[Systems] [rt.prgmr.com #311] sugarlabs1 routing problem

[Bernie Innocenti]

Nick Schmalenberger via RT wrote:
> On Mon, Feb 23, 2009 at 11:36:28AM -0800, bernie at codewiz.org via RT wrote:
>> Mon Feb 23 11:36:27 2009: Request 311 was acted upon.
>> Transaction: Ticket created by bernie at codewiz.org
>>        Queue: Customer Support
>>      Subject: sugarlabs1 routing problem
>>        Owner: Nobody
>>   Requestors: bernie at codewiz.org
>>       Status: new
>>  Ticket <URL: http://rt.prgmr.com/rt/Ticket/Display.html?id=311 >
>>
>> there seems to be a routing problem with sugarlabs1:
>>
>> bernie at giskard:~$ ping sugarlabs1.xen.prgmr.com 
>> PING sugarlabs1.xen.prgmr.com (216.218.223.88) 56(84) bytes of data.
>> From 64.62.133.65 icmp_seq=1 Time to live exceeded
>> From 64.62.133.65 icmp_seq=2 Time to live exceeded
>> From 64.62.133.65 icmp_seq=3 Time to live exceeded
>>
> The routing loop can occur if the default gateway is set to the
> gateway of one of the other subnets on the same ethernet. The
> gateway for your subnet is 216.218.223.65.

Today sugarlabs1 was not replying to pings at all.
I logged in through the console and restarted networking.
Now everything works fine, but I wonder what happened.


>> BTW, can we move the remaining machines to stables?
>>
> There might be capacity for one other but not all. So is the
> performance on stables similar to boar now that you have more
> ram and cpu_weight than before, and another cpu? Thanks.

If we don't have room for all, please move only sugarlabs2.

regarding performance, a user reported sugarlabs4 being a lot slower
than the others.  So I ran my own CPU benchmark, and it turned out to
be slightly faster than the others.  Disk I/O might still be slower,
though, I don't know.

Thanks for all the work you're doing for us!

-- 
   // Bernie Innocenti - http://www.codewiz.org/
 \X/  Sugar Labs       - http://www.sugarlabs.org/