The best and most efficient way to analyze FusionReactor data is to use FusionAnalytics.
FusionAnalytics is an application performance analysis tool, which integrates seamlessly with FusionReactor. FusionAnalytics is all about giving you the data to help you improve the performance and quality of your applications as well as measure exactly how your applications are performing on a continuous day-to-day basis.
The Analytics menu provides access to the configuration of the FusionAnalytics Connector, a system which provides automatic log shipping between FusionReactor instances and FusionAnalytics.
FusionReactor logs contain very detailed metrics on various aspects of a J2EE container: request runtimes, memory loads, CPU occupancy, JDBC activity and CPU-Thread usage, to name but a few. FusionReactor derives immediate value from this data by way of sophisticated charting and tables. Getting a longer perspective - to allow resource planning, trend discovery etc. - is outside the scope of FusionReactor, but is well suited to the FusionAnalytics product.
FusionReactor, using the FusionAnalytics Connector, is able to ship logs automatically to a FusionAnalytics DataCollector, where it can be easily stored in a database and made available for a comprehensive, extensive set of analytics operations.
More information on FusionAnalytics is available here.
This page contains details about the logging and archiving system employed by FusionReactor. It also contains extended details about the FusionAnalytics workers and Connector status.
|Set Home Page||Will set the current users home page to this page, making this the first page that is displayed on login, for that user.|
|Rotate Now||This will instruct FusionReactor to start the log rotation.
Note: This can take some time, and the archive will only appear in the list after the log rotation has completed.
|Purge Now||This will instruct FusionReactor to start the purge process.
Note: This will not affect the current days logs.
|Clear All Archives||This will clear all the archives from the disk, for this instance.|
This will delete ALL archives from the system.
Make sure you are 100% sure before removing all archives.
Under the page header and controls, there is a some information about the current state of the logging system. This Displays the next scheduled rotation time and period, as well as the current state of the rotation engine.
View Logs from a Different Date¶
To view the archives generated on another date, use the date picker to change the date.
The First table on this page is the Archive Status table, this table shows a summary of information about each log archive.
This table has the Transfer All button this button will instruct FusionReactor to transfer all the displayed logs to the FusionAnalytics Targets. This will send ALL the log archives for the selected date, it is then up to the data collector in FusionAnalytics to decided to import those logs or not.
Table Column Headings¶
|Archive File||This is the name of the archive.|
|Size||This is the size in of the archive (MB)|
|Created||This displays the date and time the archive was created.|
|Source||The name of the server that the logs are based on.|
|Archive Comment||This is explanatory text created by the subsystem that created the archive.|
|Keep||If the text reads Don't Keep then that archive will be removed after a purge has taken place. If it reads Keep then the archive will remain on the drive after a purge has taken place. Simply click the text to change between the two.|
|FusionAnalytics Connector Lifestyle||Lists where the logs are generated as well as what they were processed by.|
|FusionAnalytics Connector Status||This will show the status of the connection, No Exception means everything is OK.|
|Buttons||Transfer - This will transfer this archive to all of the FusionAnalytics Targets. To set up a target, visit the FusionAnalytics Targets page
Details - This will open a detailed view of the archive. (see Archive Details)
Delete - This will delete the archive from the disk.
Archival System Status¶
The second table displays a summary of the current sizes of the Archive System. This is only updated on user request, use the Refresh button above this table to collect the current sizes of the archives.
|Archived File Size||This is the current size of all the .zip files on this disk.|
|Housekeeping Size||This is the size of all the property files used by the system.|
|Total Archive Size||This is the total size of all the property and zip files.|
FusionAnalytics Connector Status¶
This table displays the current status of the FusionAnalytics Connector threads.
|Worker||This is the name of the worker thread.|
|Status||This displays the current status of this thread.|
|Current Archive||This shows the current archive that is being processed by this thread,|
|Targets||This lists the targets that the currently processed archive is being sent too, Also displays the progress for each target.|
FusionAnalytics Connector Transfer Queue¶
This table will display any Archives that are currently waiting to be processed by the workers,
|Archive||This is the name of the archive file that is waiting to be processed.|
|Targets||This is the list of targets that this archive is waiting to be sent to.|
|Buttons||Delete - This will remove the Archive from the queue
History - This will open the extended details for this archive.
To Head - This will move this archive to the front of the queue.
FusionAnalytics DataCollector Targets¶
Add FusionAnalytics Targets¶
Enabling the FusionAnalytics Connector
To use the FusionAnalytics connector it first needs to be enabled. To do this go to Analytics > FusionAnalytics Settings in FusionReactor and change the FusionAnalytics Connect Mode to Sent to targets by this instance, Sent to targets by FRAM or Sent by this instance (uncompressed), for more information on what these settings do please see FusionAnalytics Settings.
