currently, the ec2 datasource has some code that selects the mirror location by looking at the availability-zone.
I think this is generally a pretty good idea. I'd like to extend that to other datasources, and not limit to ".ec2.archive.ubuntu.com".
The goal would be to basically check for a mirror at:
<availability-zone>[.<cloud>][dns-subdomain]
dns subdomain would default to '.cloud.archive.ubuntu.com'.
if the datasource provided the 'cloud' value, we'd use that, but otherwise just use availability-zone.
This would fit in with what we're currently doing on ec2, but would extend to others.
I'd just treat the "cc-direction-[0-9]" availability-zones specially, to know that this is an ec2 one. Then basically append ".ec2" in that case.
currently, the ec2 datasource has some code that selects the mirror location by looking at the availability-zone. ubuntu. com".
I think this is generally a pretty good idea. I'd like to extend that to other datasources, and not limit to ".ec2.archive.
The goal would be to basically check for a mirror at: -zone>[ .<cloud> ][dns-subdomain ]
<availability
dns subdomain would default to '.cloud. archive. ubuntu. com'.
if the datasource provided the 'cloud' value, we'd use that, but otherwise just use availability-zone.
This would fit in with what we're currently doing on ec2, but would extend to others. [0-9]" availability-zones specially, to know that this is an ec2 one. Then basically append ".ec2" in that case.
I'd just treat the "cc-direction-