[Capistrano] problem with cat /etc/passwd do |ch, stream, data|

2009-01-15 Thread Gerhardus Geldenhuis

Hi
I saw the following code on 
http://blog.innerewut.de/2007/9/28/capturing-output-in-capistrano
and thought it was capistrano

run cat /etc/passwd  do |ch, stream, data|
  if stream == :err
logger.debug capured output on STDERR: #{data}
  else # stream == :out
logger.debug capured output on STDOUT: #{data}
  end
end

but I get a syntax error when trying to run it as a task.
capfile:53:in `load': compile error (SyntaxError)
capfile:53: syntax errorfrom /usr/lib/ruby/gems/1.8/gems/
capistrano-2.5.3/lib/capistrano/configuration/loading.rb:172:in
`load_from_file'
from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
capistrano/configuration/loading.rb:89:in `load'
from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
capistrano/configuration/loading.rb:86:in `load'
from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
capistrano/configuration/loading.rb:86:in `each'
from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
capistrano/configuration/loading.rb:86:in `load'
from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
capistrano/cli/execute.rb:64:in `load_recipes'
from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
capistrano/cli/execute.rb:64:in `each'
from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
capistrano/cli/execute.rb:64:in `load_recipes'
from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
capistrano/cli/execute.rb:30:in `execute!'
from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
capistrano/cli/execute.rb:14:in `execute'
from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/bin/cap:4
from /usr/bin/cap:16:in `load'
from /usr/bin/cap:16

Is this ruby code I am trying to run or is there something I am doing
wrong?

Regards
--~--~-~--~~~---~--~~
To unsubscribe from this group, send email to 
capistrano-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/capistrano
-~--~~~~--~~--~--~---



[Capistrano] Capistrano::ConnectionError

2009-01-15 Thread Xazoola

Hi,
I am getting a ConnectionError on Solaris. Works find on Linux. anyone
know what causes this error?

Capistrano::ConnectionError, connection failed for: IPADDRESS
(Net::SSH::Exception: could not settle on encryption_client algorithm)

--~--~-~--~~~---~--~~
To unsubscribe from this group, send email to 
capistrano-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/capistrano
-~--~~~~--~~--~--~---



[Capistrano] Re: Capistrano::ConnectionError

2009-01-15 Thread Lee Hambley
This might be ssh v1 vs. v2 issues, are you definately using SSH v2, with v2
keys? (i think it is a difference between DSA, and RSA keys)- Lee

2009/1/15 Xazoola col...@gmail.com


 Hi,
 I am getting a ConnectionError on Solaris. Works find on Linux. anyone
 know what causes this error?

 Capistrano::ConnectionError, connection failed for: IPADDRESS
 (Net::SSH::Exception: could not settle on encryption_client algorithm)

 


--~--~-~--~~~---~--~~
To unsubscribe from this group, send email to 
capistrano-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/capistrano
-~--~~~~--~~--~--~---



[Capistrano] Re: problem with cat /etc/passwd do |ch, stream, data|

2009-01-15 Thread Jamis Buck

Make sure you aren't getting some fancy quotes from the copy/paste.
Often, code copy/pasted from a blog post will include fancy quotes and
mdashes instead of ascii quotes and hyphens, and ruby does not recognize
those characters as valid in its syntax.

- Jamis

On 1/15/09 4:30 AM, Gerhardus Geldenhuis wrote:
 Hi
 I saw the following code on 
 http://blog.innerewut.de/2007/9/28/capturing-output-in-capistrano
 and thought it was capistrano
 
 run cat /etc/passwd  do |ch, stream, data|
   if stream == :err
 logger.debug capured output on STDERR: #{data}
   else # stream == :out
 logger.debug capured output on STDOUT: #{data}
   end
 end
 
 but I get a syntax error when trying to run it as a task.
 capfile:53:in `load': compile error (SyntaxError)
 capfile:53: syntax errorfrom /usr/lib/ruby/gems/1.8/gems/
 capistrano-2.5.3/lib/capistrano/configuration/loading.rb:172:in
 `load_from_file'
 from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
 capistrano/configuration/loading.rb:89:in `load'
 from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
 capistrano/configuration/loading.rb:86:in `load'
 from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
 capistrano/configuration/loading.rb:86:in `each'
 from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
 capistrano/configuration/loading.rb:86:in `load'
 from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
 capistrano/cli/execute.rb:64:in `load_recipes'
 from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
 capistrano/cli/execute.rb:64:in `each'
 from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
 capistrano/cli/execute.rb:64:in `load_recipes'
 from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
 capistrano/cli/execute.rb:30:in `execute!'
 from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/lib/
 capistrano/cli/execute.rb:14:in `execute'
 from /usr/lib/ruby/gems/1.8/gems/capistrano-2.5.3/bin/cap:4
 from /usr/bin/cap:16:in `load'
 from /usr/bin/cap:16
 
 Is this ruby code I am trying to run or is there something I am doing
 wrong?
 
