Ketan wrote:
> Problem is that the virtual interfaces are configured with application ips
> which are hardcoded for e1000g2361000:1 and when the data comes back from
> the e1000g2361000 interface it does not work
I think you may need to explain more details about the problem.
There's nothing obviously wrong about the packets being sent by this
system, and no reason at all that it should fail to work properly.
Can you describe:
- What is different about the addresses that work versus the ones that
do not?
- What you mean by data "coming back from" the e1000g2361000
interface? Coming back where?
- What does the test case consist of? Do you have traces, error logs,
or some other output showing a problem occurring?
Given the configuration you posted, both of those addresses are
equivalent and will be treated as equivalent -- completely
interchangeable -- on that system. If you need some other treatment,
then you'll need to specify it.
You didn't answer my final question from last time, so I'll ask again:
since you're using IPMP, did you mean to set some of the addresses up as
test addresses? If so, then you may be missing "-failover" flag, which
will also mark the interface as "DEPRECATED," and exclude it from use by
most applications.
I suspect this last bit really is the problem, but without more details,
it's going to be hard to determine.
--
James Carlson 42.703N 71.076W <[email protected]>
_______________________________________________
networking-discuss mailing list
[email protected]
