Release Notes for RHQ 1.1
This release introduces the high availability and failover feature set. It is available as a binary download and with upgrade capability from both 1.0.0 and 1.0.1 versions.
Download
Release notes with links
New Features
- [RHQ-3] - Have agent support Java 6
- [RHQ-57] - Display package audit trail in the UI
- [RHQ-176] - deployed content request history should list name of package being deployed
- [RHQ-198] - Finish IIS Plugin
- [RHQ-351] - Enable content discoveries to in the Embedded Console
- [RHQ-512] - add Vendor, Model, MHz, and CacheSize traits to CPU resource type
- [RHQ-595] - Pinging the agent should not end in an ERROR message
- [RHQ-645] - Send full alert information in SNMP trap
- [RHQ-650] - Timeline event grouping
- [RHQ-678] - HA adminstrative dashboard
- [RHQ-829] - JVM Remoting auto-discovery
- [RHQ-872] - Provide a Resourcehub popup shortcut
- [RHQ-880] - add new agent operation: download-server-failover-list
Sub-Features
- [RHQ-348] - On "Servers" tab of "Browse Resources", many items can have the same name. No way to tell them apart
- [RHQ-401] - DynaGroups Expression Builder
- [RHQ-435] - Group plugin config edit allows changes without override checked
- [RHQ-494] - Include resource's ancestry as a column in the "add to group" page
- [RHQ-588] - he name of dynagroup should not be editable from the link "Group Definition's Managed Resource Group "DynaGroup -"
- [RHQ-671] - agent failover mechanism
- [RHQ-672] - server-side "maintenance mode"
- [RHQ-677] - HA installer
- [RHQ-682] - add support for substring matching on bracketed expressions
- [RHQ-686] - support listing available property names for bracketed expressions
- [RHQ-756] - Dynagroup calculated name should change when group name is changed.
- [RHQ-770] - list servers
- [RHQ-771] - list agents
- [RHQ-772] - view server details
- [RHQ-773] - view agent details
- [RHQ-774] - edit server detalis
- [RHQ-775] - view affinity group details
- [RHQ-776] - list affinity groups
- [RHQ-777] - edit affinity group
- [RHQ-790] - HA event timeline view
- [RHQ-791] - HA event details
- [RHQ-794] - include reference to parent resource for compatible groups on monitor > visibility > resources sub-sub-tab
- [RHQ-795] - include reference to parent resource on metric details page
- [RHQ-796] - include reference to parent resource on resource group inventory page
- [RHQ-852] - HA event details - improvements
Improvements
- [RHQ-425] - group plugin config update history should be displayed by default w/ newest entries first
- [RHQ-527] - PageList missing serialVersionUID
- [RHQ-542] - have Server support running on Java (JRE or JDK) 6
- [RHQ-573] - refactor how and when alert definitions are created from their corresponding alert templates
- [RHQ-582] - WARN message implies Implementation-Version should be required in manifest
- [RHQ-633] - Out-of-the-box RHQ Server log monitoring
- [RHQ-666] - Change create package-backed resource response APIs to provide resource name
- [RHQ-669] - Add package metadata to indicate if it supports multiple architectures
- [RHQ-683] - Don't ask for package name when creating a new package-backed resource; use file name from file upload field
- [RHQ-691] - Support compiling RHQ with JDK1.6
- [RHQ-720] - improve cascade handling for alert templates
- [RHQ-742] - reduce the time it takes to load/reload the alert cache
- [RHQ-788] - Add parent name to resource names when displaying resources in a list
- [RHQ-792] - asynchronous updating of metric tempates / schedules
- [RHQ-866] - Need filter mechanism on Partition Events page.
