Description of problem: openshift-tc's init script has a default start priority of 7. This means that it's started before the 'network' service (prio 10) which causes it to fail to start when the NIC configured for openshift service (EXTERNAL_ETH_DEV) depends on the network service to start. Version-Release number of selected component (if applicable): rubygem-openshift-origin-node-1.34.1.1-1.el6op.noarch How reproducible: Alwasys Steps to Reproduce: 1. Install a node with NIC bonding and onfigure EXTERNAL_ETH_DEV=bond0 2. Enable TC (TRAFFIC_CONTROL_ENABLED=true, chkconfig openshift-tc on) 3. boot Actual results: openshift-tc fails to start. Extract from boot.log: ...... Starting systemtap: [WARNING] Calling the system activity data collector (sadc)... Starting monitoring for VG root_vg: 5 logical volume(s) in volume group "root_vg" monitored [ OK ] Starting cgconfig service: [ OK ] Starting multipathd daemon: [ OK ] /opt/rh/ruby193/root/usr/share/gems/gems/openshift-origin-node-1.23.9.21/lib/openshift-origin-node/utils/tc.rb:107:in `get_interface_mtu': Unable to determine external network interface IP address. (RuntimeError) from /opt/rh/ruby193/root/usr/share/gems/gems/openshift-origin-node-1.23.9.21/lib/openshift-origin-node/utils/tc.rb:80:in `initialize' from /usr/sbin/oo-admin-ctl-tc:28:in `new' from /usr/sbin/oo-admin-ctl-tc:28:in `<main>' iptables: Applying firewall rules: [ OK ] iptables: Loading additional modules: nf_conntrack [ OK ] Bringing up loopback interface: [ OK ] Bringing up interface bond0: Determining if ip address 22.114.234.40 is already in use for device bond0... [ OK ] Bringing up interface bond1: Determining if ip address 22.115.210.67 is already in use for device bond1... [ OK ] Starting auditd: [ OK ] ..... Expected results: Additional info: [root@node4 init.d]# grep chkconfig network openshift-tc network:# chkconfig: 2345 10 90 openshift-tc:# chkconfig: 345 7 93
https://github.com/openshift/origin-server/pull/6146
Verified and pass 1) The chkconfig priority have been changed [root@broker init.d]# grep chkconfig network openshift-tc network:# chkconfig: 2345 10 90 openshift-tc:# chkconfig: 345 11 89 2) network can be boot. Starting monitoring for VG vg_dhcp12945: 2 logical volume(s) in volume group "vg_dhcp12945" monitored ^[[60G[^[[0;32m OK ^[[0;39m]^M Starting cgconfig service: ^[[60G[^[[0;32m OK ^[[0;39m]^M ip6tables: Applying firewall rules: ^[[60G[^[[0;32m OK ^[[0;39m]^M iptables: Applying firewall rules: ^[[60G[^[[0;32m OK ^[[0;39m]^M Bringing up loopback interface: ^[[60G[^[[0;32m OK ^[[0;39m]^M Bringing up interface bond0: Determining IP information for bond0... done. ^[[60G[^[[0;32m OK ^[[0;39m]^M 3) openshift-tc is started [root@broker init.d]# service openshift-tc status Bandwidth shaping status: qdisc htb 1: root refcnt 17 r2q 10 default 0 direct_packets_stat 2075 Sent 249309 bytes 2079 pkt (dropped 0, overlimits 0 requeues 0) rate 0bit 0pps backlog 0b 0p requeues 0 class htb 1:1 root prio 0 rate 800000Kbit ceil 800000Kbit burst 1600b cburst 1600b Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0) rate 0bit 0pps backlog 0b 0p requeues 0 lended: 0 borrowed: 0 giants: 0 tokens: 250 ctokens: 250 [OK]
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHBA-2015-1463.html