't know why, in my security group there is only 1 'default' and the one that I added 'icmp_ssh' | 13:38 |
arkan | anyway, I've added all the 'default' items to network:router_gateway 192.168.40.8 port in the provider network | 13:38 |
arkan | and now I can ping my floating ip from the router netns | 13:39 |
-openstackstatus- NOTICE: Package mirrors should be back in working order; any jobs which logged package retrieval failures between 19:35 UTC yesterday and 13:20 UTC today can be safely rechecked | 13:40 | |
arkan | so basicly vms can access internet and can ping each other and can ping router | 13:41 |
arkan | router can ping vms and ping floating ip | 13:41 |
arkan | vms can ping floating ips | 13:42 |
arkan | but my local machine can not ping the floating ip | 13:42 |
arkan | now I'm investigating | 13:42 |
arkan | ok, I put tcpdump inside the interface of the router | 13:46 |
arkan | tcpdump -i qg-5fbfed04-35 port not 22 | 13:46 |
arkan | and I pinged floating ip from my macbook, I can see that it received the ECHO request | 13:47 |
arkan | IP 192.168.1.125 > 192.168.40.155: ICMP echo request, id 53120, seq 1, length 64 | 13:47 |
arkan | I need to check if the vm is recieving it | 13:47 |
arkan | but there is no tcpdump on cirros image | 13:48 |
arkan | I need to use another image | 13:48 |
*** aedc has joined #openstack-ansible | 14:49 | |
*** aedc has quit IRC | 14:50 | |
*** aedc has joined #openstack-ansible | 14:55 | |
*** aedc has quit IRC | 14:57 | |
*** aedc has joined #openstack-ansible | 14:58 | |
*** aedc has quit IRC | 15:01 | |
*** aedc has joined #openstack-ansible | 15:02 | |
*** aedc has quit IRC | 15:03 | |
*** aedc has joined #openstack-ansible | 15:06 | |
*** aedc has quit IRC | 15:10 | |
*** aedc has joined #openstack-ansible | 15:10 | |
arkan | now I can ping my floating ip | 15:12 |
*** aedc has quit IRC | 15:12 | |
arkan | I've deleted all the security groups associated with ports from the provider network, and I added floating ip to the instance with the security groups assigned to the instance | 15:13 |
arkan | and now I can ping it | 15:13 |
arkan | :)) | 15:13 |
arkan | great, thanks to all of you guys | 15:13 |
*** jcath has quit IRC | 15:14 | |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible master: Bump SHAs for master https://review.opendev.org/735467 | 15:21 |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/stein: Bump SHAs for stable/stein https://review.opendev.org/735468 | 15:21 |
openstackgerrit | Dmitriy Rabotyagov (noonedeadpunk) proposed openstack/openstack-ansible stable/train: Bump SHAs for stable/train https://review.opendev.org/735469 | 15:22 |
*** arkan has quit IRC | 16:02 | |
*** markvoelker has joined #openstack-ansible | 17:29 | |
*** markvoelker has quit IRC | 17:34 | |
*** markvoelker has joined #openstack-ansible | 17:34 | |
*** markvoelker has quit IRC | 17:39 | |
openstackgerrit | Merged openstack/openstack-ansible-galera_server master: Add Centos-8 support https://review.opendev.org/724721 | 17:56 |
noonedeadpunk | cores, would be great to get another review for https://review.opendev.org/#/c/734888/13 | 18:03 |
noonedeadpunk | if anyone around | 18:03 |
noonedeadpunk | this blocks recovery of upgrade jobs | 18:04 |
*** jawad_axd has joined #openstack-ansible | 18:35 | |
*** jawad_axd has quit IRC | 18:36 | |
*** markvoelker has joined #openstack-ansible | 18:55 | |
*** markvoelker has quit IRC | 19:00 | |
*** arkan has joined #openstack-ansible | 19:28 | |
*** arkan has quit IRC | 20:52 | |
*** markvoelker has joined #openstack-ansible | 20:56 | |
*** markvoelker has quit IRC | 21:00 | |
*** markvoelker has joined #openstack-ansible | 22:57 | |
*** markvoelker has quit IRC | 23:02 | |
*** markvoelker has joined #openstack-ansible | 23:24 | |
*** tosky has quit IRC | 23:26 | |
*** markvoelker has quit IRC | 23:29 | |
*** mrda has joined #openstack-ansible | 23:57 | |
*** markvoelker has joined #openstack-ansible | 23:59 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!