Default the agent to a 1 minute collection time.

Change-Id: I1be628ad6076069784b58224b3044ce3a23973fc
This commit is contained in:
Tim Kuhlman 2015-03-30 10:41:43 -06:00
parent 1c4f93f7f1
commit 2b5fc32f2b
2 changed files with 3 additions and 3 deletions

View File

@ -504,7 +504,7 @@ instances:
* service_name is the name of the metric
* check_command is the full command to run. Specifying the full path is optional if the checks are located somewhere in check_path. These above examples are a copy-and-paste from existing service checks in /etc/cron.d/servicecheck-* files, so migration is fairly easy.
* check_interval (optional) If unspecified, the checks will be run at the regular collector interval, which is 15 seconds by default. You may not want to run some checks that frequently, especially if they are resource-intensive, so check_interval lets you force a delay, in seconds, between iterations of that particular check. The state for these are stored in temp_file_path with file names like nagios_wrapper_19fe42bc7cfdc37a2d88684013e66c7b.pck where the hash is an md5sum of the service_name (to accommodate odd characters that the filesystem may not like).
* check_interval (optional) If unspecified, the checks will be run at the regular collector interval, which is 60 seconds by default. You may not want to run some checks that frequently, especially if they are resource-intensive, so check_interval lets you force a delay, in seconds, between iterations of that particular check. The state for these are stored in temp_file_path with file names like nagios_wrapper_19fe42bc7cfdc37a2d88684013e66c7b.pck where the hash is an md5sum of the service_name (to accommodate odd characters that the filesystem may not like).
## Host Alive Checks
An extension to the Agent can provide basic "aliveness" checks of other systems, verifying that the remote host (or device) is online. This check currently provides two methods of determining connectivity:
@ -1133,7 +1133,7 @@ The Monasca Agent ships with a Statsd daemon implementation called monasca-stats
monascastatsd will accept metrics submitted by functions in either the standard statsd Python client library, or the monasca-agent's [monasca-statsd Python client library](https://github.com/stackforge/monasca-statsd). The advantage of using the python-monasca-statsd library is that it is possible to specify dimensions on submitted metrics. Dimensions are not handled by the standard statsd client.
Statsd metrics are not bundled along with the metrics gathered by the Collector, but are flushed to the agent Forwarder on a separate schedule (every 10 seconds by default, rather than 15 seconds for Collector metrics).
Statsd metrics are not bundled along with the metrics gathered by the Collector, but are flushed to the agent Forwarder on a separate schedule (every 10 seconds by default, rather than 60 seconds for Collector metrics).
Here is an example of metrics submitted using the standard statsd Python client library.

View File

@ -69,7 +69,7 @@ def main(argv=None):
parser.add_argument('--project_id', help="Keystone project id for keystone authentication", required=False, default='')
parser.add_argument('--ca_file', help="Sets the path to the ca certs file if using certificates. " +
"Required only if insecure is set to False", required=False, default='')
parser.add_argument('--check_frequency', help="How often to run metric collection in seconds", type=int, default=15)
parser.add_argument('--check_frequency', help="How often to run metric collection in seconds", type=int, default=60)
parser.add_argument('--config_dir', help="Configuration directory", default='/etc/monasca/agent')
parser.add_argument('--dimensions', help="Additional dimensions to set for all metrics. A comma seperated list " +
"of name/value pairs, 'name:value,name2:value2'")