Patch-ID# 122122-02 NOTE: *********************************************************************** READ THE TERMS OF THE AGREEMENT ("AGREEMENT") IN THE LEGAL_LICENSE.TXT FILE CAREFULLY BEFORE USING THIS SOFTWARE. BY USING THE SOFTWARE, YOU AGREE TO THE TERMS OF THIS AGREEMENT. IF YOU DO NOT AGREE TO ALL OF THE TERMS, PROMPTLY DESTROY THE UNUSED SOFTWARE. *********************************************************************** Keywords: esm enterprise storage manager storagetek operations manager Synopsis: Sun StorageTek Operations Manager 4.1 Manager Patch Date: Jul/10/2006 Install Requirements: None Solaris Release: 8 9 SunOS Release: 5.8 5.9 Unbundled Product: Sun StorageTek Operations Manager Unbundled Release: 4.1 Xref: Topic: Relevant Architectures: sparc NOTE: This patch does not address or correct issues pertaining to a standard SVR4 package. Therefore: 1) This patch is NOT installed via /usr/sbin/patchadd nor removed via /usr/sbin/patchrm. 2) This patch will NOT be reflected in the Solaris patch database or in /var/sadm/patch. 3) This patch will NOT be searchable or traceable via /usr/bin/showrev. Complete installation instructions are found below. BugId's fixed with this patch: 420001 420002 420003 420004 420005 420006 420007 420008 420009 420010 420011 420012 420013 420014 420015 420016 420017 420018 420020 420023 420024 420025 420026 420028 420029 420030 420032 420034 420035 420036 420038 420039 420040 420041 420042 420043 420044 420045 Changes incorporated in this version: Patches accumulated and obsoleted by this patch: Patches which conflict with this patch: Patches required with this patch: Obsoleted by: Files included with this patch: README.122122-02 SP_4_2_0_2.bin Problem Description: Issues corrected in this patch -- 420023 Very slow replication in large NetApp environments Problem: The get all element details (GAED) process for a large NetApp environment is excessively long. One known customer case has 16 filers taking over 17 hours to complete a full GAED. Symptom: Long GAED times in discovery groups with NetApp devices. Solution: The NetAppComputerSystem class is now cached so that the version of ONTAPI doesn't have to continually be requested from the device. System names of each NetApp device being managed are now stored in a static map so they aren't continually requested via ONTAPI. Logging has now been added for all NetApp ONTAPI calls so better modeling can be done if there are further performance problems. The NetAppConnection lookup method has been improved to take better advantage of the hash map of device connections within CXWS. Individual NetAppComputerSystem instances are now being filtered. 420024 Host Security Group creation fails with CommandView/EVA 5.0 Problem: Creation of new host security groups (HSGs) for EVA using the version 5.0 provider always fail with an error code 4. Symptom: HSG creation fails with an error code 4. Solution: A workaround was identified by the EVA provider team to not pass any null valued input parameters when executing an extrinsic method. In this case, the CreateProtocolControllerWithPorts method specifies that some values can legally be null. However, the EVA provider cannot handle null valued parameters. The workaround has been tested against several 5.0 provider versions as well as 4.x. An unpublished system property will turn off the new behavior in case there is a version that the workaround doesn't work for. It is a global setting for all EVA providers. The system property is 'smi.hp.eva.useNullParamWorkaround'. Setting the value to anything non-null except 'true' will turn off this new default behavior. 420025 McData i10k miscounts trunk ports Problem: The McDATA i10K switch can trunk multiple ISL ports together. In this configuration, we are showing all trunked ports in the port count. We should only be showing the primary port in the trunk. All subordinate ports should not be counted. Symptom: The port table on i10K switches shows subordinate ports in a trunk configuration. Also, the port counts for the switch are incorrect. Solution: When the port information is received from SWAPI, each port will be checked to see if it is a subordinate port in a trunk configuration. If the port is subordinate (SwapiPortState=12) the port will not be added. All other ports are added. 420026 HP MPIO for EVA GL active/active and EVA XL active/active Problem: Support for HP MPIO on Windows 2000 and 2003 will be added as a new feature We will support HP's Device Support Module (DSM) for MPIO on EVA. We will support both XL and GL EVA arrays. (XP will not be tested, except in R5.1, but the code is in SP2.) Symptom: Prior to these changes, topology is limited usually one lun is displayed and others are shown unmounted. Results are confusing. Solution: To support HP MPIO, 9 new multipathing CIM classes are created. To get the information to populate these classes, the CLI provided with the full-featured DSM installations is used. (For this reason, since basic DSM does not provide a CLI, it is not supported.) The CLI commands used to retrieve MPIO multipathing data are: hpdsm devices hpdsm paths device=x Where x equals the device number returned from the previous command. The registry is used to find the location of the CLI and to get version information. WMI classes installed when HP DSM for MPIO is installed is also used to get basic information. 