| popularity | question | answers | views | user | |
|---|---|---|---|---|---|
| 8 |
TypeError: can't convert Array into String - ruby 1.9.2
Hey all, I’m trying to setup a server with ruby 1.9.2 instead of the default chef version. I’ve got the bootstrap file in chef-repo/boostrap/ubuntu10.04_ruby192.erb.
|
3 | 4.8k | ||
| 6 |
Run chef-client on the new client node
FATAL: 1 node found, but do not have the required attribute to stablish the connection. Try setting another attribute to open the connection using —attribute. |
3 | 1.9k | ||
| 4 |
Cannot install chef-server 11 on centos 6
Hello, trying to install chef-server 11 on centos 6 running the provided installer, but getting the following error message. I am stuck. Anyone can help? [2013-02-13T16:49:30+01:00] FATAL: Mixlib::ShellOut::ShellCommandFailed: execute[/opt/chef-server/embedded/bin/chpst -u chef_server -U chef_server /opt/chef-server/embedded/bin/rabbitmqctl wait /var/opt/chef-server/rabbitmq/db/rabbit@localhost.pid] (chef-server::rabbitmq line 80) had an error: Mixlib::ShellOut::ShellCommandFailed: Expected process to exit with [0], but received ‘2’ —— Begin output of /opt/chef-server/embedded/bin/chpst -u chef_server -U chef_server /opt/chef-server/embedded/bin/rabbitmqctl wait /var/opt/chef-server/rabbitmq/db/rabbit@localhost.pid —— STDOUT: Waiting for rabbit@localhost … pid is 2586 … STDERR: Error: process_not_running —— End output of /opt/chef-server/embedded/bin/chpst -u chef_server -U chef_server /opt/chef-server/embedded/bin/rabbitmqctl wait /var/opt/chef-server/rabbitmq/db/rabbit@localhost.pid —— Ran /opt/chef-server/embedded/bin/chpst -u chef_server -U chef_server /opt/chef-server/embedded/bin/rabbitmqctl wait /var/opt/chef-server/rabbitmq/db/rabbit@localhost.pid returned 2 Thanks, claudio |
0 | 420 | ||
| 4 |
Cannot get updated ipaddress for EC2 instance started after stopped
I’m developing a recipe targeting to get the update IP address for an EC2 instance as it’s just started. To achieve this, I make chef-client a system service which located under /etc/init.d/chef-client (this is default when ec2 instance is bootstrapped) and assign the run_list to the node. However, as a result I found that node[‘ipaddress’] still gets its old address rather than the new one. The reason might because chef node retrieved the value from chef server directly which still have old value. And if I run chef-client on the node again, node[‘ipaddress’] can get the updated ip value. My question is 1. How to get updated ipaddress for the first time when the chef-client gets started as a system service. 2. If node[‘ipaddress’] get value from chef-server then how to get ip address from ohai run? I ask this because I observe that ohai always output correct value. 3. Is there any way to delay chef-client execution as a system service? Greatly appreciate your comments! |
1 | 1.0k | ||
| 4 |
Sharing code, definitions and other helpers between cookbook repos
I’m writing a couple of cookbook repos and would like to share some recipes and helpers (like definitions/* of cookbooks) between them. As I understand from CHEF-2308 the file-by-file merging and shadowing feature of the site-cookbooks directory is deprecated. What would a DRY and simple to understand approach to sharing cookbook code be? I’m pushing my cookbooks with rsync and running them with chef-solo, so I’m not bound by what the current Chef server supports. I’m thinking of something modular like the site-cookbooks was, so that it’d be clear what helpers are shared and what’s repo specific. I’m trying to avoid mixing them into one file/directory like ‘knife cookbook vendor’ does with its git branch and merge approach. Cheers |
1 | 2.4k | ||
| 3 |
Problem with deploy_revision and symlinks
Edit: Is this a bug or intended behavior somehow? I’ve not used Capistrano before and am not sure how closely deploy / deploy_revision fits into its paradigm. I’m having a similar problem as described at this SO post where Chef is not able to create symlinks in deploy_revision. It seems as though Chef tries to create the symlinks to the releases/{HASH} directory instead of the repository_cache folder, since it’s done pre-migration. I can only get the error to go away if I manually create the hash release folder prior to running chef-client.
I have created the shared/config.json file prior to deploy_revision being run. You can see if I create the releases/98faa472067f4fb120d97a7cf0eb486a1435f2c9 directory before running chef-client, there is no error:
How am I supposed to make symlinks before the migration with deploy_revision? FWIW, using symlink or symlink_before_migrate seems to make no difference. Interestingly, using the force_deploy action fixes this problem, but I really don’t want to have my code redeploy every time chef runs if I’m not upgrading the git reference point. Is this a bug in chef-client or am I not understanding the correct usage of deploy_revision? Edit: here’s the compiled resource that fails without making that directory first:
|
0 | 175 | ||
| 3 |
Abandoned cookbooks
There seem to be cookbooks that have been uploaded at one point in time, and have never been updated, nor respond to any inquiries regarding updates. I have taken on primary maintenance of one such cookbook, have updated it considerably, and would like to share it back. Is there a method or process to determine ‘ownership’ of a given cookbook, and delegate that to another interested party if it is neglected? Thanks, -M |
0 | 361 | ||
| 3 |
Why does my profile picture not update?
I changed my profile picture, but it didn’t update in the various thumbnails where it should, it just looks like the faceless Opscode guy. |
2 | 1.9k | ||
| 3 |
How to find out `os` name?
Where can i find the complete list of |
1 | 1.7k | ||
| 3 | 3 | 2.0k |
