Amazon EC2 Service

Nathan Blackham kemotaha at
Wed Jun 8 17:04:07 MDT 2011

On Wed, Jun 8, 2011 at 4:00 PM, Merrill Oveson <moveson at> wrote:
> On Wed, Jun 8, 2011 at 1:58 PM, Nathan Blackham <kemotaha at> wrote:
>> On Wed, Jun 8, 2011 at 11:11 AM, Eric Jacobs <eric at> wrote:
>>> On Wednesday, June 8, 2011 at 12:02 PM, Merrill Oveson wrote:
>>>> I've set up instances to Windows 2008 R1 servers.
>>>> I've connected, but then after a while of working, the thing just freezes.
>>>> So I reboot and then it takes forever to let me remote back in.
>>>> So based on my experience I would say the service is:
>>>> a) unreliable
>>>> b) slow
>>>> c) unstable
>>>> Has anybody else had issues?
>>> That just sounds like every Windows experience that I've ever had. Doesn't matter if it's EC2 or on a machine in my office.
>>> I've never used EC2 though. I do know a few people that use Linux on EC2 and I haven't heard of any problems from them. Do you have a *nix instance on EC2 that does the same thing?
>> I work for Amazon on the OS team.  My team focuses on the Linux side
>> of things but I have work a bit with the windows side of things.  The
>> most common reason for someone seeing lockups after a period of time,
>> is that they are running too small an instance type.  It usually
>> happens with t1.micros which are only 613 MB of RAM and people run
>> applications that constantly increase the memory they use or need more
>> that is available.
>> I assume you are using the Amazon provided Windows AMIs rather than a
>> community one.  What instance type are you running on and what are you
>> trying to do?
>> Have you tried the 2008R2 AMIs?
>> If you want to privately send me the instance ids, I can take a quick
>> look at them and see if there are any other problems that I can spot.
>> Nathan
>> /*
>> PLUG:, #utah on
>> Unsubscribe:
>> Don't fear the penguin.
>> */
> that solved the problem.  I started an instance that was windows 2008
> R2.  Everything works good now.
> Those windows 2008 R1 micro or small aren't worth anything.

Glad to help.  Your case isn't common but isn't unheard of.  For
future reference, if you stop and start the instance it will move it
to different hardware.  This helps when the underlying hardware is
becoming degraded.  It allows to get back up and going with out
relaunching an instance.


More information about the PLUG mailing list