On Dec 19, 2007, at 9:07 AM,
alexander.graefe-gM/Ye1E23mwN+BqQ9rBEUg@xxxxxxxxxxxxxxxx wrote:
On Dec 19, 4:00 pm, Jamis Buck <ja...-uHoyYlH2B+GakBO8gow8eQ@xxxxxxxxxxxxxxxx>
wrote:
If all you want to do is push the code to that role, I'd suggest
using
deploy:update, instead of a full deploy. The ROLES=import setting on
the command-line does not mean "only run tasks associated with this
role", it means "use this role for all tasks". It's not really very
useful for deployment, as you discovered. It's primary purpose is to
restrict tasks like "invoke" or "deploy:setup".
Thanks, that worked to get the server up and running.
But I am curious, if I want to do a full code push on all servers, do
I just run "cap production deploy", and the roles restriction takes
care of the mongrel issue?
By "mongrel issue", do you mean "not starting mongrel on the import
role"? If so, yes. Your servers in the import role will not be used to
run mongrel.
- Jamis
smime.p7s
Description: S/MIME cryptographic signature
