Limitations

Installation and configuration

• Limited information about version of some components

  The component version of some components that is not installed through RPM package version is displayed as n/a.

  There are multiple methods to install a software. The one method is through the RPM package installation and another is by directly downloading the executables to the disk. For RPM based installation BigFix Inventory can detect the component version and signatures are available in BigFix Inventory. If an application is installed by directly copying the executables or downloading the files to the disk, the component version is not detected. Only only installed software is detected in this case. On Linux and Unix Operating System, executables do not include metadata to read the header for obtaining the version details. To avoid potential security breach, BigFix Inventory does not discover applications by running application files or executables.

• Support for Distributed Server Architecture (DSA) is limited to one backup server

  Distributed Server Architecture (DSA) allows for setting up multiple BigFix servers that can replicate data between each other. BigFix Inventory does not support failover but BigFix Inventory can work with BigFix Inventory configured in DSA mode. It is recommended to configure BigFix Inventory server with primary BigFix Inventory server. If the configured BigFix Platform server fails, BigFix Inventory server imports continues to fail. It is recommended to restore BigFix Inventory and discontinue using BigFix Inventory as a primary server. If you need to switch to BigFix Inventory server as primary server, you need to reconfigure existing data source with BigFix Inventory and then resynchronize datasources through resynchronize_datasources_once parameter in Advanced Server Settings."

• Microsoft™ SQL Server used as the BigFix database cannot be case sensitive

  Only case insensitive Microsoft™ SQL Server database is supported.

• Changing the host name of the BigFix server is not supported

  The host name of the BigFix server cannot be changed because it is recorded in the license certificate during the installation. To change the host name of BigFix, reinstall the application.

• Creating clones of virtual machines that are running is not supported

  Cloning virtual machines that are running might produce unexpected results in BigFix Inventory.

• Language of the installer depends on the system locale

  To run the BigFix Inventory installer in a language other than English, system locale must be set to that language.

• Permissions that are defined for the Administrator role cannot be changed

  The Administrator role is assigned all available permissions. It prevents from creating separate Administrator roles, for example one responsible for data, the other for users.

Using BigFix Inventory

• The scanner does not scan paths that contain certain whitespace characters and other special characters
  The scanner does not scan file paths that contain the following whitespace characters:

  • Carriage return (CR, character code 13)
  • Line feed (LF, character code 10)
  • Whitespace characters whose decimal codes are in the following ranges: 0- 8, 11-12, 14-31

  It also does not scan shared disks whose mount point paths contain an asterisk (*) or a question mark (?). To avoid a situation in which software is not discovered, ensure that files paths in your infrastructure do not contain the unsupported characters.
• Installation date and installation path not collected by the Initiate Software Scan/Disconnected Scanner

  The scanner does not scan the installation date and installation path for Debian 9, Ubuntu 18, and their earlier versions.

• Software installed in the /Libraries directory on Mac OS X is not discovered

  Only software that is installed in the /Applications directory is discovered.

• Information about Mac OS X computers is not presented on the IBM Capacity Data Completeness widget

  Computers running on Mac OS X are not included in the computer count on the IBM Capacity Data Completeness widget. It happens because the types of scans that produce results displayed on this widget are not run on Mac OS X. However, these computers are included in the computer count on the Scan Health widget. It causes that the overall number of computers displayed on each widget is different.

• Calculating utilization of the Managed PVU metric requires an additional configuration of BigFix Inventory

  Managed PVU metric is based on the number of processor cores available on computers that are managed by an application, for example by BigFix Spectrum Protect for Virtual Environments. Calculating utilization of the Managed PVU metric is not available out of the box. To calculate its utilization, place an appropriate isotag file on the computers that are managed by the particular application.

• Creating contracts and catalog customizations for IBM products is limited
  IBM provides catalog content that allows for discovering practically all IBM products. Due to internal reasons, entries for some products are duplicated in this catalog. Thus, it is not possible to modify the correct entry and creating contracts for IBM products might not work correctly. Moreover, BigFix Inventory supports contracts only for installation-based licenses which are not applicable to the majority of IBM products. Creating contracts for IBM products and customizing BigFix products in the software catalog is discouraged.