- [RHQ-939] - clean up the server's log4j xml file - make more categories
Bugs
- [RHQ-30] - Full Alert Condition Set is not displayed on the "alert detail" page
- [RHQ-67] - on the Alert>Definitions tab, the New Definition button should be disabled or hidden if the current user does not have the MANAGE_ALERTS permission
- [RHQ-119] - a group with 0 members shows an availability of UNKNOWN, but I think it should show an avail of UP
- [RHQ-129] - Event creation Exception on Agent
- [RHQ-167] - Resouce Timeline failure in Load Resource Events - unterminated literal
- [RHQ-194] - After page refreshes, ANY button that is conditionally enabled is reset to disabled, even thought the corresponding checkboxes remain selected
- [RHQ-205] - Jon Agent shows * [ version
- [RHQ-221] - Weird configuration caching on resource create
- [RHQ-222] - Resubmitting a failed configuration-backed resource create results in locked attributes
- [RHQ-253] - Unable to inventory a postgres instance owned by 'postgres' user
- [RHQ-262] - unable to manually add Apache server installed via yum
- [RHQ-279] - Dashboard page: icons need hovertext
- [RHQ-308] - "(* [.core.pc.inventory.InventoryManager)- Could not transmit availability report to server" error in agent log
- [RHQ-322] - Error after import of a JBossAS 4.0.5
- [RHQ-329] - Timeline view shows first 100 events, instead of the last (most recent) 100
- [RHQ-330] - Misleading cursor on events page
- [RHQ-333] - event list shows oldest events by default, not newest
- [RHQ-379] - Url has no more id after uninventoy
- [RHQ-416] - Native system fails to gather cpu metrics for dual core opterons
- [RHQ-427] - IIS plugin generates ugly warning in agent log when IIS not installed.
- [RHQ-436] - Creating a new configuration-backed resource needs cancel button in UI
- [RHQ-438] - Connection properties not synced from Server to Agent after clean Agent start
- [RHQ-447] - alert definition page does not show all details of the condition set
- [RHQ-463] - inventory merge after discovery does not remove "missing" resources
- [RHQ-466] - resource container goes null when deleting a platform from inventory
- [RHQ-476] - apache response time data off by an order of magnitude
- [RHQ-480] - Cancelling while adding a new resource doenst work if validation fails
- [RHQ-483] - agent-side inventory not sync'ed with the server
- [RHQ-515] - Metadata: Persistent List indexes update improperly
- [RHQ-557] - Metadata handling for templates seems to be missing
- [RHQ-569] - Off-by-one in Super User Role count
- [RHQ-605] - on Configuration>History tab, the "Rollback Selected" button should be titled "Rollback to Selected" or "Revert to Selected" (i.e. the word "to" needs to be added to clarify)
- [RHQ-613] - Recent Operations portlet shows the oldest operations executed, not the most recent
- [RHQ-614] - Manually added resources lost from agent inventory during upgrade
- [RHQ-622] - JMX plugin: JMXDiscoveryComponent does not support manual-add, and JMXServerComponent does not support JMX servers w/ type "JBoss"
- [RHQ-624] - Operations portlet: error in recent group ops
- [RHQ-627] - Apache plugin operations show success even when they fail due to lack of SNMP support
- [RHQ-629] - Adding user notification adds the user twice
- [RHQ-630] - Execute now on a scheduled operation causes a NPE
- [RHQ-631] - Resource intersection of group members not working in browse resources
- [RHQ-635] - Scheduled Group Operations not shown in RecentOperations portlet ...
- [RHQ-637] - * [RHQ-server.sh start/stop behavior doesn't match the printed output
- [RHQ-649] - ACL: Users w/o access to inventory mgmt shouldn't be allowed to view "Create New Group Definition"
- [RHQ-655] - Error when creating two content sources with the same name
- [RHQ-664] - Applying alert template to existing inventory blows up
- [RHQ-680] - Cancel button on create new package-backed resource tries to validate configuration values
- [RHQ-689] - "Greg's EJB3 Entity browser" title of page when you click an Availability icon
- [RHQ-69] - Authz: Monitor>Configure tab should not be displayed if current user does not have the MANAGE_MEASUREMENTS permission
- [RHQ-696] - removing resources from recursive groups removes too much
- [RHQ-706] - 'avail' agent prompt command does not respect "--verbose" option
- [RHQ-715] - AlertConditionCacheMonitor mbean cache element counts are incorrect
- [RHQ-717] - Parentheses in JAVA_HOME path prevents * [RHQ-server.bat from running
- [RHQ-722] - Agent wrapper script does not work as start script on RHEL 5
- [RHQ-727] - After starting Windows service, triggering manual discovery from server causes agent to throw NoClassDefFoundError: javax/xml/bind/Unmarshaller
- [RHQ-728] - Tracebacks when trying to add new content source feeds.
