Logging is a feature that provides the capability to send detailed logs to several kinds of 3rd party external log aggregation services. Services connected to this data feed serve as a useful means in gaining insight into Tower usage or technical trends. The data can be used to analyze events in the infrastructure, monitor for anomalies, and correlate events from one service with events in another. The types of data that are most useful to Tower are job fact data, job events/job runs, activity stream data, and log messages. The data is sent in JSON format over a HTTP connection using minimal service-specific tweaks engineered in a custom handler or via an imported library.
Installing Ansible Tower will install a newer version of rsyslog, which will replace the version that comes with the RHEL base. The version of rsyslog that is installed by Ansible Tower does not include the following rsyslog modules:
After installing Ansible Tower, use only the Tower provided rsyslog package for any logging outside of Tower that may have previously been done with the RHEL provided rsyslog package. If you already use rsyslog for logging system logs on Tower instances, you can continue to use rsyslog to handle logs from outside of Tower by running a separate rsyslog process (using the same version of rsyslog that Tower is), and pointing it to a separate /etc/rsyslog.conf.
For systems that use rsyslog outside of Tower (on the Tower VM/machine), consider any conflict that may arise with also using new version of rsyslog that comes with Tower.
You can configure from the
/api/v2/settings/logging/ endpoint how the Tower rsyslog process handles messages that have not yet been sent in the event that your external logger goes offline:
LOG_AGGREGATOR_MAX_DISK_USAGE_GB: specifies the amount of data to store (in gigabytes) during an outage of the external log aggregator (defaults to 1). Equivalent to the
rsyslogd queue.maxdiskspace setting.
LOG_AGGREGATOR_MAX_DISK_USAGE_PATH: specifies the location to persist logs that should be retried after an outage of the external log aggregator (defaults to
/var/lib/awx). Equivalent to the
rsyslogd queue.spoolDirectory setting.
For example, if Splunk goes offline, rsyslogd stores a queue on the disk until Splunk comes back online. By default, it will store up to 1GB of events (while Splunk is offline) but you can make that more than 1GB if necessary, or change the path where you save the queue.
Below are special loggers (except for
awx, which constitutes generic server logs) that provide large amount of information in a predictable structured or semi-structured format, following the same structure as one would expect if obtaining the data from the API:
job_events: Provides data returned from the Ansible callback module
activity_stream: Displays the record of changes to the objects within the Ansible Tower application
system_tracking: Provides fact data gathered by Ansible
setup module (i.e.
gather_facts: True) when job templates are ran with Enable Fact Cache selected
awx: Provides generic server logs, which include logs that would normally be written to a file. It contains the standard metadata that all logs have, except it only has the message from the log statement.
These loggers only use log-level of INFO, except for the
awx logger, which may be any given level.
Additionally, the standard Tower logs are be deliverable through this same mechanism. It is apparent how to enable or disable each of these five sources of data without manipulating a complex dictionary in your local settings file, as well as adjust the log-level consumed from the standard Tower logs.
To configure various logging components in Ansible Tower, select System from the () menu located on the left navigation bar.
Common schema for all loggers:
cluster_host_id: Unique identifier of the host within the Tower cluster
level: Standard python log level, roughly reflecting the significance of the event All of the data loggers as a part of this feature use INFO level, but the other Tower logs will use different levels as appropriate
logger_name: Name of the logger we use in the settings, for example, “activity_stream”
@timestamp: Time of log
path: File path in code where the log was generated
(common): This uses all the fields common to all loggers listed above
actor: Username of the user who took the action documented in the log
changes: JSON summary of what fields changed, and their old/new values.
operation: The basic category of the changed logged in the activity stream, for instance, “associate”.
object1: Information about the primary object being operated on, consistent with what we show in the activity stream
object2: If applicable, the second object involved in the action
This logger reflects the data being saved into job events, except when they would otherwise conflict with expected standard fields from the logger, in which case the fields are nested. Notably, the field host on the
job_event model is given as
event_host. There is also a sub-dictionary field,
event_data within the payload, which contains different fields depending on the specifics of the Ansible event.
This logger also includes the common fields.
These contain a detailed dictionary-type fields that are either services, packages, or files.