420028 HDvM 5.0 and CV/AE 5.0 Device Cache refresh fails Problem: When refreshing a device, the password is not required. However, since our software was expecting a password, it incorrectly supplies the password as 'null'. This is an error which results in the array not being refreshed. Symptom: When getting all details with the 'Force Device Manager Refresh' option checked the operation fails to refresh the HDvM database. Also, during certain provisioning failures, which provoke a refresh of the array in HDvM, the same failure will occur. Solution: when refreshing a device, the HDS provider queries the connection parameters for the storage array and then passes those parameters into the API function that refreshes the device. If a password is not present in the connection parameters, then it should not be supplied to the call to refresh the device. This allows backwards compatibility with earlier versions of HDvM. 420029 Include Element name and WWn in SNMP traps Problem: Storage Essentials generates SNMP traps from events in the Event Manager. In the SNMP trap definition there are fields for Element Name and WWN. These fields were not populated, even for events where we have a corresponding element (device) and WWN. Symptom: Outgoing SNMP traps do not have the Element Name or WWN filled in. This can be seen in 3rd party SNMP trap receiver software. It makes it impossible to correlate events from Storage Authority with events from other sources.Solution: n SnmpServiceBean.java the event type will be examined. If it is a event relating to a host, switch, or storage system, the element name will be included in the trap. If the event is for a switch or storage system, the device WWNN will also be placed in the outgoing SNMP trap. 420030 Hosts disappear from mv_hostfsvolumevw Problem: Hosts disappear from the materialized view MV_HOSTFSVOLUMEVW Symptom: Servers missing from view mv_ MV_HOSTFSVOLUMEVW randomly after Report cache refresh. This could affect several reports, such as the host utilization reports and the Host Use of SAN Storage Details and Host Use of SAN Storage-Summary. Solution: Modified MV_HOSTFSVOLUMEVW and MV_HOSTSTORAGEALLOCATIONVW to take into consideration the raw volume flag while comparing the timestamp to max time stamp. Previously, this time stamp could be incorrect, which would cause the volume to be skipped when the view was populated. 420032 Slow startup registering for indications Problem: The management server takes a long time (more than usual) to start Symptom: Delay in service startup. The delay usually appears in the appiq.log file as a long gap immediately prior to the line that says 'INFO [main:com.appiq.service.event.eventService.IndicationHandler] Indication Handler Started'. This line will appear once for each discovery group that is used. Solution: Eliminated an unnecessary enumeration of the CIM_System class when subscribing to indications from the product's cimoms. Enumerating this class could result in excessive contacting of the managed devices. 420034 Secure admin pages Problem: Several web pages were not secured against access by a user with a less-privileged role. Symptom: Users can access web pages denied by his role. If the user enters the URL for these pages in the browser's address bar, the pages get loaded. Solution: Edited SecurityRestrictionFilter to secure pages. 420035 Change Brocade serial # to return chassis serial number Problem: The serial number for Brocade switches was not appearing in the product. This change happened in R4.2. In previous versions, Brocade switches would display the manufacturing serial number. If that was not available, the WWNN of the switch would be displayed. With this change, Brocade switches now display the chassis serial number rather than the manufacturing serial number. Symptom: The serial number for Brocade switches was not showing up on the navigation page, the properties page, in reports, and in the CLI. Solution: Instead of fetching the manufacturing serial number from the switch attributes, chassis serial number is fetched from the SwitchEnclosure attributes. Information as per Brocade: Switch enclosure is not supported by FOS 2.6.x, but it is supported by FOS 3.2.x (& above) 4.2.x (& above) and 5.x. Neither serial number (MfgSerialNumber or ChassisSerialNumber) is reliably populated and if a value is present, it does not necessarily match the SN on the physical chassis. 420036 FSRM zip file corruption Problem: File SRM data can fail when unzipping the compressed data on the management server when the data is collected from a Windows File Server. Symptom: File SRM data collected is not uploaded into the Management server database. Because of this, no File SRM data is reported in the product. Solution: The logic flow is that file scans save their data in a text file on the File Server. This text file is compressed to a zip format and then encrypted prior to being sent to the management server via FTP. On the management server, the file is decrypted and unzipped. The unzip was failing because the decryption process created a corrupt file. The solution is to no longer add the extraneous 32 bits during the encryption process. Additionally, a system property 'fileSrm.blockEncryption' is checked. If its values is 'true', encryption will not be used at all. 420038 File SRM can use incorrect UNC paths for network file scans Problem: The Windows provider can report incorrect UNC paths for network drives. Symptom: File SRM scans for network drives can either fail or work at the share level rather than the directory of a mapped drive letter. The File SRM scan UI will display the incorrect UNC path for mapped network drives. Solution: The Windows provider will now get the UNC share name from the Win32_NetworkConnection CIM class if that information is not returned by Win32_MappedLogicalDisk. 