fab setup_all fails with traceback in setup-vnc-keepalived

[wenqing liang]

Seen in r2.21.x-89 juno, fab setup_all fails with traceback in setup-vnc-keepalived. The control_data network is not up.

2018-04-15 20:52:38:939711: [root@10.87.118.43] out: Traceback (most recent call last):
2018-04-15 20:52:38:940596: [root@10.87.118.43] out: File "/usr/bin/setup-vnc-keepalived", line 9, in <module>
2018-04-15 20:52:38:940868: [root@10.87.118.43] out: load_entry_point('ContrailProvisioning==0.1dev', 'console_scripts', 'setup-vnc-keepalived')()
2018-04-15 20:52:38:941166: [root@10.87.118.43] out: File "/usr/local/lib/python2.7/dist-packages/contrail_provisioning/common/keepalived_setup.py", line 119, in main
2018-04-15 20:52:38:941821: [root@10.87.118.43] out: keepalived.setup()
2018-04-15 20:52:38:942093: [root@10.87.118.43] out: File "/usr/local/lib/python2.7/dist-packages/contrail_provisioning/common/base.py", line 253, in setup
2018-04-15 20:52:38:943610: [root@10.87.118.43] out: self.fixup_config_files()
2018-04-15 20:52:38:944664: [root@10.87.118.43] out: File "/usr/local/lib/python2.7/dist-packages/contrail_provisioning/common/keepalived_setup.py", line 51, in fixup_config_files
2018-04-15 20:52:38:944811: [root@10.87.118.43] out: ext_device=self.get_device_by_ip(self._args.mgmt_self_ip)
2018-04-15 20:52:38:946344: [root@10.87.118.43] out: File "/usr/local/lib/python2.7/dist-packages/contrail_provisioning/compute/network.py", line 64, in get_device_by_ip
2018-04-15 20:52:38:946845: [root@10.87.118.43] out: raise RuntimeError, '%s not configured, rerun w/ --physical_interface' % ip
2018-04-15 20:52:38:946974: [root@10.87.118.43] out: RuntimeError: 10.87.118.43 not configured, rerun w/ --physical_interface
2018-04-15 20:52:38:949876: [root@10.87.118.43] out:
2018-04-15 20:52:38:965862:
2018-04-15 20:52:38:994320: Disconnecting from 10.87.118.44... done.
2018-04-15 20:52:39:110977: Disconnecting from 10.87.118.42... done.
2018-04-15 20:52:39:127450: Disconnecting from 10.87.118.40... done.
2018-04-15 20:52:39:191945: Disconnecting from 10.87.118.43... done.
2018-04-15 20:52:39:307064: Disconnecting from 10.87.118.41... done.
2018-04-15 20:52:39:422081:
2018-04-15 20:52:38:994655: Fatal error: sudo() received nonzero return code 1 while executing!
2018-04-15 20:52:38:994655:
2018-04-15 20:52:38:994655: Requested: setup-vnc-keepalived --self_ip 10.10.0.4 --internal_vip 10.10.0.200 --mgmt_self_ip 10.87.118.43 --self_index 1 --num_nodes 3 --role openstack --internal_virtual_router_id 100 --external_vip 10.0.0.200 --external_virtual_router_id 100
2018-04-15 20:52:38:994655: Executed: sudo -S -p 'sudo password:' /bin/bash -l -c "cd /opt/contrail/bin && setup-vnc-keepalived --self_ip 10.10.0.4 --internal_vip 10.10.0.200 --mgmt_self_ip 10.87.118.43 --self_index 1 --num_nodes 3 --role openstack --internal_virtual_router_id 100 --external_vip 10.0.0.200 --external_virtual_router_id 100"
2018-04-15 20:52:38:994655:
2018-04-15 20:52:38:995433: Aborting.

Installation logs (installlog.tar) are uploaded to /cs-shared/bugs/.

[wenqing liang (wliang)]

The autogenerated testbed.py for which the cluster VMs were created has the FIPs instead of the interface IPs. The problem went away after the correction and the provision went through fine. Closed the bug.