  Note: This limitation is applicable to the old contracts that were created before BigFix Inventory version 10.0.5.

  The new contracts are shown only for first 20 computer groups. The contracts for any computer group that is beyond 20 is excluded from the report. Computer groups are selected as per the order they were created (computer ID). It excludes the computer groups that have metrics disabled. However, you can select the computer groups using the 'new_contracts_max_computer_groups' and 'new_contracts_selected_computer_groups' hidden parameters from the Advanced Server settings.

• Signatures with the same GUID and different rules might be overwritten

  When you import a signature that has the same GUID as a signature that already exists in BigFix Inventory and the new signature has a different number of rules than the existing one, the signatures are not merged. Instead, the new signature overwrites the existing one. The problem does not occur when the imported signature has a different or no GUID.

• Software is detected twice if a custom catalog entry is the same as an entry in the BigFix catalog or an entry generated from an isotag file

  If you create a custom catalog entry which has the same publisher name, software name, and version as an entry in the BigFix catalog or an entry generated from an isotag file, both the custom and original entry is visible as detected.

• slmtag files with overlapping entries are not supported

  Start time of one entry in an slmtag file cannot overlap with the previously reported period. It should start at the end time of the previous entry or later.

• Adding excluded directories fails on Solaris when the number of characters in the task exceeds 1000

  When you use the Add Excluded Directories task and the specified list of directories is longer than 1000 characters, the task fails on Solaris.

• Specific start time cannot be set for actions by using REST API

  BigFix REST API requires that the start time for an action is specified as an offset from the time on the BigFix server. The solution is prone to network latency problems. It might cause that the action starts at time different from the intended one.

Reports

• Custom rule to suppress the discovery of software components of publishers from SWID tags

  With improvements for ISO SWID tags processing in versions 10.0.2 and 10.0.4, BigFix Inventory is capable to read information from most ISO SWID tags released by vendors. This includes the ability to recover from issues within the tag itself.

  Reporting of software through ISO SWID tags is not normalized and the reports are shown as usual. Currently, the following ISO SWID tags were identified that results in duplicate discovery of existing software catalog entries with below names:

  • with the published defined as: Adobe Systems Incorporated
  • with the published defined as: Microsoft Corporation
  • with the component name of: Red Hat Enterprise Linux

  Starting from version 10.0.4, predefined custom rules have been created to suppress the duplicate discovery of software which have been already discovered once. With version 10.0.9, the issue of not marking already discovered software have been addressed. In future releases, there might be added capability to reject ISO SWID tags overlapping with catalog content.

• Reports show the current PVU per core value instead of the value that occurred during the license peak time

  All Metrics reports including IBM Subcapacity reports and the audit snapshot show the current PVU per core value instead of the value that occurred during the license peak time. It does not affect the calculation of PVU subcapacity and full capacity. To display the value that occurred during the license peak on the reports, change the start date of the report to a more recent date. Try changing the date until you find the date when the PVU per core value changed. It allows you to have consistent information displayed on the reports.

• Data on the Software Installations report is inconsistent with data on other reports

  Information on the Software Installations report regarding software installed on shared disks or components installed multiple times on the same computer is inconsistent with information on other reports. The inconsistency occurs also when a column on a report links to a filtered view of the Software Installations report such as in case of the Installed Software column on the Computers report. It happens because the Software Installations report is based on the old data model that does not take shared disks and multiple component instances into account.

  Starting from application update 9.2.15, the old Software Installations report is deprecated and substituted with a new report view. It is based on a new data model and consistent with other reports. The limitation no longer applies.

• Scheduling multiple report e-mails requires ensuring an appropriate interval between each report e-mail

  When you schedule multiple reports to be sent to your email, you must ensure that they are not being sent at the same time. The reason is that each report is managed in a different session that starts when the report is being sent, and ends when the action is completed. Each closed session, however, terminates also other sessions, in this case the delivery of other reports. Therefore, the schedule must be different for each report. The sufficient interval is about 5-10 minutes. You can specify it while setting the start time.

• Audit Trail limitations

  Audit Trail report cannot be filtered by the Details column and does not contain information about uploads and imports of part numbers. What is more, audit trail does not contain information about software classification when either component, product or both are non-IBM.

