Monitoring chef-clients

Chef Infra (archive)
1

Folks,

is there any monitoring solution, that allows to collect statuses of
chef-client runs? So, for example, if cookbook was broken I’ll see this
immediately?

Thanks,
Kirill.

2

You could register a chef exception handler[0] that simply makes an API call to register the failure with whatever monitoring system you use.

[0] http://docs.opscode.com/essentials_handlers_available_handlers.html

On Friday, February 8, 2013 at 4:48 PM, Kirill Timofeev wrote:

Folks,

is there any monitoring solution, that allows to collect statuses of
chef-client runs? So, for example, if cookbook was broken I'll see this
immediately?

Thanks,
Kirill.

3

I've got a cookbook posted that integrates the Chef error handler system with Sentry, and I think there is a similar one for Hoptoad.

--Noah

On Feb 8, 2013, at 4:48 PM, Kirill Timofeev wrote:

Folks,

is there any monitoring solution, that allows to collect statuses of chef-client runs? So, for example, if cookbook was broken I'll see this immediately?

Thanks,
Kirill.

4

We use Sensu to check the status of the chef-client log for
exceptions, and also the mtime of the file.

Cheers,

AJ

On 9 February 2013 13:52, Noah Kantrowitz noah@coderanger.net wrote:

I've got a cookbook posted that integrates the Chef error handler system with Sentry, and I think there is a similar one for Hoptoad.

--Noah

On Feb 8, 2013, at 4:48 PM, Kirill Timofeev wrote:

Folks,

is there any monitoring solution, that allows to collect statuses of chef-client runs? So, for example, if cookbook was broken I'll see this immediately?

Thanks,
Kirill.

5

There's a few handlers out there for throwing chef errors to jabber or IRC. That's a great way to know you broke something.

On Feb 8, 2013, at 5:10 PM, AJ Christensen aj@junglist.gen.nz wrote:

We use Sensu to check the status of the chef-client log for
exceptions, and also the mtime of the file.

Cheers,

AJ

On 9 February 2013 13:52, Noah Kantrowitz noah@coderanger.net wrote:

I've got a cookbook posted that integrates the Chef error handler system with Sentry, and I think there is a similar one for Hoptoad.

--Noah

On Feb 8, 2013, at 4:48 PM, Kirill Timofeev wrote:

Folks,

is there any monitoring solution, that allows to collect statuses of chef-client runs? So, for example, if cookbook was broken I'll see this immediately?

Thanks,
Kirill.

6

Hi Kirill,

I am working on a service that I am about to open to wider testing
that shows the last run status of all your nodes, shows the exception
if one occurred, and emails you the first time a new exception occurs

Sent from my iPhone

On 09/02/2013, at 11:48 AM, Kirill Timofeev kvt@hulu.com wrote:

Folks,

is there any monitoring solution, that allows to collect statuses of chef-client runs? So, for example, if cookbook was broken I'll see this immediately?

Thanks,
Kirill.

7

We just started using the splunk handler, and it makes watching for errors
and setting email triggers a snap.
See here for more:
http://docs.opscode.com/community_plugin_report_handler.html
On Feb 8, 2013 8:26 PM, "Alan Harper" alan@aussiegeek.net wrote:

Hi Kirill,

I am working on a service that I am about to open to wider testing
that shows the last run status of all your nodes, shows the exception
if one occurred, and emails you the first time a new exception occurs

Sent from my iPhone

On 09/02/2013, at 11:48 AM, Kirill Timofeev kvt@hulu.com wrote:

Folks,

is there any monitoring solution, that allows to collect statuses of
chef-client runs? So, for example, if cookbook was broken I'll see this
immediately?

Thanks,
Kirill.

8

We use the exception handler to submit an error via Nagios' NSCA.
Basically the same thing that everyone else does.

On Fri, Feb 8, 2013 at 11:29 PM, Jesse Campbell hikeit@gmail.com wrote:

We just started using the splunk handler, and it makes watching for errors
and setting email triggers a snap.
See here for more:
http://docs.opscode.com/community_plugin_report_handler.html
On Feb 8, 2013 8:26 PM, "Alan Harper" alan@aussiegeek.net wrote:

Hi Kirill,

I am working on a service that I am about to open to wider testing
that shows the last run status of all your nodes, shows the exception
if one occurred, and emails you the first time a new exception occurs

Sent from my iPhone

On 09/02/2013, at 11:48 AM, Kirill Timofeev kvt@hulu.com wrote:

Folks,

is there any monitoring solution, that allows to collect statuses of
chef-client runs? So, for example, if cookbook was broken I'll see this
immediately?

Thanks,
Kirill.

--
Ian Marlier | Manager of Platform Operations
Brightcove, Inc.
290 Congress Street, 4th Floor, Boston, MA 02110
imarlier@brightcove.com

9

Hi,

I use a very simple Pushover[1] handler to get notification on my
phone when a Chef run fails:

[1] https://pushover.net/

On Mon, Feb 11, 2013 at 2:30 AM, Ian Marlier imarlier@brightcove.com wrote:

We use the exception handler to submit an error via Nagios' NSCA. Basically
the same thing that everyone else does.

On Fri, Feb 8, 2013 at 11:29 PM, Jesse Campbell hikeit@gmail.com wrote:

We just started using the splunk handler, and it makes watching for errors
and setting email triggers a snap.
See here for more:
http://docs.opscode.com/community_plugin_report_handler.html

On Feb 8, 2013 8:26 PM, "Alan Harper" alan@aussiegeek.net wrote:

Hi Kirill,

I am working on a service that I am about to open to wider testing
that shows the last run status of all your nodes, shows the exception
if one occurred, and emails you the first time a new exception occurs

Sent from my iPhone

On 09/02/2013, at 11:48 AM, Kirill Timofeev kvt@hulu.com wrote:

Folks,

is there any monitoring solution, that allows to collect statuses of
chef-client runs? So, for example, if cookbook was broken I'll see this
immediately?

Thanks,
Kirill.

--
Ian Marlier | Manager of Platform Operations
Brightcove, Inc.
290 Congress Street, 4th Floor, Boston, MA 02110
imarlier@brightcove.com

--
Juanje