(common): This uses all the fields common to all loggers listed above
services: For services scans, this field is included and has keys based on the name of the service. NOTE: Periods are disallowed by elastic search in names, and are replaced with “_” by our log formatter
package: Included for log messages from package scans
files: Included for log messages from file scans
host: Name of host scan applies to
inventory_id: Inventory id host is inside of
This is a intended to be a lower-volume source of information about changes in job states compared to job events, and also intended to capture changes to types of unified jobs other than job template based jobs.
In addition to common fields, these logs include fields present on the job model.
In addition to the common fields, this contains a
msg field with the log message. Errors contain a separate
traceback field. These logs can be enabled or disabled in the Configure Tower User Interface
ENABLE EXTERNAL LOGGING setting.
The logging aggregator service works with the following monitoring and data analysis systems:
Ansible Tower’s Splunk logging integration uses the Splunk HTTP Collector. When configuring a SPLUNK logging aggregator, add the full URL to the HTTP Event Collector host, like in the following example:
Splunk HTTP Event Collector listens on 8088 by default so it is necessary to provide the full HEC event URL (with port) in order for incoming requests to be processed successfully. These values are entered in the example below:
For further instructions on configuring the HTTP Event Collector, refer to the Splunk documentation.
To set up the sending of logs through Loggly’s HTTP endpoint, refer to https://www.loggly.com/docs/http-endpoint/. Loggly uses the URL convention described at http://logs-01.loggly.com/inputs/TOKEN/tag/http/, which is shown inputted in the Logging Aggregator field in the example below:
In Sumologic, create a search criteria containing the json files that provide the parameters used to collect the data you need.
Standing up your own version the elastic stack requires no modification to the logstash
Backward-incompatible changes were introduced with Elastic 5.0.0, and different configurations may be required depending on what versions you are using.
To set up logging to any of the aggregator types:
Click the Settings () icon from the left navigation bar.
In the System screen, select the Logging tab.
Set the configurable options from the fields provided:
Enable External Logging: Click the toggle button to ON if you want to send logs to an external log aggregator.
Logging Aggregator: Enter the hostname or IP address you want to send logs.
Logging Aggregator Port: Specify the port for the aggregator if it requires one.
When the connection type is HTTPS, you can enter the hostname as a URL with a port number and therefore, you are not required to enter the port again. But TCP and UDP connections are determined by the hostname and port number combination, rather than URL. So in the case of TCP/UDP connection, supply the port in the specified field. If instead a URL is entered in host field (Logging Aggregator field), its hostname portion will be extracted as the actual hostname.
Logging Aggregator Type: Click to select the aggregator service from the drop-down menu:
Logging Aggregator Username: Enter the username of the logging aggregator if it requires it.
Logging Aggregator Password/Token: Enter the password of the logging aggregator if it requires it.
Loggers to Send Data to the Log Aggregator Form: All four types of data are pre-populated by default. Click the tooltip icon next to the field for additional information on each data type. Delete the data types you do not want.
Log System Tracking Facts Individually: Click the tooltip icon for additional information whether or not you want to turn it on, or leave it off by default.
Logging Aggregator Protocol: Click to select a connection type (protocol) to communicate with the log aggregator. Subsequent options vary depending on the selected protocol.
TCP Connection Timeout: Specify the connection timeout in seconds. This option is only applicable to HTTPS and TCP log aggregator protocols.
Logging Aggeregator Level Threshold: Select the level of severity you want the log handler to report.
Enable/Disable HTTPS Certificate Verification: Certificate verification is enabled by default for HTTPS log protocol. Click the toggle button to OFF if you do not want the log handler to verify the HTTPS certificate sent by the external log aggregator before establishing a connection.
Review your entries for your chosen logging aggregation. Below is an example of one set up for Splunk:
When done, click Save to apply the settings or Cancel to abandon the changes.
To verify if your configuration is set up correctly, click Save first then click Test. This sends a test log message to the log aggregator using the current logging configuration in Ansible Tower. You should check to make sure this test message was received by your external log aggregator.
If the Test button is disabled, it is an indication that the fields are different than their initial values so save your changes first, and make sure the Enable External Logging toggle is set to ON.
If you have sent a message with the test button to your configured logging service via http/https, but did not receive the message, check the
/var/log/tower/rsyslog.err log file. This is where errors are stored if they occurred when authenticating rsyslog with an http/https external logging service. Note that if there are no errors, this file will not exist.