The high availability xCAT management node lost its virtual IP address after reboot

[neo954]

This bug is against xcatha.py commit 7620439004af8683a6b69bfcfc6125e95310e78d.

After activate one of the high availability xCAT management node, it seems xcatha.py does not write any permanent configuration file under directory /etc/sysconfig/network-scripts. Thus, after an operating system reboot, the xCAT management node lost its settings.

# ./xcatha.py -a -p /media/u/gongjie/ha-test -i eth0:99 -v 10.3.1.99 -m 255.0.0.0  -t sqlite
2018-06-12 01:07:33,183 - INFO - Activating this node as xCAT primary MN
############################################################################################
2018-06-12 01:07:33,183 - INFO - Activate stage
============================================================================================
2018-06-12 01:07:33,213 - INFO - Check virtual ip stage
2018-06-12 01:07:33,214 - INFO - ping -c 1 -w 10 10.3.1.99
PING 10.3.1.99 (10.3.1.99) 56(84) bytes of data.
From 10.3.1.7 icmp_seq=1 Destination Host Unreachable
From 10.3.1.7 icmp_seq=2 Destination Host Unreachable
From 10.3.1.7 icmp_seq=3 Destination Host Unreachable

--- 10.3.1.99 ping statistics ---
3 packets transmitted, 0 received, +3 errors, 100% packet loss, time 2004ms
pipe 3
2018-06-12 01:07:36,218 - INFO - virtual ip can be used.
============================================================================================
2018-06-12 01:07:36,218 - INFO - Configure virtual ip as alias ip stage
2018-06-12 01:07:36,220 - INFO - ifconfig eth0:99 10.3.1.99  netmask 255.0.0.0 [Passed]
============================================================================================
2018-06-12 01:07:36,236 - INFO - Configure hostname stage
2018-06-12 01:07:36,237 - INFO - hostname c910f03c01p99 [Passed]
2018-06-12 01:07:36,238 - INFO - Check if xCAT data is in shared data directory
2018-06-12 01:07:36,240 - INFO - There is xCAT data /media/u/gongjie/ha-test/install in shared data /media/u/gongjie/ha-test
============================================================================================
2018-06-12 01:07:36,240 - INFO - Configure shared data directory stage
2018-06-12 01:07:36,264 - INFO - systemctl stop goconserver [Passed]
2018-06-12 01:07:36,276 - INFO - systemctl stop conserver [Passed]
2018-06-12 01:07:36,287 - INFO - systemctl stop ntpd [Passed]
2018-06-12 01:07:36,299 - INFO - systemctl stop dhcpd [Passed]
2018-06-12 01:07:36,312 - INFO - systemctl stop named [Passed]
2018-06-12 01:07:36,324 - INFO - systemctl stop xcatd [Passed]
Failed to stop mariadb.service: Unit mariadb.service not loaded.
2018-06-12 01:07:39,338 - INFO - Retry 1 ... ...systemctl stop mariadb
Failed to stop mariadb.service: Unit mariadb.service not loaded.
2018-06-12 01:07:42,353 - INFO - Retry 2 ... ...systemctl stop mariadb
Failed to stop mariadb.service: Unit mariadb.service not loaded.
2018-06-12 01:07:42,364 - ERROR - systemctl stop mariadb [Failed]
Failed to stop postgresql.service: Unit postgresql.service not loaded.
2018-06-12 01:07:45,378 - INFO - Retry 1 ... ...systemctl stop postgresql
Failed to stop postgresql.service: Unit postgresql.service not loaded.
2018-06-12 01:07:48,392 - INFO - Retry 2 ... ...systemctl stop postgresql
Failed to stop postgresql.service: Unit postgresql.service not loaded.
2018-06-12 01:07:48,404 - ERROR - systemctl stop postgresql [Failed]
2018-06-12 01:07:48,405 - INFO - Creating symlink .../install
2018-06-12 01:07:48,405 - INFO - Creating symlink .../etc/xcat
2018-06-12 01:07:48,405 - INFO - Creating symlink .../root/.xcat
2018-06-12 01:07:48,405 - INFO - Creating symlink .../var/lib/pgsql
2018-06-12 01:07:48,405 - INFO - Creating symlink .../var/lib/mysql
2018-06-12 01:07:48,405 - INFO - Creating symlink .../tftpboot
2018-06-12 01:07:48,411 - INFO - cat /tmp/ha_mn >> /etc/xcat/ha_mn [Passed]
============================================================================================
2018-06-12 01:07:48,411 - INFO - Start all services stage
2018-06-12 01:07:50,696 - INFO - systemctl start xcatd [Passed]
    domain=pok.stglabs.ibm.com
2018-06-12 01:07:51,070 - INFO - lsdef -t site -i domain|grep domain [Passed]
2018-06-12 01:07:51,071 - WARNING - Long hostname is not in "/etc/hosts". "named" service will not be started
Renamed existing dhcp configuration file to  /etc/dhcp/dhcpd.conf.xcatbak

Warning: No dynamic range specified for 10.0.0.0. If hardware discovery is being used, a dynamic range is required.
2018-06-12 01:07:51,595 - INFO - makedhcp -n [Passed]
2018-06-12 01:07:51,792 - INFO - makedhcp -a [Passed]
2018-06-12 01:07:51,838 - INFO - systemctl start ntpd [Passed]
2018-06-12 01:07:51,839 - INFO - This machine is set to primary management node successfully...

[bybai]

After reboot, the VIP is lost, user need to run "--activate" process to activate this MN, it is as the same with current design.

[neo954]

I suggest, document this issue in xCAT document.

[neo954]

Refer to xcat2/xcat2-task-management#163