Here you can add new FusionAnalytics targets and see you existing targets.
Adding a new target¶
Click the Add New Target button and you will be presented with the Add FusionAnalytics Target dialog, which contains the following fields:
If you don't know where to find the information for the fields, see this.
|Host||The machine where FusionAnalytics is installed. This example uses the local machine, so the IP address is 127.0.0.1 or localhost.|
|Port||The FusionAnalytics port. If you changed the default port during the installation, use that instead; if not, use the default (8400).|
|Application Name||The name of the DataCollector Application that you want to collect the data. The default for this field is the FusionReactor instance name.|
|DataCollector||The name of the actual DataCollector that you want to receive the data. Each DataCollector application can have multiple DataCollectors that receive data. The default setting is one DataCollector named faDataCollector-1.|
|Username||The username for a user who has access to the DataCollector application. This is only used if the DataCollector requires authentication. If you are using the default setting, this is not needed.|
|Password||The password for the above username.|
|HTTPS||Check this box if your DataCollector uses HTTPS. For more information on this see Setting Up FusionAnalytics with HTTPS.|
Click Save Target Settings. The target will be entered into the existing targets table and if the connection was successful the status column will display:
- 200 OK
Editing, Verifying and Deleting targets¶
On the FusionAnalytics Targets page you will see your existing targets listed in a table below the Add New Target button. The columns of the table are defined as follows:
|URL||The url that FusionReactor uses to connect to this FusionAnalytics DataCollector. This is automatically constructed from the fields in the Add New Target dialog.|
|Username||The username that is used by FusionReactor to connect to this DataCollector. A username and password can be set up for a DataCollector inside FusionAnalytics. See Username and Password|
|Status||When the test button is clicked, the resulting status of the target will be displayed here. Clicking on the status text will display some more detailed information about the test including any exceptions, response text, test duration and a few others.|
There are three buttons to help you manage your DataCollector targets:
|Allows you to modify the options you entered in the Add New Target dialog.|
|Tests the connection to the target. The status column will display the results.|
|Deletes the target from FusionReactor.|
Seeing your data in FusionAnalytics
When you connect FusionReactor to FusionAnalytics, the data will not instantly appear. You will need wait for a transfer to occur.
If you want to force a transfer, in FusionReactor go to Analytics → Connector Status. Click Transfer to send the logs to FusionAnalytics. You will need to wait for each log to be processed by FADC before the data will be available in the Client.
Naming Applications and DataCollectors¶
When adding a FusionAnalytics target to FusionReactor's Analytics > FusionAnalytics Targets, you'll be asked for some information about the FusionAnalytics DataCollector target.
An in-depth explanation of FusionAnalytics targets, and the FusionAnalytics DataCollector which defines them, is out of the scope of this article; however it is possible to show you where the data for each field in the wizard comes from.
Host and Port¶
This is the hostname and port number of the FusionAnalytics DataCollector service. If you used the default port, the target will be available on 8400. The host must be either an IP address or a DNS-resolvable name, which points to a reachable interface, located on the target system.
This refers to the name of the individual data collector app running within FusionAnalytics.
You can find it by logging into the FusionAnalytics DataCollector Administrator, and looking at the Application Summary page. The application name is located in the Current Applications table, in the column Name. In the figure below, the application apple has been defined.
Each Application running in FusionAnalytics can define one or more DataCollectors. These should be used to cluster together logically-related FusionReactor datasources. For example, all FusionReactor instances on a single machine could feed the same Application, but must feed different DataCollectors.
You must not point two different FusionReactor instances at the same DataCollector, as this will make the analytics invalid since there is no differentiating field for the different sources.
By using a different DataCollector, FusionAnalytics is able to correctly partition the data by including a uniquely-named field (the name of the DataCollector).
If you accepted the default DataCollector name, or you haven't customized your FusionAnalytics DataCollector Application yet, the default DataCollector name will have been set up for you. This is faDataCollector-1.
If you're unsure of which names are already in use, you can see which DataCollectors are installed as follows:
Log in to the FusionAnalytics DataCollector Administrator and, on the Application Summary page click the Application Detail button for your application. On the page which appears, click Show/Hide DataCollectors to reveal the names of the data collector in use. In the figure below, we've defined four: lin1, mac, solaris and win1. .
Pick the one you want to send logfiles to, and enter it's name into the DataCollector box in the FusionReactor Connector Manager Wizard.
Username and Password¶
These are the credentials used to log in automatically to the FusionAnalytics DataCollector. The default username is admin and the password is that used to log in to the FusionAnalytics DataCollector Administrator.
Check this box if your FusionAnalytics installation is configured to use HTTPS transports.
