Chat now with support
Chat with Support

Foglight 5.9.1 - Release Notes

Resolved issues and enhancements

Resolved issues and enhancements

The following is a list of issues addressed and enhancements implemented in this release. For a list of resolved issues for the Foglight Agent Manager and the Infrastructure cartridge at the time of the Foglight 5.9.1 release, see the Foglight Agent Manager 5.9.1 Release Notes and Foglight for Infrastructure 5.9.1 Release Notes.

Management Server

Defect ID

Resolved Issue

WCF-15912 When clicking Administration under Homes, the Management Server did not return to the parent page.
FGL-19415 Fixed the CVE-2016-2183Birthday vulnerability.
FGL-19384 After upgrading to the Management Server, performance slowness and FglAM upgrade failures occurred.
FGL-19337 Fixed issue where the values of baseline_MAX and threshold were inconsistent.
FGL-19247 Fixed the LDAP performance issue existing in environments with a large number of users and user groups.
FGL-19236 Added auditing to FSMService add/remove/modify operations.
FGL-19003 It was slow to calculate or render Alarm Count details during the generation of FMS bundle.
FGL-18890 AlarmAggregateService failed to receive alarm notifications from local Alarm Service on Federation master.
FGL-18554 When aggregating the history of a non-numeric observation into timeslices, the timeslice values may not be correct if the raw value is at a more coarse granularity than the timeslices.
FGL-17267 During the FMS installation using Windows Authentication to an external SQL Server instance on a separate SQLHOST, an error was thrown when the LAN Manager Authentication level security policy is set to "Send NTLMv2 response only. Refuse LM & NTLM".


Known issues

Known issues

The following is a list of issues known to exist at the time of this release.

Note: Check the release notes for each cartridge you will be using for known and late-breaking issues with that cartridge. 

For a list of known issues for Foglight Agent Manager in this release, see the Foglight Agent Manager 5.9.1 Release Notes.

Workflow automation not supported in Federation or High Availability modes

Foglight for Automation is not supported:

  • On a Foglight Federation Master.
  • When Foglight is configured to use the High Availability features.


See below for details. These limitations will be addressed in a future release.

Workflow automation not supported in High Availability mode

When a Foglight Management Server running in High Availability mode fails over, all in-progress tasks fail and the system cannot recover them.

Workflow automation not supported on Foglight Federation Masters

The Foglight Cartridge for Automation should not be installed, and workflow automation must not be used, on a Foglight Federation Master. This is because active tasks that update topology objects are not permitted on the Federation Master. Workflow automation may be installed and used on the Federated Child(ren) as usual.

Management Server

Defect ID

Known Issue


If you toggled enable/disable state for Alarm Integration, Incident Integration, Clear Integration Alarm, or QMX Agent Host Connection Rule rule in the Integration cartridge, its enable/disable state will be reset after upgrading to 5.9.1. 


When upgrading the Management Server from to 5.9.1, unexpected CartridgeDeploymentException shows in the log file.

Workaround: This exception only occurs in environments with those cartridges which define Java Archive or JBoss Artifacts component, like Integration and IntelliProfile cartridges. JBoss Server has been removed from the Management Server 5.9.1. As a result, JBossDeploymentHandler, which was used to handle above components, is also removed from the Management Server 5.9.1. If you encounter this issue, you could ignore it as this exception does not impact upgrade of the Management Server and relevant cartridges. 

FGL-18953 After first installing this version of the Foglight Management Server, EU appliances may fail to connect with the Foglight Management Server.
Workaround: Restart the Foglight Management Server after the installation. 
FGL-18611 LDAP accounts may fail to bind to the OU group whose name includes backslash characters '\'. This can prevent Foglight to authenticate LDAP users.
FGL-18470 FSMService availability calculations do not work as expected in a service hierarchy.

FoglightShutdownAction.checkInUsedPort() does not load port settings from server.config during an upgrade.


In a Windows Single-Sign ON (SSO) with untrusted multiple domains, only the user with the keytab file-corresponding group can be imported.


The Management Server data-driven multiple severity rules may fire multiple times for the same severity in case of concurrent data submissions.


The rule engine may pass non-persisted alarm IDs to rule actions. This can prevent some alarms, that are generated multiple times against the same scoping object, from firing.

FGL-18034 PostgreSQL fails to start up while installing the Management Server with an embedded PostgreSQL database on a Korean OS.

Failing to start the Management Server with the --host=*.*.*.* option due to a failed attempt to start the embedded database.


LDAP users may have to log in twice to get proper role assignments.


Foglight is configured with a list of default ciphers. Some of these ciphers are potentially vulnerable ciphers. A review of the ciphers that will be available in a future release is ongoing.

