- A cluster configuration can define global resources (declared outside of cluster service groups) and in-line resources (declared inside a service group). The names of resources must be unique regardless of whether the resource is a global or an in-line resource. Previously, luci allowed a global resource with a name, which was already used by a resource that had been declared in-line, and a name of a resource within a service group with a name that was already used by a global resource. As a result, luci could terminate unexpectedly with error 500 or the cluster configuration could be modified improperly. With this update, luci fails gracefully under these circumstances and reports the problems, and the cluster configuration remains unmodified.
- If a cluster configuration was not valid, luci terminated unexpectedly without any report about the cluster configuration problem. It was therefore impossible to use luci for administration of clusters with invalid configurations. With this update, luci detects invalid configurations and returns warnings with information about possible mistakes along with proposed fixes.
- The user could not debug problems on-the-fly if debugging was not enabled prior to starting luci. This update adds new controls to allow the user to change the log level of messages generated by luci according to the message type while luci is running.
- When the user created a cluster resource with a name that contained a period symbol (
.), luci failed to redirect the browser to the resource that was just created. As a result, error 500 was displayed, even though the resource was created correctly. This update corrects the code that handles redirection of the browser after creating such a resource and luci redirects the browser to a screen that displays the resource as expected.
- Previously, luci did not require any confirmation on removal of cluster services. Consequently, the user could remove the services by accident or without properly considering the consequences. With this update, luci displays a confirmation dialog when the user requests removal of cluster services, which informs the user about the consequences, and forces them to confirm their action.
- Since Red Hat Enterprise Linux 6.3, authenticated sessions automatically expired after 15 minutes of inactivity. With this update, the user can now change the time-out period in the who.auth_tkt_timeout parameter in the
- Previously, the default value of the monitor_link attribute of the IP resource agent was displayed incorrectly: when not specified explicitly, its value was displayed as enabled while it was actually disabled, and vice versa. When the user made changes to the monitor_link value using luci, an incorrect value was stored. With this update, the monitor_link value is display properly, and the user can now view and modify the value as expected.
force_unmountoption was not shown for file-system resources and the user could not change the configuration to enable or disable this option. A checkbox that displays its current state was added and the user can now view and change the force_unmount attribute of file-system resources.
- A new attribute,
tunneled, was added to the VM (Virtual Machine) resource agent script. This update adds a checkbox displaying the current value of the
tunneledattribute to the VM configuration screen so that the user can enable or disable the attribute.
- Previously, an ACL (Access Control List) system was added to allow delegation of permissions to other luci users. However, permissions could not be set for users until they had logged in at least once. With this update, ACLs can be added and changed before the user logs in to luci for the first time.
- Due to a regression, the Intel Modular and IF MIB fencing agents were removed from the list of devices for which users could configure new instances. Consequently, users could not create a new instance of these fencing devices. The Intel Modular and IF MIB fencing device entries have been added back to the list of fence devices and users are again able to create new instances of Intel Modular and IF MIB fencing devices.
- In the Create and edit service groups form, the relationships between groups could not always be easily discerned. Solid borders were added along the side of resources within the forms to make the relationships between resources clearer. Also, when adding a resource to a service group, the screen is scrolled to the resource that was added.
- While creating and editing fail-over domains, the user could select and unselect checkboxes and enter values into text fields whose values were ignored. Such checkboxes and text fields are now disabled and become enabled only when their values are used.
- To provide an interface for working with the RRP (Redundant Ring Protocol) configuration in luci, Technology Preview support was added for RRP in the corosync cluster engine that the Red Hat HA stack is built upon. The
Redundant Ringconfiguration tab is now available in the
Configuretab of clusters to allow RRP configuration from luci.
- A new resource agent was added to provide high-availability of condor-related system daemons. This update adds support for viewing, creating, and editing the configuration of Condor resources to allow the user to configure the Condor resource agent.
rebooticon was similar to the
refreshicon and the user could have mistakenly rebooted a cluster node instead of refreshing the status information. With this update, the
rebooticon has been changed. Also, a dialog box is now displayed before reboot so the user must confirm their reboot request.