[TriLUG] Networking and Fedora

[Joseph Tate]

That would require a home phone line, which I don't have.  An
unplanned power outage of more than 30 minutes would probably "fix" it
(would have to drain the UPS).  While those are not unheard of in my
neighborhood, I don't think I can count on it happening on cue.

On Wed, May 26, 2010 at 3:25 PM, Clay Stuckey <claystuckey at gmail.com> wrote:
> Time to buy a modem and go old school on it!
>
>
>
> --
> Clay Stuckey
> Email: clay.stuckey at sas.com, claystuckey at gmail.com
> Office: (919) 301-1792
> Cell: (919) 600-0486
>
>
> -----Original Message-----
> From: trilug-bounces at trilug.org [mailto:trilug-bounces at trilug.org] On Behalf
> Of Joseph Tate
> Sent: Wednesday, May 26, 2010 3:04 PM
> To: Triangle Linux Users Group General Discussion
> Subject: Re: [TriLUG] Networking and Fedora
>
> Well, that went poorly.
>
> As soon as I issued the brctl command my whole network went dark.  The
> shell I was working in hung, and I forcefully disconnected to the top
> level ssh shell instead of the sub shell.  Now I can't connect to that
> host machine, or even the other machine that I was using to connect
> through it.  I guess it'll have to wait until I get home to fix it up.
>  That'll teach me to troubleshoot network issues remotely... :/
>
> On Wed, May 26, 2010 at 2:50 PM, Justis Peters <jtrilug at indythinker.com>
> wrote:
>> Jeff Schornick wrote:
>>> Preparing to make a fool of myself...
>>>
>>>> 10.2.2.0        0.0.0.0         255.255.255.0   U     0      0        0
> eth0
>>>> 10.2.2.0        0.0.0.0         255.255.255.0   U     0      0        0
> br0
>>>>
>>> You appear to have two routes to the same destination subnet, equal
>>> metric to two different interfaces.  Since eth0 and br0 aren't
>>> currently bridged, I'd imagine this would lead to rather inconstant
>>> delivery.  I'm honestly not sure how the kernel choses which route to
>>> use under these circumstances, but I'm 99% sure it won't deliver each
>>> packet to both.
>>>
>>> Do you perhaps want eth0 part of the br0 bridge group?
>>>
>> Didn't see your reply before I sent mine. I think you've got it nailed,
>> Jeff. I hadn't noticed that eth0 is not yet attached to br0.
>>
>> The way that Xen users frequently do this is to rename the "real" eth0
>> to peth0 and then name the bridge itself eth0. You would then add peth0
>> to the bridge, along with all the virtual interfaces for the guest VMs.
>> Some people find this confusing; others like it.
>>
>> Joseph: The quick way to test our theory would be to remove the second
>> route and then add eth0 to the br0 bridge. The commands should look
>> something like this:
>>  route del -net 10.2.2.0/24 br0
>>  brctl addif br0 eth0
>>
>> I'm not sure of the preferred way to make those changes persistent in
>> Fedora, but I'm sure the answer is online and easy to find. Also note
>> that you probably want to run them in the order listed above. I suspect
>> that reversing the order might create a feedback loop between the
>> routing table and the bridge.
>>
>> Best of luck and let us know how it goes.
>>
>> Kind regards,
>> Justis
>> --
>> TriLUG mailing list : http://www.trilug.org/mailman/listinfo/trilug
>> TriLUG FAQ          :
> http://www.trilug.org/wiki/Frequently_Asked_Questions
>>
>
>
>
> --
> Joseph Tate
> Personal e-mail: jtate AT dragonstrider DOT com
> Web: http://www.dragonstrider.com
> --
> TriLUG mailing list : http://www.trilug.org/mailman/listinfo/trilug
> TriLUG FAQ          : http://www.trilug.org/wiki/Frequently_Asked_Questions
>
>
>
> --
> TriLUG mailing list : http://www.trilug.org/mailman/listinfo/trilug
> TriLUG FAQ          : http://www.trilug.org/wiki/Frequently_Asked_Questions
>
>



-- 
Joseph Tate
Personal e-mail: jtate AT dragonstrider DOT com
Web: http://www.dragonstrider.com