Workaround: Edit the list of ciphers by completing the following steps:

  1. Open the following file for editing:
    • Releases prior to 5.7.1: <FMS_HOME>/server/deploy/jboss-web.deployer/server_https.xml
    • 5.7.1 release and later: <FMS_HOME>/server/tomcat/server.xml
  2. In this file, specify a list of ciphers in the ciphers attribute of the secure connector:

    <Connector executor="tomcatThreadPool" maxHttpHeaderSize="8192" URIEncoding="UTF-8" scheme="https" secure="true" SSLEnabled="true" clientAuth="false" keystoreFile="../../config/tomcat.keystore" keystorePass="nitrogen" sslProtocol="TLS" sslEnabledProtocols="TLSv1.2,TLSv1.1,TLSv1" bindOnInit="false" ciphers="COMMA_SEPARATED_LIST_OF_CIPHERS"/>

For complete information, visit


The Management Server distribution layout contains some confusing directories under server/jboss.

FGL-17415 ObservationQueryPlan$QueryStep.getDayIndex(...) can result in ArrayIndexOutOfBoundsExceptions during a Daylight Savings Time (DST) change.

Core monitoring policy derivations are not deployed on a Federated Master.


Excluded objects are sometimes included in a Federation topology synchronization.


A Foglight Agent Manager that is no longer in use maintains a stale disconnected RemoteObject instance under the host object. This causes the host to appear as undefined on the Hosts dashboard, even though the host is monitored by Foglight for Infrastructure.


When federation.config does not contain any queries, TopologySynchService mistakenly logs the following warning:

WARN Obsolete federated topology object purge will not be performed because the following federated servers are not connected: ...


Topology loading can fail for topology objects that have been converted to DataObjects.


Type-based exclusions in federation.config do not apply to referenced objects.


String observations containing reserved characters in script agents are not supported.


The alarm email notification code does not persist any information about notification alarms that were previously sent. If the server is restarted, all alarms appear as new, and the notifications are resent.


Hibernate/JDBC errors occurring while loading rules return an empty list.


Cleared credential alarms are found when querying for alarms that are not cleared.


Simple rules that report alarms directly are not affected by blackouts.

FXM-880 Diagnose link doesn't work on the Master FMS

Browser interface

Defect ID

Known Issue


An attempt to import external groups results in the following error:

Could not execute the requested operation, error(s) had occurred: Failed to retrieve groups, please make sure you correctly configured the Directory Services, check log for details.

The log, however, does not contain any errors.


After changing the value of the registry value, some of the performance calendars may disappear from the list.
Workaround: To correct the display and allow performance calendars to be deleted, return to the main Manage Registry Variables page, then navigate back to the appropriate registry value.


When editing properties from the Agent Status dashboard or the Agent Properties dashboard in Chrome 17, some edit fields are not available.
Workaround: Use a different browser when editing agent properties through these dashboards.


The Cartridge Inventory page is currently unable to alert users if a cartridge install fails due to lack of space in the server's temp directory. An error is written to the log files.


When editing secondary agent property tables from either the Agent Status or Agent Properties dashboards, performing Add and Delete operations at the same time results in all Added items being deleted.
Workaround: Add items to the table, save them, and close the popup window, then edit the secondary property again and delete the rows that need to be deleted. Finally, save and close the popup window. This ensures that the added items remain.


Selecting the time range "All" on the Service Details dashboard can result in the page getting stuck refreshing. Avoid using the "All" time range on this page.


Selections made on the Service Selector tab of the Select services and tiers to monitor dialog box of the SOC become unselected when you switch to the Tier Selector tab and back again.


The first time the Environment Report is run it may generate a warning: " - Text ... did not fit. in view ..."
This is a benign warning that is caused by one of the values read from the environment being too large for a page.


Credential alarm messages sometimes contain UUIDs for objects related to the alarm.
Workaround: Show the Alarm Detail dialog for these alarms to resolve the objects to identifiable names.


When using a Federation Master that is a newer version than the Federated Child, it is possible to get warnings about missing metrics. This occurs when new metrics have been added to the Federation Master's version. The warnings indicate that the Federated Child does not provide these metrics, and they will cease when the Federated Child is upgraded to the same version as the Federation Master.


When editing secondary agent property lists, the Delete Selected Row(s) button marks the rows to be deleted by graying them out. Save changes must be selected to delete the selected rows. If a required field is blank, then a value must be entered before the row can be marked for deletion.


In the Service Builder and Service Operation Console (SOC) dashboards, when alarm filters are configured, they are applied to the main dashboards (Service Builder and SOC dashboards) and to some first level popups only. The alarm filters are not applied throughout all popups/drilldowns at any level.


When reviewing Alarms on a Federation Master, you are able to open a link to view the rule that fired the alarm on the Federated Child. This involves opening the console of the Federated Child. If the Federated Child is of an older version than the Federation Master, an error page is displayed.


Under rare conditions, if the session is allowed to expire while viewing some of the views in the Administration dashboard, the process of logging in again displays the Administration view without any decoration around the side panels.

Workaround: To clear this condition, close the browser. Do not save your session if the browser prompts you to save the last session configuration. Instead, start the browser and log in again.


The column layout on the Agent Status Page does not allow resizing of columns, which can result in the host name appearing truncated.


An error message stating "Error getting value of context entry with key:timeRange" can appear in the logs when using the popup menu on the Origin field of the Alarm Details dialog.