• Non-IBM software discovered both on a local system and a shared disk is shown only as installed locally

  When non-IBM software is installed on a local system and on a shared disk, it is shown only as installed locally on the Software Classification panel.

• Filtering computers based on their IP addresses does not work under certain conditions

  When you filter your computers based on their IP addresses and specify the relation as ends with, no computers are displayed even if some computers match the criteria. To work around this issue, you can choose other relations, such as contains or begins with. This limitation concerns all reports.

• Filtering by installation path does not work under certain conditions

  When the installation path is specified as report filtering criteria, and contains -1, for example: C:\My_App -1, the filter returns inaccurate results. The filter returns not only the content of the specified directory, but also all non-IBM products that are installed on the computer.

• Filtering causes that the content of the window is scrolled up

  When you filter a report by using the in set filter, the content of the entire window is scrolled up to the top. You have to scroll down to add another filter.

• Incorrect number of rows can be displayed on reports that are saved as PDF files

  The number of rows that is displayed in the upper right corner of a PDF file can differ from the number of rows that is displayed on the user interface. This problem was only encountered with Mounted Shared Disks column in the Computers report because it contains multi-line entries.

• Time values are provided in different time zones on the user interface and in PDF or CSV reports

  On some reports, time values are expressed in the local time zone when the report is viewed on the user interface but in the UTC time when the report is exported to a PDF or CSV file. What is more, the format of the date and time is hardcoded in PDF reports.

• Signature definitions are not available on the Catalog Audit report for custom signatures that were deleted

  When you delete a custom signature from the software catalog, definition of this signature is not available on the Catalog Audit report. The definition section is empty.

• Package Data limitations

  The version format on Package Data report may differ for disconnected software scan results when compared to regular scan results. For disconnected scan, the dash (-) is replaced with a period (.).

  For example, the version shows as follows:

  • Disconnected scanner reports: 1.8.3.1.11.el7
  • Regular scanner reports: 11.8.3.1-11.el7
  Note: The limitation is applicable to Linux Disconnected Scanner version 10.0.12 and lower, and all versions of Unix disconnected scanner.

  On Solaris platform, inconsistencies in Package Data report version may happen because of collecting data from various package managers. Disconnected scanner does not support IPS package manager.

  On Windows platform, InstallDate for Uninstall Registry may have various formats such as "20190801", "Jan 26, 2023", "2023/1/25", or "Tue Sep 27 16:12:01 CEST 2022". The resulting date will be in the local time zone due to technical limitations of VBS.

  The character encoding used for the install location is determined by the regional settings of the operating system. If the character encoding of the registry entry does not match the regional settings, the path may contain question marks. However, MSI packages fully support character encoding in their location paths.

• Non-English characters are incorrectly displayed on the Package Data report for IBM i systems

  If the name of a package that is discovered on an IBM i system contains non-English characters, the characters are incorrectly displayed on the Package Data report.

• Sorting by name does not work on the predefined reports.

Security

• SP800-131a compliance cannot be enabled with Microsoft™ SQL Server

  If the SP800-131a cryptographic standard is enabled in strict mode in BigFix Inventory and the enhanced security mode is configured on the BigFix and MS SQL servers, connection between the BigFix Inventory server and the SQL Server cannot be established.

• The pvk format of the encrypted private key is not supported

  You can use an encrypted private key in the openSSL or pkcs8 formats.

• Key pair generated for BigFix Inventory can be used for Web Reports only if the private key is not password-protected

Other

• Custom page URL does not work on Microsoft Edge

  The Custom Rule URL, accessed throughout Management > Custom Rules > Rule Details page may not work on Microsoft Edge because of the browser's limitation in handling the length of URL.

• The To Do List displays the names of outdated scanning actions always in English.
• Catalog search does not work during the import of data - This is applicable to versions lower than 10.0.4.

• Fixlet for distributing scanner catalogs to the endpoints is not translated

  The Catalog Download (Version: version) fixlet is a custom fixlet that is not delivered with the BigFix Inventory fixlet site. Thus, it is not translated.

• Process IDs are not present in audit snapshots

• Eastern Arabic numerals are not supported