hi everyone
i created new virtual machines (centos6) on vmware esx5 but i have problem in a specific vlan (20). when i want to start or restart an interface it shows me "Error, some other host already uses address x.x.x.x".
i thought it is related to OS version but when i tested another os version (like red hat 5 , centos5) it showed the same error. however when i change vlan number it will be correct but when i return to vlan20 it has the same problem. i dont know why but this error only happened in Linux and when i change a windows vm to vlan20 it will work correctly. thanks in advance
On the windows VM, you are using the same ip address when testing? What type of router do you have? You might have a arp issue on that VLAN.
Hi seighalani,
Welcome to the communities! I agree with timlane on the arp topic. If all else is normal, a common cause for this behavior (which is by design in such cases) is proxy arp on the firewall if you use that.
http://ccnpsecurity.blogspot.com/2011/11/proxy-arp-on-firewall.html
yeah i used the same ip in windows vm . we use cisco 6500 core sw . i dont have any problem in physical systems in vlan20 but linux vms get errors.
hi again
i double check interface vlan20. it doesnt use proxy arp ( no proxy arp ).
/etc/sysconfig/network-scripts/ifcfg-eth0 just comment out the mac address of Linux, and restart network service.
then check, is it giving same error message??
-A
hi
it giving me the same error after comment out HWADDR . ;(
NSLOOKUP the ID to find the name of device attached with.adn change any of the VM IPaddress to conflict.Prode a free IP to avoid the IP conflict.
Can you detailed log info from the logs..
Hi ,
I am facing the same issue here.I have deployed red-hat guess over Exsi 5 in my vcenter. Can anybody guide me to resolve the issue.Thanks in advance.
Regards,
Tamil
I've got the same problem and I believe I know why I've got it: I have recycled an old IP-Adress and somehow VM seems to have noticed the Adress although it is no longer in use.
We have found a kind of workaround: The script ifup-eth (etc/sysconfig/network-scripts/) contains one query:
# if ! /sbin/arping -q -c 2 -w 3 -D -I ${REALDEVICE} ${ipaddr[$idx]} ; then
# net_log $"Error, some other host already uses address ${ipaddr[$idx]}."
# exit 1
# fi
If one comments this block out we manage to start the machine with obviously no warning and everything works fine (as long as you are sure the IP-Adress is really nowhere to find!)
But that can't be the "real" solution, can it?
Neither worked something like this
arping -I eth0 -c 3 eth0 90.xx.xx.xx
nor the following
cp ifcfg-eth0 /etc/sysconfig/networking/devices/
cp ifcfg-eth0 /etc/sysconfig/networking/profiles/default/
Hi Foltom,
Thanks for work around , It has worked for me .
Cheers,
Bhanu Prakash K
Hope My problem same , also will be solved by this approach.
Best Regards,
arun