If a Federation Master Server connects to an older version Federated Child, the Federated Child may have some of the objects that are no longer associated with the Management Server Performance Report and Foglight Server Diagnostic views on the Federation Master. This results in some data about the Federated Child being unavailable to this report and views.
Workaround: Go to the Federated Child to view detailed information.


Inconsistency in the number of alerts shown in the "All Alarms Chart" and "Alarm List with Filter" in the Alarms dashboard.


Display all available scoping objects in the condition editor for copied and user-created rules.


On systems with a large numbers of hosts, the Host Table view can be slow to display.
Workaround: Use the hide column feature to hide the sparkline (small graph) columns. This significantly improves the display performance.


Foglight 4 Tabular views generate an error when the time range has "Auto" granularity.


The edit rule definition for an alarm from a Federated Child brings up a page for editing a rule on the Federation Master, rather than for the rule that triggered the alarm. This occurs because it is not possible to perform administration from the Federation Master.


When the user pastes data using right-click > Paste to a box that automatically updates itself (for example, the edit agent properties screen), the user interface does not automatically refresh itself.
Workaround: After pasting into an Agent Properties text field, press the Spacebar > Backspace, and click the Save button.


When agent activation fails, the "Activating" status box does not show the status of the failed start agent.


The Service Levels Details page does not show cleared alarms for Service Level Policy.


Seconds should be displayed in the startTime and endTime in the WCF under Browse > Data when granularity is set to "raw".


QP5 needs to be upgraded to the latest version.


In Internet Explorer 8, when the number of dashboard nodes in the navigation panel exceeds the maximum that can be displayed without a scrollbar, the scrollbar does not immediately appear.


Tree Table list nodes should not be clickable or dwellable.


PDF charts and files that were requested through a proxy fail and result in a Connection Refused error. Note that this error only applies to requests made by the Management Server.


Vertical scrollbar not shown for the text area.


Third party known issues

Third party known issues

The following is a list of third party issues known to exist at the time of this release. For a list of third party issues related to the Foglight Agent Manager at the time of the Foglight 5.9.1 release, see the Foglight Agent Manager 5.9.1 Release Notes.

Defect ID

Known Issue


The Foglight Management Server generates an "Unable to load Java VM shared library to start JVM" message when attempting to start the Management Server on a Red Hat Enterprise Linux Server 5.0 VMware machine.

This is caused by the SELinux security code on newer RedHat systems. There are two workarounds:

  • Enter the following command (as root):

    chcon -t textrel_shlib_t <foglight_home>/jre/lib/i386/server/

  • Disable SELinux on the system in question.

This issue is addressed in the Sun JDK 1.6.0 release notes:
Bug: 6538311
This bug affects both Foglight Management Server executables (fms and fmsha).


Installation: Console Mode. Returning to a previous step in the installation process after being prompted for a user password, the input is not recognized.
During a console mode install, it is possible to return to the previous step by typing "back" and then pressing the Enter key. If done while the installer is asking the user to enter a password, all subsequent user inputs are automatically changed to asterisk symbols (* ) and are unreadable by the console.
Workaround: Type quit and reinstall again.


Installation: Console Mode. Some characters of the Administrator's password are not hidden if typed too quickly.
This issue has been observed when installing Foglight on some older machines.
Workaround: Avoid typing the password fast.


When typing words containing "ss" in the online help search, a null pointer exception is generated.
This is a known issue for applications that use Oracle's default implementation of the Java Help Search functionality.


Upgrade and compatibility

Upgrade and compatibility


  • To upgrade from a 5.5.x Management Server to 5.7.1 or later, you must first upgrade to 5.6.2 (or a more recent 5.6.x version), and then to 5.7.1 or later.
  • To upgrade from a 5.6.11 Management Server to 5.7.5.x, you must first upgrade to 5.7.1, and then to 5.7.5.x.
  • To upgrade from a 5.7.5.x Management Server to 5.9.1, you must first upgrade to, and then to 5.9.1.
  •  Foglight Management Server (5.7.0 to no longer supports authentication of Federation actions (such as clearing alarms) without specifying a secure token in the federation.config file. Previously configured installations in which Federated Master and Children were authenticated without a security token no longer allow authenticated actions. For more information, see the Federation Field Guide.
  • In this 5.9.1 release, configuration of SecurityToken in the federation.config file (adopted from Foglight 5.7.0 to has been removed. Therefore, configuring an SSL for Federation is required when you upgrade the Federation environment to release 5.9.1, if your Federation environment configures SecurityToken before. For more information, see the Federation Field Guide.
  • The following cartridges are not compatible with this release. If you have these products installed, you must NOT upgrade to this version of the Management Server.
    • Foglight for Transaction Recorder and Player
    • Foglight APM for End User Experience
    • Foglight for SQL Azure

See the Foglight 5.9.1 Upgrade Guide for information about upgrading Foglight and for a list of the latest cartridge versions that are compatible with Foglight Management Server 5.9.1.


Related Documents

The document was helpful.

Select Rating

I easily found the information I needed.

Select Rating