Keepalived 1.2.7 is bad version. Please, see comments in this bug https://bugs.launchpad.net/neutron/+bug/1497272. I suggest you to try one of the latest version of Keepalived.
On Wed, Jun 22, 2016 at 5:03 PM, fabrice grelaud < fabrice.grel...@u-bordeaux.fr> wrote: > Hi, > > keepalived 1:1.2.7-1ubuntu > > > Le 22 juin 2016 à 15:41, Anna Kamyshnikova <akamyshnik...@mirantis.com> a > écrit : > > Hi! > > What Keepalived version is used? > > On Wed, Jun 22, 2016 at 4:24 PM, fabrice grelaud < > fabrice.grel...@u-bordeaux.fr> wrote: > >> Hi, >> >> we deployed our openstack infrastructure with your « exciting » project >> openstack-ansible (mitaka 13.1.2) but we have some problems with L3HA after >> create router. >> >> Our infra (closer to the doc): >> 3 controllers nodes (with bond0 (br-mgmt, br-storage), bond1 (br-vxlan, >> br-vlan)) >> 2 compute nodes (same for network) >> >> We create an external network (vlan type), an internal network (vxlan >> type) and a router connected to both networks. >> And when we launch an instance (cirros), we can’t receive an ip on the vm. >> >> We have: >> >> root@p-osinfra03-utility-container-783041da:~# neutron >> l3-agent-list-hosting-router router-bim >> >> +--------------------------------------+-----------------------------------------------+----------------+-------+----------+ >> | id | host >> | admin_state_up | alive | ha_state | >> >> +--------------------------------------+-----------------------------------------------+----------------+-------+----------+ >> | 3c7918e5-3ad6-4f82-a81b-700790e3c016 | >> p-osinfra01-neutron-agents-container-f1ab9c14 | True | :-) | >> active | >> | f2bf385a-f210-4dbc-8d7d-4b7b845c09b0 | >> p-osinfra02-neutron-agents-container-48142ffe | True | :-) | >> active | >> | 55350fac-16aa-488e-91fd-a7db38179c62 | >> p-osinfra03-neutron-agents-container-2f6557f0 | True | :-) | >> active | >> >> +--------------------------------------+-----------------------------------------------+----------------+-------+—————+ >> >> I know, i got a problem now because i should have :-) active, :-) >> standby, :-) standby… Snif... >> >> root@p-osinfra01-neutron-agents-container-f1ab9c14:~# ip netns >> qrouter-eeb2147a-5cc6-4b5e-b97c-07cfc141e8e6 >> qdhcp-0ba266fb-15c4-4566-ae88-92d4c8fd2036 >> >> root@p-osinfra01-neutron-agents-container-f1ab9c14:~# ip netns exec >> qrouter-eeb2147a-5cc6-4b5e-b97c-07cfc141e8e6 ip a sh >> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group >> default >> link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 >> inet 127.0.0.1/8 scope host lo >> valid_lft forever preferred_lft forever >> inet6 ::1/128 scope host >> valid_lft forever preferred_lft forever >> 2: ha-4a5f0287-91@if6: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1450 qdisc >> pfifo_fast state UP group default qlen 1000 >> link/ether fa:16:3e:c2:67:a9 brd ff:ff:ff:ff:ff:ff >> inet 169.254.192.1/18 brd 169.254.255.255 scope global ha-4a5f0287-91 >> valid_lft forever preferred_lft forever >> inet 169.254.0.1/24 scope global ha-4a5f0287-91 >> valid_lft forever preferred_lft forever >> inet6 fe80::f816:3eff:fec2:67a9/64 scope link >> valid_lft forever preferred_lft forever >> 3: qr-44804d69-88@if9: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1450 qdisc >> pfifo_fast state UP group default qlen 1000 >> link/ether fa:16:3e:a5:8c:f2 brd ff:ff:ff:ff:ff:ff >> inet 192.168.100.254/24 scope global qr-44804d69-88 >> valid_lft forever preferred_lft forever >> inet6 fe80::f816:3eff:fea5:8cf2/64 scope link >> valid_lft forever preferred_lft forever >> 4: qg-c5c7378e-1d@if12: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc >> pfifo_fast state UP group default qlen 1000 >> link/ether fa:16:3e:b6:4c:97 brd ff:ff:ff:ff:ff:ff >> inet 147.210.240.11/23 scope global qg-c5c7378e-1d >> valid_lft forever preferred_lft forever >> inet 147.210.240.12/32 scope global qg-c5c7378e-1d >> valid_lft forever preferred_lft forever >> inet6 fe80::f816:3eff:feb6:4c97/64 scope link >> valid_lft forever preferred_lft forever >> >> Same result on infra02 and infra03, qr and qg interfaces have the same >> ip, and ha interfaces the address 169.254.0.1. >> >> If we stop 2 neutron agent containers (p-osinfra02, p-osinfra03) and we >> restart the first (p-osinfra01), we can reboot the instance and we got an >> ip, a floating ip and we can access by ssh from internet to the vm. (Note: >> after few time, we loss our connectivity too). >> >> But if we restart the two containers, we got a ha_state to « standby » >> until the three become « active » and finally we have the problem again. >> >> The three routers on infra 01/02/03 are seen as master. >> >> If we ping from our instance to the router (internal network >> 192.168.100.4 to 192.168.100.254) we can see some ARP Request >> ARP, Request who-has 192.168.100.254 tell 192.168.100.4, length 28 >> ARP, Request who-has 192.168.100.254 tell 192.168.100.4, length 28 >> ARP, Request who-has 192.168.100.254 tell 192.168.100.4, length 28 >> >> And on the compute node we see all these frames on the various interfaces >> tap / vxlan-89 / br-vxlan / bond1.vxlanvlan / bond1 / em2 but nothing back. >> >> We also have on ha interface, on each router, the VRRP communication >> (heartbeat packets over a hidden project network that connects all ha >> routers (vxlan 70) ) . Priori as normal, each router thinks to be master. >> >> root@p-osinfra01-neutron-agents-container-f1ab9c14:~# ip netns exec >> qrouter-eeb2147a-5cc6-4b5e-b97c-07cfc141e8e6 tcpdump -nl -i ha-4a5f0287-91 >> tcpdump: verbose output suppressed, use -v or -vv for full protocol decode >> listening on ha-4a5f0287-91, link-type EN10MB (Ethernet), capture size >> 65535 bytes >> IP 169.254.192.1 > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 50, >> authtype simple, intvl 2s, length 20 >> IP 169.254.192.1 > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 50, >> authtype simple, intvl 2s, length 20 >> IP 169.254.192.1 > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 50, >> authtype simple, intvl 2s, length 20 >> IP 169.254.192.1 > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 50, >> authtype simple, intvl 2s, length 20 >> >> root@p-osinfra02-neutron-agents-container-48142ffe:~# ip netns exec >> qrouter-eeb2147a-5cc6-4b5e-b97c-07cfc141e8e6 tcpdump -nt -i ha-4ee5f8d0-7f >> tcpdump: verbose output suppressed, use -v or -vv for full protocol decode >> listening on ha-4ee5f8d0-7f, link-type EN10MB (Ethernet), capture size >> 65535 bytes >> IP 169.254.192.3 > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 50, >> authtype simple, intvl 2s, length 20 >> IP 169.254.192.3 > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 50, >> authtype simple, intvl 2s, length 20 >> IP 169.254.192.3 > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 50, >> authtype simple, intvl 2s, length 20 >> IP 169.254.192.3 > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 50, >> authtype simple, intvl 2s, length 20 >> IP 169.254.192.3 > 224.0.0.18: VRRPv2, Advertisement, vrid 1, prio 50, >> authtype simple, intvl 2s, length 20 >> >> >> Someone could tell me if he has already encountered this problem ? >> The infra and compute nodes are connected to a nexus 9000 switch. >> >> Thank you in advance for taking the time to study my request. >> >> Fabrice Grelaud >> Université de Bordeaux >> >> >> __________________________________________________________________________ >> OpenStack Development Mailing List (not for usage questions) >> Unsubscribe: >> openstack-dev-requ...@lists.openstack.org?subject:unsubscribe >> <http://openstack-dev-requ...@lists.openstack.org/?subject:unsubscribe> >> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev >> >> > > > -- > Regards, > Ann Kamyshnikova > Mirantis, Inc > __________________________________________________________________________ > OpenStack Development Mailing List (not for usage questions) > Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev > > > > __________________________________________________________________________ > OpenStack Development Mailing List (not for usage questions) > Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev > > -- Regards, Ann Kamyshnikova Mirantis, Inc
__________________________________________________________________________ OpenStack Development Mailing List (not for usage questions) Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