420039 Allow periods in custom names Problem: Custom names do not allow periods. This causes difficulties for customers that want to use fully-qualified DNS names as custom names. Symptom: Attempts to set a custom name for an element that contain periods will fail with a JavaScript alert message. Solution: The JavaScript validation method used for custom names has a parameter indicating whether periods are accepted. The value for this parameter was changed to true. This change was only made for the element properties page. 420040 Cannot save topology if tape libraries have been discovered Problem: If a tape library was discovered, users cannot save the topology layout from System Explorer. Symptom: If a tape library was discovered, attempts to save the topology layout from System Explorer fail. The Java applet console reports the following: Exception in thread "AWT-EventQueue-30" org.jdom.IllegalDataException: The data "null" is not legal for a JDOM attribute: A null is not a legal XML value.' Solution: The code to get an identifying WWN for an element only handled hosts, switches, and storage systems. This method was extended to handle tape libraries as well. 420041 Brocade 48000 port numbers are incorrect Problem: The port numbers for Brocade 48000 were inaccurate and some of the ports were missing from the list of ports displayed on navigation page. This would occur for switches with more than 16 ports in a slot. Symptom: The port list on the navigation page was displaying inaccurate information. Port numbers could be incorrect, and some ports might be missing. The missing ports could also break any paths that used those ports. These paths would not be reported by the product, and the mount points that used them would appear as if they were local. This problem can easily be detected if output of 'switchShow' CLI command is compared with information on the navigation page. Solution: There is a specific way to calculate port numbers for Brocade director switches. Prior to this fix, the same algorithm that was used for Brocade 12000 and 24000-series director switches was also used for the 48000 line. In this fix, a new algorithm provided by Brocade has been implemented to calculate port numbers for Brocade 48000 switches. This algorithm uses slot number and the port number information provided by the Brocade API to calculate unique port numbers. 420042 Need to identify hosts with unresponsive SMI-S providers Problem: SMI-S discovery task does not log access point information in case of discovery failure / timeout. It's very difficult to find out which access point is failing if you are discovering multiple access points. This is important because a unresponsive SMI-S server can cause discovery to take a very long time (several hours) before timing out. These servers need to be identified and either corrected or removed from the discovery list. Symptom: Discovery failure exceptions in the appiq.log files. Solution: The access point information is included in the log messages in case of failure 420043 Disk statistics collection for windows host is highly inefficient Problem: Disk statistics collection for Windows hosts is highly inefficient. The Windows provider contacts all the discovered hosts when attempting to query disk statistics for particular host. Symptom: In an environment with many Windows hosts in the same discovery group, this can cause problems in getting any Windows information. The requests can create a backlog which could cause requests to time out. In extreme cases, it also might exhaust the system supply of sockets, leading to networking problems. Solution: Instead of querying every Windows host, the provider now identifies the host from disk object path and queries only that host. 420044 EVA Volume creation fails Problem: Volume creation against SMI-S providers using job control returns an error even though the operation is ultimately successful. This only affects the CommandView/EVA SMI-S provider, since that is the only supported SMI-S provider that uses job control. Symptom: The message 'Unexpected result from provisioning call: null' is displayed to the user. Solution: The job control status code checking was tweaked slightly to check for a status of 2 (OK) AND 17 (Completed) in order for a job to be deemed successfully complete. Previously this was an OR and the status 2 was being inadvertently picked up as complete when it may not have been. 420045 Support fo CIM Extensions deployment tool Problem: HP-SIM includes a tool to automate deployment of Storage Essentials CIM Extensions. In service pack 1A, the Windows CIM Extension could not be deployed using this tool. Symptom: Attempts