- [RHQ-730] - Add JMX Server does not fill connection properties from selected template
- [RHQ-733] - Upgrade from 2.0.0 to 2.1.0 fails with postgres
- [RHQ-734] - alert audit trial should not include action filter of its original alert definition
- [RHQ-736] - Config: config component incorrectly maintains the Configuration and ConfigurationDefinition attributes in its own state
- [RHQ-737] - Could not load ./* [RHQ-agent-env.sh on Solaris machine while running the script ./* [RHQ-agent.sh
- [RHQ-740] - ConcurrentModificationException is being seen in the agent log
- [RHQ-750] - recovery alerts not working if recovery happens too quickly
- [RHQ-751] - can't view non-measurement-based alert definitions or alert history after * [RHQ-742
- [RHQ-752] - content deployment creates duplicate entries in the "full package audit trail" table
- [RHQ-753] - PersistenceUtility warning in server log when viewing package upload confirmation page
- [RHQ-755] - Members count is off in Browse Resources page (for view-only user)
- [RHQ-757] - duplicate entries in recent alerts portlet for non-admin user
- [RHQ-758] - resource summary portlet counts are incorrect for non-admin user
- [RHQ-759] - compatible and mixed group totals at the bottom of res browser are incorrect for non-admin user
- [RHQ-760] - Apache live check never happens
- [RHQ-763] - hibernate exceptions from Agent.status field after dbupgrade to 1.1.0-SNAPSHOT
- [RHQ-781] - Installer program does not write server name into * [RHQ-server.properties (regression)
- [RHQ-782] - missing resources on "Read Only" option for alerts created with an alert template
- [RHQ-802] - server should add the server name in * [_server if missing
- [RHQ-806] - EJB Exception when trying to view the resources for a mixed group
- [RHQ-807] - Alert Definition Template UI should filter out invalid condition types for any given resource.
- [RHQ-808] - "Super User" (admin) is shown when adding users to an alert template
- [RHQ-809] - HA Affinity Group can be created with blank name.
- [RHQ-812] - Very long strings submitted for Affinity Group name fails and returns cryptic message
- [RHQ-816] - No submit button/nothing to do on listAgents.xhtml
- [RHQ-818] - UI Allows user to create multiple Affinity Groups with the same name.
- [RHQ-823] - Breadcrump too close to content
- [RHQ-826] - NPE in * [ log on startup, fails to create directory
- [RHQ-827] - Availability alerts not working on HA install
- [RHQ-828] - Detailed chart view shows all events
- [RHQ-830] - "Number of Concurrent Connections" metric no longer collected on apache servers
- [RHQ-831] - can't delete HA server
- [RHQ-834] - alert cache not reload after agent reconnection
- [RHQ-835] - agent needs to reconnect to the cloud properly
- [RHQ-837] - duplicate alerts generated
- [RHQ-846] - ContentContext should be passed to any resource component that is creatable w/ content-based creation
- [RHQ-849] - server-specific jobs only seem to be running on one server in the cloud, instead of on EACH server
- [RHQ-850] - agent doesn't update current server relation in table upon connect after failover
- [RHQ-851] - missing audit trail for some server OPERATION_MODE changes
- [RHQ-853] - webapp reloads cause error in the ejb3 timer service-based jobs
- [RHQ-855] - Clicking the simile "copyright" logo embeds the simile demo site within * [ itself.
- [RHQ-857] - Cancelling the creation of an affinity group results in stack trace.
- [RHQ-862] - mtime column constraint violation during server install
- [RHQ-865] - Unnecessary operation mode changing when system known to be down.
