Hi all!

I have a bonded Infiniband (basically super-fast ethernet) interface on my oracle linux 5.8 server, made of ib0 and ib1. Old address was 192.168.10.1/22, new address is 192.168.10.201/22. I changed the ip addresses on the bonded interface by manually editing the ifcfg-bondib0 file, and now the server cannot ping anything else on the same switch (no vlans involved - it's a flat network). It's still the same subnet, so routing should not be an issue - the infiniband network is a private network that is internal to the database system and does not get routed outside the server rack, so we don't have a default gateway for it - that's on a different interface.

When I do an "ip addr show" I see both the new address AND the old address:
12: bondib0: <NO-CARRIER,BROADCAST,MULTICAST,MASTER,UP> mtu 65520 qdisc noqueue
link/infiniband 80:00:00:49:fe:80:00:00:00:00:00:00:00:21:28:00:01:cf:12:8c brd 00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff
inet 192.168.10.201/22 brd 192.168.11.255 scope global bondib0
inet 192.168.10.1/22 brd 192.168.11.255 scope global secondary bondib0

I've tried flushing and re-assigning the addresses on the bondib0 interface, checked all the config files I know of, but I'm missing something somewhere. Somehow that interface is getting the 'old' address assigned somewhere. Any ideas or advice on how I can remove it permanently?

My ifcfg-bondib0 file:
DEVICE=bondib0
USERCTL=no
BOOTPROTO=none
ONBOOT=yes
#IPADDR=192.168.10.1
IPADDR=192.168.10.201
NETMASK=255.255.252.0
NETWORK=192.168.8.0
BROADCAST=192.168.11.255
BONDING_OPTS="mode=active-backup miimon=100 downdelay=5000 updelay=5000 num_grat_arp=100"
IPV6INIT=no
MTU=65520


My /proc/net/bonding/bondib0 file:
Ethernet Channel Bonding Driver: v3.6.0 (September 26, 2009)

Bonding Mode: fault-tolerance (active-backup) (fail_over_mac active)
Primary Slave: None
Currently Active Slave: None
MII Status: down
MII Polling Interval (ms): 100
Up Delay (ms): 5000
Down Delay (ms): 5000

Slave Interface: ib0
MII Status: down
Link Failure Count: 0
Permanent HW addr: 80:00:00:48:fe:80
Slave queue ID: 0

Slave Interface: ib1
MII Status: down
Link Failure Count: 1
Permanent HW addr: 80:00:00:49:fe:80
Slave queue ID: 0

Which slave is active at present as per below none of slave is active

Currently Active Slave: None

Also does this two address appear even after restarting network service ?


Thanks

OK, a coworker helped me ont his one. It was a stupid newbie mistake - when I created a backup of the original ifcfg file, I named it "ifcfg-bond0.old" so.. when the machine booted it read that file since it starts with ifcfg - not caring that it ended with ".old". That's how I was getting two different addresses on the same bond interface.

I renamed the 'old' file to old.ifcfg-bond0 - and it worked great!