to deploy Windows CIM Extension service packs using the deployment tool would fail. Solution: The deployment tool assumes a fixed name for the CIM Extensions. Starting with this service pack, the Windows CIM Extensions will have this name. The deployment tool also assumes that the CIM Extensions are complete installations. Previous service packs only included the changed files for Windows CIM Extensions. With this change, the entire Windows CIM Extension is included. Issues correct in previous versions of this patch -- 420001 Inability to delete 3.x storage system Problem: After upgrading pre-4.0 databases storage element delete will result in error. Symptom: After upgrading pre 4.0 databases storage element delete will yield java.sql.SQLException: ORA-02292: integrity constraint (APPIQ_SYSTEM.CARD_CONTAINER_ID_FK) violated - child record found. Solution: Created sql script SP1DBChanges.sql to clean up the orphan records and fix the card containership. This script is run during installation of this service pack, and during upgrade to future releases. 420002 Inability to view configuration/reports after deleting user who scheduled report Problem: Deleting a user breaks Configuration | Reports . The underlying problem is that when a user is deleted, schedules that were created by the user are not updated to reflect the deletion. Symptom: 1. Create a user and log in as that user 2. Schedule a report for delivery 3. Log in as admin, and delete the user created in step 1 4. Try to go to Configuration | Reports. You will see an error like: " Cannot find data model with id:com.appiq.toolkit.security.impl.UserImpl:1001" Solution: There are 2 parts to the solution: 1.A script is run as part of installation of this service pack and upgrade to delete any scheduled tasks owned by deleted users. 2.All user deletion is now trapped and their scheduled tasks are now deleted. 420003 Inability to restore position of unmanaged elements with WWN of all digits Problem: For unmanaged elements, System Explorer saves their positions by using their WWN. When retrieving information from the database, System Explorer mistakenly treads the WWN of all digits as an element ID. This causes the position of corresponding elements to not be restored. Symptom: Users can not restore the positions for the unmanaged elements that have the WWN of all digits Solution: When System Explorer gets a saved WWN of all digits, first search all managed elements to see if there is a match. If no matching element exists, the WWN will be used to retrieve the unmanaged elements. Note that WWNs contain 16 characters, which effectively makes these values impossible to be considered element IDs. Element IDs are numbers that increment by 1. 420004 Failure in chargeback of currency validation for some currencies Problem: User input validation fails for Asian currency symbols. Symptom: When Asian currency symbols such as the Yen symbol are entered in Chargeback, a Javascript error message appears informing the user that the currency symbol is invalid. Solution: There are 2 different unicode representations/code points for most Asian currencies. For example, there is both a full-width Yen sign and half-width Yen sign. Java always returns full-width Yen; however, the previous validation was done against the half-width Yen sign, thus it failed. The fix introduced in the code covers validation for both code points. We now validate against a list containing full-width currency symbols obtained from Java and the half-width symbols we had previously supported. 420005 Inability to display element topology if there are unmounted paths Problem: Element Topology only displays mounted external (form mount point to storage system) paths. Unmounted external paths are not displayed. Symptom: User can not see external, unmounted paths in Element Topology. Solution: Display the external paths for an unfiltered and unmounted path. 420006 Inability to discover NetBackup on Linux Problem: NetBackup discovery on Linux fails Symptom: Backup details could not be collected for Linux Hosts. Solution: The solution is to remove DOS end-of-line characters from the scripts. This change is made in cimiq.xml after copying the files to the agent directory under cimiq_dist. A special ant tag removes all the EOL characters from all the files in the directory. 420007 Support for Brocade 48000 and 200E Problem: Brocade 48000 switches did not accurately report the number of ports.Symptom: There were errors in the sync process and the topology was not be correct for models: 48000 4100 200E 2000 3016 4012 3014 4020 7420. Solution: Brocade 48000 is now added to the list of supported director switches to get the port information correctly. Director switches calculate ports differently than other Brocade switches because director switches have multiple port modules. Support has also been added for displaying the model property for Brocade switches listed above. 