This has been fixed in http://dev.rubyonrails.org/changeset/6702. It
will be part of Preview Release 2, later this week.
The above change implements the following:
1. If you set neither :scm_command nor :local_scm_command, the
default will be used for both local and remote hosts.
2. If you
As a way of managing my sanity while keeping Capistrano 1 and 2
installed for different projects, I created a few Bash aliases to make
the process more obvious. My "cap" wrapper script is installed in /
opt/local/bin/cap, and I've setup the following:
alias cap='echo "You must run either cap1 o
> It uses the same known_hosts file as the OpenSSH command-line client
> (~/.ssh/known_hosts).
That's good to know. It turns out that the fault was mine: I hadn't
put the right user keys in all the right places. Once I did, it
worked. It took a while to figure out partially because there was
Thank you for you hints. Will try to taylor something for my needs. :)
the plugin is new to me. Will have a look into it.
thanks
On 8 Mai, 18:32, wolfmanjm <[EMAIL PROTECTED]> wrote:
> thats one of the things I love about postgresql, you can wrap
> migrations in a transaction, so failed migrati
Agreed. OSX has the same problem and ssh_options[:paranoid] = false
works around it.
Casey
On May 2, 9:24 am, "[EMAIL PROTECTED]" <[EMAIL PROTECTED]> wrote:
> I'm also remote deploying from OS X.
>
> Adding "ssh_options[:paranoid] = false" does solve the problem.
> Looking through the 1.0.10 d
On May 8, 2007, at 11:35 AM, Michael Hartl wrote:
> I found this thread by dropping the error message into Google. I'm
> having the same problem on Linux (CentOS). Setting
> ssh_options[:paranoid] = false works for me as well.
>
> My command-line ssh works fine currently, but errors like this d
I found this thread by dropping the error message into Google. I'm
having the same problem on Linux (CentOS). Setting
ssh_options[:paranoid] = false works for me as well.
My command-line ssh works fine currently, but errors like this do
happen when hosts move; fixing them usually involves editi
thats one of the things I love about postgresql, you can wrap
migrations in a transaction, so failed migrations leave the database
in a determinable state.
coupled with this plugin, it makes life easier...
http://www.redhillonrails.org/#transactional_migrations
However (at least with cap 1) if
Migrations are tricky beasts, because if they fail halfway through,
your database will be left with part of the migration applied. Most
DBMS's will not allow you to do things like table creations or column
additions inside a transaction, so there is no easy way to atomically
apply a migrat
Thanks for suggesting that fix, Chris. I'm going to get that in as
soon as I can.
- Jamis
On May 8, 2007, at 4:47 AM, Chris Andrews wrote:
> Ah, thanks for testing that out. This getnameinfo issue on Windows
> seems to be causing a few problems that have been reported here,
> and this see
I'm not sure I understand what you mean by "reverse tunnel."
Is the problem that your production environment cannot access your
subversion repository? If so, then the solution is in Capistrano 2.0,
where you can say:
set :deploy_via, :copy
And deployments will then do a local checkout, t
Hello,
i'm just looking for a solution to rollback my app after a migration
failed.
I'd like to run the migration with every deploy command. but if the
migration fails, i'd like to rollback to the last version before the
migration.
can i do a "dry run" to check if the migration is ok? or do i ha
Hello,
i'm just looking for a solution to rollback my app after a migration
failed.
I'd like to run the migration with every deploy command. but if the
migration fails, i'd like to rollback to the last version before the
migration.
can i do a "dry run" to check if the migration is ok? or do i ha
Ah, thanks for testing that out. This getnameinfo issue on Windows seems to
be causing a few problems that have been reported here, and this seems a
reasonable fix -- it doesn't cause trouble on either OSX or Linux.
Chris.
On 08/05/07, dweinand <[EMAIL PROTECTED]> wrote:
>
>
> Thank you, just fou
Thank you, just found your Thread and wanted to tell it here. But
you've been faster.
Thank you for the fix. Works now on this machine.
Great Work!
On May 8, 12:16 pm, "Chris Andrews" <[EMAIL PROTECTED]> wrote:
> Hi,
>
> Could you try the suggestion I've just made in that thread?
>
> "I believe
Hi,
Could you try the suggestion I've just made in that thread?
"I believe you should also be able to fix this problem without editing your
services file, by changing line 108 of net/ssh/transport/session.rb to:
ip_address = Socket.getnameinfo(addr, Socket::NI_NUMERICHOST |
Socket::NI_NUMERICSER
Doing the things described here won't fix the problem.
http://rubyforge.org/forum/forum.php?thread_id=13883&forum_id=1177
Is it a problem with Capistrano or net-ssh?
is Capistrano using an different implementation?
On 8 Mai, 10:18, dweinand <[EMAIL PROTECTED]> wrote:
> Sorry, wrong Information.
Sorry, wrong Information. It isn't working on this machine after
adding the ssh_options[:paranoid] = false option.
just the add_key method works. I was remote on the wrong box as i
tested again.
after entering my password i get an SocketError.
Full error here:
http://pastie.caboo.se/59791
On 8
Hello,
i'm getting an error on my windows box if i try to run capistrano from
this laptop. this is the error:
host-key-verifier.rb:64:in `add_key': undefined method `mkdir' for
File:Class (NoMethodError)
the error occurs if i try to run cap with my windows ruby
installation, but on the same machi
Currently I have an local netwerk containing my development
environment.
This includes the computer I am running Capistrano on and an Apache-
svn server.
On my test server the repository is reachable and Capistrano works
perfect.
I would like to be able to run Capistrano to setup my live
environm
20 matches
Mail list logo