 Regards
  


--~--~-~--~~~---~--~~
To unsubscribe from this group, send email to 
capistrano-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/capistrano
-~--~~~~--~~--~--~---



[Capistrano] Re: Capistrano::ConnectionError

2009-01-15 Thread Jamis Buck

This can also be caused if your SSH server is configured to allow only a
small subset of cipher algorithms, and that subset does not overlap any
of the algorithms that Net::SSH supports.

Net::SSH supports the following ciphers:

  aes128-cbc
  3des-cbc
  blowfish-cbc
  cast128-cbc
  aes192-cbc
  aes256-cbc
  rijndael-...@lysator.liu.se
  idea-cbc
  none

(Though it is strongly recommended to not use 'none'. It's really only
useful for debugging Net::SSH.)

If there is a cipher you want that isn't included there, you'll need to
see how to make Ruby's OpenSSL bindings provide it.

- Jamis

On 1/15/09 8:37 AM, Lee Hambley wrote:
 This might be ssh v1 vs. v2 issues, are you definately using SSH v2,
 with v2 keys? (i think it is a difference between DSA, and RSA keys)
 - Lee
 
 2009/1/15 Xazoola col...@gmail.com mailto:col...@gmail.com
 
 
 Hi,
 I am getting a ConnectionError on Solaris. Works find on Linux. anyone
 know what causes this error?
 
 Capistrano::ConnectionError, connection failed for: IPADDRESS
 (Net::SSH::Exception: could not settle on encryption_client algorithm)
 
 
 
 
  


--~--~-~--~~~---~--~~
To unsubscribe from this group, send email to 
capistrano-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/capistrano
-~--~~~~--~~--~--~---



[Capistrano] Re: Capistrano::ConnectionError

2009-01-15 Thread Xazoola

You have just nailed it.
There was a SSH CBC vulnerability reported by US-CERT.
http://www.kb.cert.org/vuls/id/958563
I'm not sure how the vulnerability is exploited. But to be safe, I
disable all CBC related ciphers.
After enabling all the default ciphers aes128-ctr,aes128-cbc,arcfour,
3des-cbc,blowfish-cbc, the encryption_client algorithm is solved.



do you know anything about this SSH CBC vulnerability? If it is a
security problem when CBC related cipher enabled on SSH server, any
plan for Net:SSH to support non CBC ciphers?

On Jan 15, 4:03 pm, Jamis Buck ja...@37signals.com wrote:
 This can also be caused if your SSH server is configured to allow only a
 small subset of cipher algorithms, and that subset does not overlap any
 of the algorithms that Net::SSH supports.

 Net::SSH supports the following ciphers:

   aes128-cbc
   3des-cbc
   blowfish-cbc
   cast128-cbc
   aes192-cbc
   aes256-cbc
   rijndael-...@lysator.liu.se
   idea-cbc
   none

 (Though it is strongly recommended to not use 'none'. It's really only
 useful for debugging Net::SSH.)

 If there is a cipher you want that isn't included there, you'll need to
 see how to make Ruby's OpenSSL bindings provide it.

 - Jamis

 On 1/15/09 8:37 AM, Lee Hambley wrote:

  This might be ssh v1 vs. v2 issues, are you definately using SSH v2,
  with v2 keys? (i think it is a difference between DSA, and RSA keys)
  - Lee

  2009/1/15 Xazoola col...@gmail.com mailto:col...@gmail.com

      Hi,
      I am getting a ConnectionError on Solaris. Works find on Linux. anyone
      know what causes this error?

      Capistrano::ConnectionError, connection failed for: IPADDRESS
      (Net::SSH::Exception: could not settle on encryption_client algorithm)
--~--~-~--~~~---~--~~
To unsubscribe from this group, send email to 
capistrano-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/capistrano
-~--~~~~--~~--~--~---



[Capistrano] Re: Capistrano::ConnectionError

2009-01-15 Thread Jamis Buck

I had heard of it, but I am not a crypto guru, and I have no idea how to
make OpenSSL support those (which is what Net::SSH depends on). Perhaps
someone more crypto-saavy than I will be able to pipe in and say oh,
heck, you just need to do X and Y and you're all set. :)

- Jamis

On 1/15/09 10:02 AM, Xazoola wrote:
 You have just nailed it.
 There was a SSH CBC vulnerability reported by US-CERT.
 http://www.kb.cert.org/vuls/id/958563
 I'm not sure how the vulnerability is exploited. But to be safe, I
 disable all CBC related ciphers.
 After enabling all the default ciphers aes128-ctr,aes128-cbc,arcfour,
 3des-cbc,blowfish-cbc, the encryption_client algorithm is solved.
 
