README VERSION : 1.1 README CREATION DATE : 2015-12-01 PATCH-ID : VRTScps-7.0.1.000 PATCH NAME : VRTScps 7.0.1.000 BASE PACKAGE NAME : VRTScps BASE PACKAGE VERSION : 7.0.0.000 SUPERSEDED PATCHES : NONE REQUIRED PATCHES : NONE INCOMPATIBLE PATCHES : NONE SUPPORTED PADV : sol11_sparc (P-PLATFORM , A-ARCHITECTURE , D-DISTRIBUTION , V-VERSION) PATCH CATEGORY : MEMORYLEAK PATCH CRITICALITY : CRITICAL HAS KERNEL COMPONENT : NO ID : NONE REBOOT REQUIRED : NO REQUIRE APPLICATION DOWNTIME : YES PATCH INSTALLATION INSTRUCTIONS: -------------------------------- Please refer to Install Guide for install instructions and Release Notes for other information. PATCH UNINSTALLATION INSTRUCTIONS: ---------------------------------- Please refer to Install Guide for uninstall instructions and Release Notes for other information. SPECIAL INSTRUCTIONS: --------------------- NONE SUMMARY OF FIXED ISSUES: ----------------------------------------- PATCH ID:VRTScps-7.0.1.000 3601140 (3667755) Strict permission check for Coordination point Server (CPS) configuration files. 3757264 (3758088) VCS cluster intermittently loses connection with CP server due to some memory leaks in vxcpserv process. 3851446 (3851443) Fencing fails to start in IPMP networks with certain valid IPMP group name formats. SUMMARY OF KNOWN ISSUES: ----------------------------------------- NONE KNOWN ISSUES : -------------- NONE FIXED INCIDENTS: ---------------- PATCH ID:VRTScps-7.0.1.000 * INCIDENT NO:3601140 TRACKING ID:3667755 SYMPTOM: A malicious user can alter CPS configuration files and thus the CPS fails to properly start the next time. DESCRIPTION: If CPS configuration files have relaxed (non-root) permissions, a malicious user is able to change attributes, such as the database path or certificate location, in the files. Starting CPS with changed configuration may result in client clusters failing to connect to CPS. RESOLUTION: While starting up, make sure CPS (while starting up) makes sure that the configuration files and database have the root only access; otherwise it does not start showing appropriate error. * INCIDENT NO:3757264 TRACKING ID:3758088 SYMPTOM: VCS CoordPoint agent intermittently logs error messages like below and faults- "CoordPoint:coordpoint:monitor:Unable to reach coordination point (or) local node not registered on coordination point" DESCRIPTION: VCS cluster intermittently loses connection with CP server when the memory footprint of vxcpserv process on CP server grows too large. RESOLUTION: The code has been modified to plug the memory leaks in vxcpserv process. * INCIDENT NO:3851446 TRACKING ID:3851443 SYMPTOM: Fencing fails to start in IPMP networks and cpsadm command fails with following error: I/O error: 6 DESCRIPTION: Fencing fails to start in an IPMP network when an IPMP group is configured with a name more than certain length (14 characters) and/or the name has a character such as _.The cpsadm command throws the following error on execution:C I/O error: 6 This limitation comes from a certain version of POCO library with which CPS is compiled. RESOLUTION: The POCO library version is upgraded for CPS, with which the above problem no more occurs if the IPMP group name is valid. INCIDENTS FROM OLD PATCHES: --------------------------- NONE