When the settings are complete a test is run automatically, you can also click Test Target to attempt a test connection. If all is well, OK appears. If an error occurs, Error appears. Clicking on the Error opens a window in which diagnostic information can be used to locate the error.
This part of the page is used to control how the FusionAnalytics Connector will handle log files for transfer to FRAM or FusionAnalytics DataCollector. The options in the drop down box are listed below in the note.
|FusionAnalytics Connect Mode||Controls how the FusionAnalytics Connector will handle log files for transfer to FRAM or FusionAnalytics DataCollector.
FusionAnalytics Connect Modes:
Sent to targets by this instance - Log files are compressed and sent locally. This mode does not require FRAM to be available.
Sent to targets by FRAM - Log files are compressed by FRAM. If FRAM is not available, falls back to Sent to targets by this instance. This mode optimizes transfer of archives by having FRAM compress and send files to FusionAnalytics.
Sent by this instance (uncompressed) - No compression is performed. Each log file is sent to FRAM or FusionAnalytics directly and uncompressed. After rotation, if this mode is active, the original folder is not deleted. No archives are created.
Disabled - The FusionAnalytics Connector is disabled. No log files are sent to FusionAnalytics.
After you have selected your option click Save FusionAnalytics Settings.
Since FRAM itself has no FusionAnalytics targets, the targets to which logs are sent are always those of the local instance, regardless of which Connector mode is active.
|Enable Proxy FusionAnalytics||Use this setting to Enable/Disable the use of the proxy server with the FusionAnalytics connections.||Do not use proxy|
This tab of the settings page allows you to change the FRAM target settings.
|Host||This is the ip or hostname of the server that is running the FRAM instance.
Currently only local FRAM targets are allowed.
|Port||This is the port on which FRAM is running, this is set during the FusionReactor setup process, and defaults to 8087||8087|
|Path||This is the web root of the FusionReactor instance on the given host and port. This is the same as the Web Root setting under FusionReactor Settings on the FRAM target.||/fusionreactor|
|User Name||This is the rest user name of the FRAM instance. This can be seem in the FusionReactor Settings page under FR-REST API Access on the FRAM target.||restuser|
|Password||This is the rest password for the FRAM instance. This can be set in the FusionReactor Settings page under Change FR-REST Password on the FRAM target.|
|HTTPS||This will indicate to FusionReactor whether or not to use HTTPS to send the request to FRAM, see the following: Setting Up FRAM with HTTPS||Disabled|
|Test FRAM||This will perform a test connection to the FRAM target to test that the FRAM target can be contacted, ensure that any changes are saved before testing the target.||Test Fram|
The FusionReactor Administration Manager (FRAM) installed as part of FusionReactor 6.0.x acts as a vital part in the transfer of logs from individual FusionReactor instances.
FRAM provides a completely isolated instance of FusionReactor which is just for monitoring and instance management, it will continue running even following a JVM crash.
FRAM will also perform process and transfer offloading, making the log transfer to FusionAnalytics a central process performed by a single instance.
Log Transfer & Archiving¶
By selecting the FusionAnalytics Connector mode Sent to targets by FRAM (See FusionAnalytics Settings for more information about log modes) the individual FusionReactor instance log archive transfers to FusionAnalytics will be performed by FRAM. This mode optimizes transfer of archives by having FRAM compress and send files to the FusionAnalytics Targets, offloading the processing power required to compress and transfer of log files onto FRAM. This means that each individual FusionReactor instance will not have to perform this costly action, making FusionReactor more transparent on the target server.
The FusionReactor Administration Manager (FRAM) will also act as a central hub for log transfer where multiple instances can send their log files for archiving and transfer via FRAM. By doing this a central archive will be maintained by FRAM containing the log files of all instances using the FusionAnalytics Connector mode Sent to targets by FRAM (assuming their FRAM and FusionAnalytics targets are correctly set up.
FRAM will also create a local archive of the log files generated by the
instance and save this in its archive folder available at
\instance\FRAM\archive. By doing this FRAM generates a central storage
of files. FRAM will handle the log transfer and archiving of log files
generated by individual FusionReactor instances. Each FusionReactor
Instance specifies the FusionAnalytics DataCollector
Targets it wishes
to send log files to, FRAM will process the log files and send the
compressed log files to the specified FusionAnalytics targets.
More information about each archive can be provided by clicking the Details button next to each listed archive. This will provide information on the life cycle of the archive, all the way from the instance that generated the logs to FusionAnalytics. By having FRAM performing this action the transfer to FusionAnalytics can be easily tracked and maintained by the user of the FusionReactor Administration Manager.
For more information about FRAM please see the online documentation for the FusionReactor Administration Manager (FRAM).