 
 
 do you know anything about this SSH CBC vulnerability? If it is a
 security problem when CBC related cipher enabled on SSH server, any
 plan for Net:SSH to support non CBC ciphers?
 
 On Jan 15, 4:03 pm, Jamis Buck ja...@37signals.com wrote:
 This can also be caused if your SSH server is configured to allow only a
 small subset of cipher algorithms, and that subset does not overlap any
 of the algorithms that Net::SSH supports.

 Net::SSH supports the following ciphers:

   aes128-cbc
   3des-cbc
   blowfish-cbc
   cast128-cbc
   aes192-cbc
   aes256-cbc
   rijndael-...@lysator.liu.se
   idea-cbc
   none

 (Though it is strongly recommended to not use 'none'. It's really only
 useful for debugging Net::SSH.)

 If there is a cipher you want that isn't included there, you'll need to
 see how to make Ruby's OpenSSL bindings provide it.

 - Jamis

 On 1/15/09 8:37 AM, Lee Hambley wrote:

 This might be ssh v1 vs. v2 issues, are you definately using SSH v2,
 with v2 keys? (i think it is a difference between DSA, and RSA keys)
 - Lee
 2009/1/15 Xazoola col...@gmail.com mailto:col...@gmail.com
 Hi,
 I am getting a ConnectionError on Solaris. Works find on Linux. anyone
 know what causes this error?
 Capistrano::ConnectionError, connection failed for: IPADDRESS
 (Net::SSH::Exception: could not settle on encryption_client algorithm)
  


--~--~-~--~~~---~--~~
To unsubscribe from this group, send email to 
capistrano-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/capistrano
-~--~~~~--~~--~--~---



[Capistrano] Re: Capistrano::ConnectionError

2009-01-15 Thread Xazoola

ok, and many thanks for your solution!

On Jan 15, 5:03 pm, Jamis Buck ja...@37signals.com wrote:
 I had heard of it, but I am not a crypto guru, and I have no idea how to
 make OpenSSL support those (which is what Net::SSH depends on). Perhaps
 someone more crypto-saavy than I will be able to pipe in and say oh,
 heck, you just need to do X and Y and you're all set. :)

 - Jamis

 On 1/15/09 10:02 AM, Xazoola wrote:

  You have just nailed it.
  There was a SSH CBC vulnerability reported by US-CERT.
 http://www.kb.cert.org/vuls/id/958563
  I'm not sure how the vulnerability is exploited. But to be safe, I
  disable all CBC related ciphers.
  After enabling all the default ciphers aes128-ctr,aes128-cbc,arcfour,
  3des-cbc,blowfish-cbc, the encryption_client algorithm is solved.

  do you know anything about this SSH CBC vulnerability? If it is a
  security problem when CBC related cipher enabled on SSH server, any
  plan for Net:SSH to support non CBC ciphers?

  On Jan 15, 4:03 pm, Jamis Buck ja...@37signals.com wrote:
  This can also be caused if your SSH server is configured to allow only a
  small subset of cipher algorithms, and that subset does not overlap any
  of the algorithms that Net::SSH supports.

  Net::SSH supports the following ciphers:

    aes128-cbc
    3des-cbc
    blowfish-cbc
    cast128-cbc
    aes192-cbc
    aes256-cbc
    rijndael-...@lysator.liu.se
    idea-cbc
    none

  (Though it is strongly recommended to not use 'none'. It's really only
  useful for debugging Net::SSH.)

  If there is a cipher you want that isn't included there, you'll need to
  see how to make Ruby's OpenSSL bindings provide it.

  - Jamis

  On 1/15/09 8:37 AM, Lee Hambley wrote:

  This might be ssh v1 vs. v2 issues, are you definately using SSH v2,
  with v2 keys? (i think it is a difference between DSA, and RSA keys)
  - Lee
  2009/1/15 Xazoola col...@gmail.com mailto:col...@gmail.com
      Hi,
      I am getting a ConnectionError on Solaris. Works find on Linux. anyone
      know what causes this error?
      Capistrano::ConnectionError, connection failed for: IPADDRESS
      (Net::SSH::Exception: could not settle on encryption_client algorithm)
--~--~-~--~~~---~--~~
To unsubscribe from this group, send email to 
capistrano-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/capistrano
-~--~~~~--~~--~--~---



[Capistrano] Re: Deploying from a specific path with a Git repo

2009-01-15 Thread Henry

Thanks for the reply.  I understand the best practice of 1 repo == 1
project.   But sometimes it just doesn't make sense.  For example, I
have another big rails app that is not ready for deployment yet, but
I'd like to display a semi-functional splash site that uses lots of
the same assets.  So I have a one-file sinatra app sitting in another
directory in the repo rather than creating another repo with duplicate
files for this purpose.  It would be nice in this case just to deploy
that dir (which could have it's own teensy Capfile).