420008 Support for multiple Sun 35xx arrays on a single DotHill provider Problem: Fixing replication of multiple Sun 3510 arrays that are managed by single Dot Hill SMI-S provider. Also, handling the port deviceId field, which the SMI-S provider was supplying in an unexpected format. Symptom: There were replication errors when multiple arrays were managed by the same provider / server. There were also array configurations that could cause the SMI-S provider to supply deviceId in unexpected format. This resulted in missing ports and host security group information on the navigation page and incorrect topology. Solution: Usage of port property DeviceID is totally removed. The multiple array errors were happening because of incorrect caching. Caching has been fixed to resolve this error. 420009 Host data not collected by SIM (HP SIM integrated only) Problem: With the integration of the HP-SIM and the HP-SE products, the initial discovery of a Host device that can be monitored by both HP-SIM and HP-SE (the CIM Extensions are present) can result in a condition whereby the HP-SIM fails to successfully collect data from that host. This fix will allow the users to successfully gather host data from either HP-SIM or HP-Se Symptom: When a host has been discovered that can be managed by both HP-SIM and HP-SE, the scheduled data collections to that host performed by HP-SIM will fail. The user will receive a notification that the data collection event was aborted for a general reason. The users will not see the expected host data within the HP-SIM reporting structure. Solution: The code was changed such that when HP-SE is populating host information back into HP-SIM it correctly provides all expected data. HP-SIM is expecting a piece of string data that HP-SE was supplying as a null value. By providing a non-null value to HP-SIM along with the host data, HP-SIM is able to correctly process the device and the ability for HP-SIM to collect data from that host is not affected. 420010 HP-SIM does not know when SE discovery is complete (HP SIM integrated only) Problem: With the integration of the HP-SIM and the HP-SE products, there is a case where devices discovered within HP-SIM may not successfully be added to a discovery task within HP-SE. When HP-SIM is able to successfully identify a device, or at least able to ping a device, the device IP address information and credential information to the HP-SE product for it to perform a Step 1 discovery step. Prior to sending this information to HP-SE, the HP-SIM product via the connector, checks to ensure that HP-SE is not currently running a discovery operation. If a discovery operation is occurring, the HP-SIM product will hold onto the newly discovered device information until such time as it determines HP-SE has completed its discovery steps. When HP-SIM checks to see if HP-SE is doing a discovery operation, it is only looking to see if the Step 1 (Discovery Setup) discovery is running. If a new device or set of devices is discovered within HP-SIM when HP-SE is in some other discovery set, such as Step 2 Topology or Step 3 Details, HP-SIM will incorrectly determine that HP-SE is not doing a discovery operation, and will hand over the information for the devices that HP-SIM just discovered. HP-SE will accept the information from HP-SIM, but it will not run a Step 1 Discovery Setup step against the devices. Symptom: The users will see devices that have been discovered within the HP-SIM product, but have not been also discovered by HP-SE. There is no visual feedback to indicate that something has not worked correctly. The users must re-identify the devices in HP-SIM when HP-SE is not in a discovery step Solution: The code was changed such that the call made by HP-SIM to find out if HP-SE is currently in a running a discovery task will now check for all possible HP-SE discovery states. HP-SIM will hold off on sending new discovery information if HP-SIM detects that HP-SE is in either a Step 1 Discovery Setup, Step 2 Topology or Step 3 Details discovery operation. Once HP-SIM is able to adequately determine that HP-SE is not performing any discovery operations, HP-SIM will hand over the new discovery information, and instigate a Step 1 Discovery Setup discovery step in HP-SE. 420011 System Explorer slow to open in large NAS configurations Problem: In a system with many NetApp shares, starting System Explorer can take an extremely long time. Symptom: In a system with many NetApp shares, starting System Explorer can take hours. Solution: The calculation of host dependencies on NetApp servers has been optimized. It had been extremely inefficient. 420012 Volume creation on CV/EVA 5.0 fails Problem: Attempts to create a Storage Volume on an EVA system managed by CommandView 5.0 will fail. Symptom: Storage Volume creation would fail. Solution: We now properly escape the CIM object path for the configuration service as per RFC 2396. We had been using java.net.URLEncoder.encode, which escapes spaces as plus signs rather than '%20' as per the RFC. (Cf http://bugs.sun.com/bugdatabase/view_bug.do'bug_id=4616184) 420013 Create Host Security Group can fail on CV/EVA 5.0 if the proxy provider had previously managed a second storage array Problem: If you attempt to create or edit a Host Security Group on an EVA system managed by CommandView 5.0, you may get errors if the CommandView host also managed other EVA systems. Symptom: Host Security Group creation or modification would fail. Error code 4 (generic failure) would be returned by the EVA provider. Solution: Until now, we were checking to see if a given WWN was known to a provider by making an EnumerateInstances call and checking the WWN against the StorageID property of CIM_StorageHardwareID. This works in the case of embedded providers, but it breaks down in the case of a proxy provider managing multiple arrays because a given initiator WWN will have separate instances for each of the arrays it is known by. Instead of the EnumerateInstances call, we are now making an AssociatorNames call from the StorageSystem to the StorageHardwareIDManagementService and then an Associators call to the StorageHardwareIDs. 