[Thincrust-devel] eth0 interface doesn't come up on latest derived F10 AMI

[Jon Steer]

Sorry, I spoke too soon!

Unbelievable.. I can't imagine where to start looking for this.  But I
explicitly setup the AKI and ARI (I have logs of  this) and now 3-4 days
later I can't run an instance.  I instantiated each instance at least once
after I created it.

This is the bundle line
ec2-bundle-vol -d /mnt -k /mnt/pk-xxxxxxxxxxx.pem -c
/mnt/cert-xxxxxxxxxx.pem -u nnnnnnnnnn -r i386  --kernel aki-a71cf9ce
--ramdisk ari-a51cf9cc -s 2048 -p myderivedimagename


Each instance worked (even 1-2 days afterwards).. but now they don't work
and fail trying to get the network up in the same fashion as my earlier
report.

It is like some token/key times out inside the kernel? or some cache gets
flushed out?

Just for grins and giggles, I instantiated the orignal F10 ami to see if
that needed to be running or in cache before these instances would run.
That did not work.  I compared console output of the two instances (the
original F10 and my derived one) and they were the same right up to where
.
.
Bringing up interface eth0:
Determining IP information for eth0...RTNETLINK answers: No such process
 done.
RTNETLINK answers: Network is down

.


Whereupon, the derived instance is dead.

Ideas? I'm happy to try out ideas.

jon



On Tue, Dec 30, 2008 at 8:31 AM, Bryan Kearney <bkearney at redhat.com> wrote:

> Jon Steer wrote:
>
>> Sorry, guess it is my bad... rebuilding and then explicitly stating the
>> AKI and ARI produces images that run.
>>
>> thanks,
>> jon
>>
>
> Jon:
>
> Thanks for looking into this. Most of us are on vacation this week. We will
> work on the docs and talk with the Amazon folks about making the default
> ami/aki be the one you are in.
>
>
> -- bk
>
> _______________________________________________
> Thincrust-devel mailing list
> Thincrust-devel at redhat.com
> https://www.redhat.com/mailman/listinfo/thincrust-devel
>



-- 
"Don't stand still, if you see me running down the road, 'cause there is
trouble right behind me".
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/thincrust-devel/attachments/20090101/1924a66b/attachment.htm>