Further complicating things is the fact that you can indeed check out
paths of submodules, which seems like it doesn't fit with the 1 repo
== 1 project philosophy.

But this is not a git discussion board so I'll leave it that, except
to say thanks for Cap, it's a fantastic tool.

On Jan 14, 4:46 pm, Jamis Buck ja...@37signals.com wrote:
 Does does not let you check out a just subdirectory of a repository. If
 you want to deploy just a subdirectory, you'll need to write your own
 tasks that check out the repo locally, and then bundle up the
 subdirectory you wish to deploy.

 Git best-practice is to use one repository for one project.

 - Jamis

 On 1/14/09 11:22 AM, Henry wrote:



  Greetings,

  I have a Git repo that contains a few small sinatra apps. So I would
  like to deploy them one at at time.  So I need to set a path within
  the repo to deploy, but I can't figure out how to do it. This is what
  I've tried:

  set :repository,  g...@github.com:me/myrepo.git/app1

  But that doesn't work.  Any suggestions?

  Thanks!
--~--~-~--~~~---~--~~
To unsubscribe from this group, send email to 
capistrano-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/capistrano
-~--~~~~--~~--~--~---



[Capistrano] Re: Deploying from a specific path with a Git repo

2009-01-15 Thread David Masover
On Thu, Jan 15, 2009 at 11:44 AM, Henry hpoy...@gmail.com wrote:


 ...I have another big rails app that is not ready for deployment yet, but
 I'd like to display a semi-functional splash site that uses lots of
 the same assets.  So I have a one-file sinatra app sitting in another
 directory in the repo rather than creating another repo with duplicate
 files for this purpose.


I realize that is the easier way of doing it, but I wouldn't call it a best
practice.

On the last major project I had, we had our own from-scratch-and-plugins
Rails app, and a Mephisto implementation, a Spree implementation, and a
Castronaut implementation.

We did two things:

First, we did actually duplicate the files between them, when necessary. In
this case, the original Rails app was on SVN, and everything else was on
Git.

Second, we used a separate asset server -- really the same machine, but it
could be anything, and we were considering doing that with Amazon S3. So we
built a separate project which was just the assets, and wrote a deploy task
that just ran deploy:update_code and deploy:symlink -- then configured nginx
to serve files directly from there, statically.

Bonus: Browser caching works, now, since the two apps will be pointing to
the same URLs, not just copies of the same file.

And finally, for the things which were too hard to do that with, or which we
simply didn't want to bother changing yet -- for example, a common layout
file -- we put that in its own repository, and used submodules.

I don't know how well this maps onto the default tasks, though. I remember
writing a lot of my own -- probably more than I needed to, but enough that
I'm sure if I wanted to do what you're suggesting, I could have pretty
easily.

Further complicating things is the fact that you can indeed check out
 paths of submodules, which seems like it doesn't fit with the 1 repo
 == 1 project philosophy.


Well, a submodule is sort of an alternative to using symlinks. I think it
does make sense that it's possible to map a subdirectory to a subdirectory
-- you might not need the entire other project, or your project might be
extremely sensitive to directory structure.

--~--~-~--~~~---~--~~
To unsubscribe from this group, send email to 
capistrano-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/capistrano
-~--~~~~--~~--~--~---



[Capistrano] Re: Deploying from a specific path with a Git repo

2009-01-15 Thread Wincent Colaiuta

On 15 ene, 18:44, Henry hpoy...@gmail.com wrote:
 Thanks for the reply.  I understand the best practice of 1 repo == 1
 project.   But sometimes it just doesn't make sense.  For example, I
 have another big rails app that is not ready for deployment yet, but
 I'd like to display a semi-functional splash site that uses lots of
 the same assets.  So I have a one-file sinatra app sitting in another
 directory in the repo rather than creating another repo with duplicate
 files for this purpose.  It would be nice in this case just to deploy
 that dir (which could have it's own teensy Capfile).

Seeing as you have two versions of the same thing, sounds like a job
for branches: master for the Rails implementation, in development;
and sinatra for the placeholder.

At least, that's the way I would do it.

If they weren't two versions of the same thing but actually two
different apps then I would do what Jamis says and have two repos.

Wincent


--~--~-~--~~~---~--~~
To unsubscribe from this group, send email to 
capistrano-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/capistrano
-~--~~~~--~~--~--~---