420014 We do not get Windows serial numbers Problem: The product does not show Serial Number for windows hosts. This is important for asset tracking. Symptom: UI and reports (host properties and chargeback) do not show the serial number for windows hosts. Solution: An association associating physical package with computer system was missing. Added the association class and provider for windows hosts. 420015 Storage system discovered with generic SMI-S have replication errors forCIM_ProtocolControllerForPort Problem: Storage systems that are discovered with the generic SMI-S infrastructure do not show the mapping to front end ports for HSGs. Symptom: Replication errors for class CIM_ProtocolControllerForPort occur during a GAED. Solution: Because the CIM_ProtocolControllerForPort association requires a ClassMap to help the mapper connect the front end ports to the protocol controller instances, we use a replication bean to set the ClassMap appropriately depending on whether the FC Port instance is a front-end or back-end port.. 420016 Error discovering SQL Server after installing Windows security pack Problem: SQL Server database applications can not be discovered from Windows management servers that have certain Microsoft security patches installed. Symptom: Discovering SQL Server application gives "ORA-12154: TNS:could not resolve service name". Solution: Forced ODBC connections to use TCP/IP instead of named pipes by modifying the registry key for ODBC connection Network Library. The default transport was changed by Microsoft in several of their security patches. 420017 File SRM fails with very long directory names Problem: If a very long directory path is scanned on a Windows File Server, the scan process will terminate silently Symptom: Scanning network drives from a Windows File Server that has paths > 260 characters will terminate the scan. Storage Essentials will not receive the scan results, and will display an asterisk next to the network drive. Solution: We expanded the directory buffer used to 1024 characters to create a UNC path. Note that Windows MAX_PATH value is only 260 characters. However, some network drives, such as those mounted on NetApp servers, can support longer paths. We also changed the code that gets the Windows SID (Security ID) to handle the long paths as well. 420018 Tru64 CIM Extensions crash if no HBA API Library Problem: The Tru64 CIM Extensions that shipped with R4.2 do not work with the HBA API Library that currently ships with Tru64. Symptom: On startup of the Tru64 CIM Extension, the following error will appear: java.lang.UnsatisfiedLinkError: dlopen: Cannot map library libhbaapi.so After that, the CIM Extension will stop running. Solution: We rebuilt the CIM Extension to work with the latest HBA API library. The CIM Extensions now require either the V5.1B-2 or V5.1B-3 Tru64 patches. 420020 CV/EVA 5.0: Removing a Volume From HSGs Problem: On an EVA accessed through CommandView 5.0, if you remove a volume from one Host Security Group, it will display in the UI as if it is not in any Host Security Groups. Symptom: When you use CommandView 5.0 to manage an EVA and remove a storage volume from a Host Security Group, all other Host Security Groups will report that the volume is not a member as well. Neither CommandView nor the EVA are adversely affected. StorageEssentials just reports incorrect information. The next Get All Elements Details request will correct the problem. Solution: The replication code was replicating the wrong CIM class. It was replicating CIM_PROTOCOLCONTROLLERFORUNIT, but should have replicated HPEVA_ProtocolControllerForVolume. This is now fixed. Patch Installation Instructions: ----------------------------- 1) Execute /etc/init.d/appstormanager stop 2) Navigate to the directory where you unzipped this patch, and execute 122122-02/SP_4.2.0_2.bin 3) Follow the resulting on-screen patch installation instructions. 4) Execute /etc/init.d/appstormanager start Special Installation Notes: ************************************************************ NOTE: Installing this Patch will overwrite the jboss.properties. If you have modified jboss.properties, you must first make a copy of the old one, and then restore it after this Patch installation. ************************************************************ Either the "-i silent" or "-i console" switch may be used if needed. Version Notes: Navigate to the directory where the Sun StorageTek Operations Manager is installed. The file Version.txt contains the following version strings: Version 4.2.0.110 of Sun StorageTek Operations Manager has been installed. SP_4.2.0_1A has been installed on . SP_4.2.0_2 has been installed on . Special Install Instructions: -------------------------------- None. README -- Last modified date: Monday, July 10, 2006