hi all. I found that it has been fixed a mouth ago: https://github.com/Juniper/contrail-fabric-utils/commit/1fbdf305f3f767b28f2fcfbdaa54b5c19e54a9ca#diff-14478d03bf08eb5a4f6ca99c693246d2
thanks for all your reply! tony CertusNet 赛特斯信息科技股份有限公司 龚瑞涛 | 高级软件工程师 上海市杨浦区国泰路11号复旦科技园大厦20层 电话：021 6109 6070 �C8004 手机：156 1895 8005 公司网址：www.certusnet.com 发件人： Ranjeet R 发送时间： 2015-08-26 00:34 收件人： [email protected]; dev 主题： RE: [opencontrail-dev] fabric error in config keepalive component in opencontrail-2.20 HA feature Hello In 2.20, we made the virtual_router_id to be configurable to facilitate running two Contrail clusters in the same L2 domain (previously, conflicts arose when the same virtual_router_id was used for the two clusters in the same L2 domain) You can configure the virtual router id’s in testbed.py as follows (example below) #env.ha = { # 'internal_vip' : '1.1.1.100', #Internal Virtual IP of the openstack HA Nodes. # 'external_vip' : '2.2.2.200', #External Virtual IP of the openstack HA Nodes. # 'contrail_internal_vip' : '1.1.1.10', #Internal Virtual IP of the contrail HA Nodes. # 'contrail_external_vip' : '2.2.2.20', #External Virtual IP of the contrail HA Nodes. # 'nfs_server' : '3.3.3.3', #IP address of the NFS Server which will be mounted to /var/lib/glance/images of openstack Node, Defaults to env.roledefs['compute'][0] # 'nfs_glance_path' : '/var/tmp/images/', #NFS Server path to save images, Defaults to /var/tmp/glance-images/ # 'internal_virtual_router_id' : 180, #Default = 100 # 'external_virtual_router_id' : 190, #Default = 100 # 'contrail_internal_virtual_router_id' : 200, #Default = 100 # 'contrail_external_virtual_router_id' : 210, #Default = 100 #} However, these values are configured to default to 100 if they were not configured in testbed.py. The code is there in R2.20 but I don’t know why it did not work in your setup. Let me test it to see what happens when I don’t give any of these fields in testbed.py. To fix this, edit /etc/keepalived/keepalived.conf and change the virtual_router_id to an unique number in all the three controllers (same virtual_router_id) and restart keepalived using “service keepalived restart”. Hope that helps! Ranjeet From: Dev [mailto:[email protected]] On Behalf Of [email protected] Sent: Tuesday, August 25, 2015 12:43 AM To: dev <[email protected]> Subject: [opencontrail-dev] fabric error in config keepalive component in opencontrail-2.20 HA feature Hi, I found that HA feature in opencontrail 2.20 can't be installed successfully. can't find the VIP. in further research, I found 'virtual_router_id' of vrrp instance in /etc/keepalived/keepalived.conf is set to be "None", then find the code in contrail-2.01: contrail_provisioning/common/keepalived_setup.py if vip_name == 'INTERNAL': router_id = 101 if 'openstack' in self._args.role: router_id = 100 external_device = internal_device else: router_id = 201 if 'openstack' in self._args.role: router_id = 200 but in the new version if vip_name == 'INTERNAL': router_id = self._args.internal_virtual_router_id external_device = internal_device else: router_id = self._args.external_virtual_router_id it needs to be identify in fabfile/tasks/ha.py: with cd(INSTALLER_DIR): cmd = "setup-vnc-keepalived\ --self_ip %s --internal_vip %s --mgmt_self_ip %s\ --self_index %d --num_nodes %d --role %s" % ( self_ip, internal_vip, mgmt_ip, (keepalived_host_list.index(self_host) + 1), len(env.roledefs[role]), role) if external_vip: cmd += ' --external_vip %s' % external_vip sudo(cmd) but there is no "--internal_virtual_router_id" or "--external_virtual_router_id" in param , so "vrouter_id" should be defined in testbed.py ? CertusNet 赛特斯信息科技股份有限公司 龚瑞涛 | 高级软件工程师 上海市杨浦区国泰路11号复旦科技园大厦20层 电话：021 6109 6070 �C8004 手机：156 1895 8005 公司网址：www.certusnet.com
_______________________________________________ Dev mailing list [email protected] http://lists.opencontrail.org/mailman/listinfo/dev_lists.opencontrail.org
