Hi,
I have a machine that is having issues connecting to a machine in onse
specific other subnet. The two subnets that are giving issues are on one
interface using an IP alias.
It seems the machine is storing hardware addresses in the ARP table for
the machines in the other subnet (reachable through gateway). The first
connection works, and then you need to wait for some time until you can
do another connection that works. I think the issues are related to an
ARP cache issue (because when the ARP entry is gone, the first
connection always works). Can/should it do that?
It sounds very weird since it will need to pass the gateway to reach the
other subnet, so why/how cache hardware adresses? (it seems that ARP
information is leaking between both subnets)
If it tries to use that HW address (in stead of the gateway) to talk to
that machine next time, then, well, that seems logical it doesn't get
any response...
I can't delete the dynamic ARP either entry because then it says (of
course):
SIOCDARP(priv): Network is unreachable
Can I disable the ARP cache manually for a certain IP range? Should I
change something at the gateway? Am I missing something?
(Note: windows machines in the same subnet do not have this issue)
-- Velpi
_______________________________________________
rhelv5-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/rhelv5-list