- [RHQ-869] - CheckForSuspectAgentsJob incorrectly marks resources RED when cloud servers are in different timezones
- [RHQ-874] - config caching causes stale data to be seen from read-only pages
- [RHQ-875] - Plugins should only be redeployed if they've changed
- [RHQ-878] - upgrade install fails on oracle
- [RHQ-881] - No trait data found ( for Xeon x5460 but also for Core Duo on Macbook)
- [RHQ-885] - Breadcrumbs not working during browse/create dynagroup
- [RHQ-889] - NumberFormatException when viewing partition event details
- [RHQ-894] - Duplicate agent connect messags in log (HA)
- [RHQ-897] - config caching causes stale data to be seen from read-only pages - redux
- [RHQ-899] - JMX plugin: MBeanResourceComponent.getAvailability() may erroneously return DOWN for an MBean Resource that was discovered by means other than a JMX query (e.g. an EAR or WAR discovered by ApplicationDiscoveryComponent's filesystem scan)
- [RHQ-903] - Dashboard Auto-Discovery portlet - default "maxiumum platforms to show" of 2 is way too small
- [RHQ-908] - Agent fails to start after upgrade
- [RHQ-922] - DELETED resources continue to be processed as part of the server resource sync info
- [RHQ-924] - Adding resource manually results in NPE
- [RHQ-925] - In affinity group edit page (Agents) "Agent name" column data shows server name, not agent name
- [RHQ-935] - Invalid agent references at server startup
- [RHQ-948] - [java.sql.BatchUpdateException: ORA-01400: cannot insert NULL into ("RHQADMIN"."RHQ_PRD_VER"."VERSION")] error in ContentSourceManagerBean during processing of sync report for JBossAS Patch content source
- [RHQ-962] - V1.0.1 Agent upgrade scripts should be removed from 1.1.0 distribution
Sub-Bugs
- [RHQ-168] - Event Alerts not working properly
- [RHQ-328] - After creating a resource group definition, the next screen should allow you to enter conditions
- [RHQ-404] - group operation history shows FAILURE when it should read IN PROGRESS
- [RHQ-411] - "Required" error message on group definition change is unreadable
- [RHQ-448] - Link from Alert email links to a page where the actual condition is not shown
- [RHQ-628] - Operations that are in progress, show an error for their results
- [RHQ-746] - Bracketed sub-expressions with '=' in them don't work
- [RHQ-747] - ClassCastException when defining GroupDefinition
- [RHQ-803] - ClassCastException when specifying a pluginConfiguration or resourceConfiguration property whose type is logically an integer
- [RHQ-821] - group def allows string functions to be suffixed onto non-string types, but then blows up during calcuation
Code Changes
- [RHQ-369] - NPE when clicking on "OK" when no config is present
- [RHQ-508] - RecursiveGroupTest swallows NPEs
- [RHQ-554] - One filesystem discovery blowing up shouldn't stop all the others being inventoried
- [RHQ-597] - Ensure purge operations handle symbolic link following, if necessary
- [RHQ-713] - ConcurrentModificationException in ResourceMetadataManagerBean during plugin registration
- [RHQ-735] - alert cache needs to be resiliant to failure
- [RHQ-785] - ResourceContainer.measurementSchedule set goes null under certain conditions, causes NPEs
- [RHQ-820] - allow alerts cache to load and unload cache elements on an agent-by-agent basis
- [RHQ-838] - EMS and overloaded operations
- [RHQ-845] - colliding plugin configs cause inventory report failure
- [RHQ-859] - logic error - check for whether a resource is new is reversed in RuntimeDiscoveryExecutor.discoverForResource()
- [RHQ-898] - ConcurrentModificationException in AvailabilityExecutor.checkInventory()
- [RHQ-926] - move updated to agent.lastAvailabilityReport into a separate transaction
- [RHQ-927] - use direct jdbc insertion for out-of-bound message data
- [RHQ-936] - reduce contention on * [_agent table by updating status bit in a single update statement in a REQUIRES_NEW txn
- [RHQ-941] - MeasurementScheduleRequest is put into a Set/HashSet, but doesn't implement required contract methods
- [RHQ-942] - oob insertion failures crop up when under high load
Tasks
- [RHQ-95] - General Jira for UI issues with IE 6.0.2, post as comments each issue
- [RHQ-556] - publish javadocs
- [RHQ-568] - Disable web services
- [RHQ-890] - ensure that plugin configurations make it back down to the agent after starting --clean
Sub-Tasks
- [RHQ-652] - button beneath alert definition table are in a column, instead of across the screen
- [RHQ-668] - alert condition cache segmentation
- [RHQ-670] - HA schema, entities, dbsetup and dbupgrade scripts
- [RHQ-674] - distribution algorithm for server primary
- [RHQ-679] - make sure quartz jobs are clusterable
- [RHQ-698] - document use cases so QA can generate a test plan
- [RHQ-703] - HA handshaking
- [RHQ-786] - use simple text summaries when displaying recovery info within alerts audit trail pages