3

I have two machines, puppet master--host name puppet--and a lone client, host name git. The puppet agent on the master machine works without issue. The agent on git fails with a '400 No required SSL certificate was sent'. First, the configuration of the puppet master, which is a thin/nginx affair:

puppet:~# ruby -v
ruby 1.9.2p0 (2010-08-18 revision 29036) [i486-linux]

puppet:~# puppet --version
2.7.9

puppet:~# cat /etc/nginx/sites-enabled/default 
server {
  listen puppet:8140;

  ssl on;
  ssl_certificate /var/lib/puppet/ssl/certs/puppet.pem;
  ssl_certificate_key /var/lib/puppet/ssl/private_keys/puppet.pem;
  ssl_ciphers ALL:-ADH:+HIGH:+MEDIUM:-LOW:-SSLv2:-EXP;
  ssl_client_certificate  /var/lib/puppet/ssl/ca/ca_crt.pem;
  ssl_verify_client on;

  proxy_redirect   off;
  proxy_set_header Host             $host;
  proxy_set_header X-Real-IP        $remote_addr;
  proxy_set_header X-Forwarded-For  $proxy_add_x_forwarded_for;
  proxy_set_header X-Client-Verify  $ssl_client_verify;
  proxy_set_header X-Client-Verify  SUCCESS;
  proxy_set_header X-Client-DN      $ssl_client_s_dn;
  proxy_set_header X-SSL-Subject    $ssl_client_s_dn;
  proxy_set_header X-SSL-Issuer     $ssl_client_i_dn;

  default_type application/x-raw;

  location /production/file_content/ {
    rewrite ^/production/file_content/modules/([^/]+)/(.*) /$1/files/$2;
    break;
    root /etc/puppet/modules/;
  }
  location / {
    proxy_pass http://puppet-production;
  }
}


# cat /etc/nginx/conf.d/puppet-production-upstream.conf 
upstream puppet-production {  
  server unix:/var/run/puppet/master.00.sock;
  server unix:/var/run/puppet/master.01.sock;
  server unix:/var/run/puppet/master.02.sock;
}


puppet:~# cat /etc/supervisor/conf.d/puppetmaster.conf 
# This file is autogenerated by Puppet. Manual changes will be overwritten!
[program:puppetmaster]
command=/usr/bin/thin start -e development --socket /var/run/puppet/master.%(process_num)02d.sock --user puppet --group puppet --chdir /etc/puppet -R /etc/puppet/config.ru
process_name=%(program_name)s_%(process_num)02d
numprocs=3
priority=999
autostart=true
autorestart=unexpected
startsecs=3
startretries=3
exitcodes=0,2
stopsignal=TERM
stopwaitsecs=10
redirect_stderr=false
stdout_logfile=/var/log/supervisor/puppetmaster/puppetmaster.out
stdout_logfile_maxbytes=250MB
stdout_logfile_backups=10
stderr_logfile=/var/log/supervisor/puppetmaster/puppetmaster.err
stderr_logfile_maxbytes=250MB
stderr_logfile_backups=10

puppet:~# cat /etc/puppet/puppet.conf 
[main]
ssldir=$vardir/ssl

[master]
certname=puppet

Applying the work-around here I'm able to get the git agent only to this point when attempting to introduce git to puppet master:

git:~# puppet agent --waitforcert 30 --test
err: Could not request certificate: Error 400 on SERVER: <html>
<head><title>400 No required SSL certificate was sent</title></head>
<body bgcolor="white">
<center><h1>400 Bad Request</h1></center>
<center>No required SSL certificate was sent</center>
<hr><center>nginx/1.1.8</center>
</body>
</html>

This resource suggests in its Simulate SSL Connection section to run, from my git box:

openssl s_client -host puppet -port 8140 -cert /var/lib/puppet/ssl/certs/git.troutwine.us.pem -key /var/lib/puppet/ssl/private_keys/git.troutwine.us.pem -CAfile /var/lib/puppet/ssl/certs/ca.pem

The problem with this being that I'm missing /var/lib/puppet/ssl/certs/git.troutwine.us.pem:

git:~# tree /var/lib/puppet/ssl/
/var/lib/puppet/ssl/
├── certificate_requests
├── certs
│   └── ca.pem
├── private
├── private_keys
│   └── git.troutwine.us.pem
└── public_keys
    └── git.troutwine.us.pem

Plain old webrick puppetmasterd works just fine--it's only the nginx/puppet combination that's failing me. Both machines are running ntpd and have an acceptable time spread. What am I doing wrong?

4
  • Use the one in public_keys instead - it's already been signed by the puppet CA, correct? Feb 2, 2012 at 20:42
  • No, I'll make this more clear in the body of the question, but the first puppet agent command you see for git is the attempt to introduce the git box to puppet master.
    – troutwine
    Feb 2, 2012 at 20:58
  • Make sure the date&time in both machines are in close sync. Feb 2, 2012 at 21:49
  • They are. I've ntp running on both machines. The time difference is ~200ms.
    – troutwine
    Feb 2, 2012 at 22:10

2 Answers 2

2

newl in the #puppet channel suggested modifying ssl_verify_client to be 'optional', rather than 'on'. I've done this and everything's peachy now.

I had myself convinced that this was a Bad Thing to do, but after newl's suggestion I couldn't recall why. If anyone does believe this to be a less than ideal configuration setting, do let me know.

1

If you run a puppet agent on the master, you should make sure that they do NOT share SSL directories. I have seen Weird Stuff result from that configuration.

Snippets from my /etc/puppet/puppet.conf:

[main]
# Where SSL certificates are kept for the puppet master and other
# subcommands.
# Note that this is a global setting because most of the subcommands
# other than 'agent' are only valid in puppetmaster context.
# The default value is '$confdir/ssl'.
vardir = /var/lib/puppetmaster
ssldir = $vardir/ssl

[agent]
# The var & SSL dir for the agent; listed explicitly because the master
# and other subcommands intended for the master should use
# the different SSL state.
# The default value is '$confdir/ssl'.
vardir = /var/lib/puppet
ssldir = $vardir/ssl
1
  • That didn't end up being the problem, but thank you. I'll modify my environment.
    – troutwine
    Feb 3, 2012 at 18:25

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .