On 20/08/2011, at 4:05 AM, Michael Solberg wrote: > ----- Justin Clift <[email protected]> wrote: >> On 19/08/2011, at 12:32 PM, Michael Solberg wrote: >>> ----- Chris Lalancette <[email protected]> wrote: >>>> The actual problem, though, probably has more to do with the fact that we >>>> couldn't properly get ssh access to the instance in a reasonable amount of >>>> time. Quick question; are you building against EC2 US east? If so, we >>>> have >>>> been seeing problems with it in the past couple of days. We still aren't >>>> sure if the problem is on our end or on Amazon's, but in any case it is not >>>> working at the moment. It might be wortwhile to try using EC2 US west and >>>> see if you have any better luck. >>> >>> I've just run into this in my testing. Is there any idea what could be >>> causing this issue and what a possible resolution would be? I'm able to >>> provision to EC2 West, but not East. >> >> Ouch, as a temporary workaround for EC2 West, try allowing SSH inbound >> connections on the EC2 West "Security Groups". >> >> The steps to do that (with screenshots) are here on an old Getting Started >> Guide page: >> >> http://www.aeolusproject.org/start_image.html >> 2. Configure EC2 for SSH connections >> >> That's all in the AWS interface. The other steps on the page, to do with >> instances and retrieving the private SSH key, are not relevant for Aeolus >> 0.3.0. >> >> Does that help? > > I got this working on EC2 West last night with the firewall changes and was > able to demonstrate Aeolus to a (very impressed, BTW) group of engineers > today. I was just curious as to when we'd have EC2 East working again or > if we might lose connectivity to EC2 West in the same way.
Ouch, I'd actually misread your email, thinking that you were able to provision to EC2 East, but not EC2 West. The SSH firewall thing I mentioned is a workaround to something I'd seen (ages ago) with regions *other* than EC2 East. Which I'd thought was actually fixed. ;) Anyway, seeing that EC2 East is being the problem for you, the firewall thing probably isn't actually relevant. :/ Sorry for the bum steer there. With AWS, they do seem to be changing instance availability in their regions quite a lot lately. EC2 East in particular. If it's any help, a Deltacloud issue was recently opened regarding improving the communication of Cloud specific errors: https://issues.apache.org/jira/browse/DTACLOUD-62 That would allow Aeolus (and other Deltacloud using things) to then handle this cloud-platform-under-us-is-changing kind of situation better. Regards and best wishes, Justin Clift -- Aeolus Community Manager http://www.aeolusproject.org
