Patch-ID# 144221-17

20360919 missing deliverable from T144221-16
 
(from 144221-16)
 
20282100 remove debug statement from 18600846 in sc32 patch
 
(from 144221-15)
 
18514674 problem with Solaris Cluster sysmgmt
18600846 problem with Solaris Cluster sysmgmt
18651812 problem with HA-API
18877854 during centralized scinstall scrcmd coredump
18887941 clzc install shows lots of  "usage:"  messages and install fails on other nodes
18889248 fail to install cluster bits using "clzc install-cluster" on S10xip ZC
20052719 thread waiting on a PxFS mutex but mutex owner does not execute PxFS code
 
(from 144221-14)
 
15791814 incorrect file size update on a directio ON file
15799330 data corruption in PxFS when using buffered sequential I/O
15802228 cl_execd_worker failfast panic due to sigkill sent to process group 9
17481033 did-driver 1.77 does not pass ddi-failfast-supported from sd/ssd-driver to SVM
17564911 quorum disk offline due to PGRe key checksum mismatch error
17835456 node panics in PxFS code with getapage: read token not held!
17876594 hang in PxFS code during a switch to redzone
17922323 scha_control giveover causes corruption in rgmd, leading to possible SEGV
18173565 fix for 17481033 incorrect: missed releasing lock
18184018 Parfait build broken by fix for 17481033
18892098 fix for 17481033 incorrect for Solaris 9
 
(from 144221-13)
 
16499276 cldev show -v reports wrong output for some disks
16722387 Cluster sysevents can cause kernel Data segment corruption
16825608 bad trap panic when reporting an exception encountered in read_from_ccr()
17004490 rpc.fed aborts a node due to a validate method being unkillable while it is not
17030436 pagefault on truncated file causes panic
6819197 HAStoragePlus should avoid force import of ZFS pool by default
7147046 HAS+ has to accommodate the zpool_import_props() interface change
 
(from 144221-12)
 
15717310 deadlock in PxFS code during a switch to redzone
15795184 sc_delegated_restarter does not take into account environment variable set in manifest
15797534 BAD TRAP kernel panic when libsecurity stores door refs in kernel
15811789 scsnapshot output does not include the Resource_dependencies_offline_restart property
15818800 reservation keys not written to some LUNs after node joins cluster
15819908 cluster create -i file.xml core dumps when configuring a single-node cluster
15821297 new clusters events in cl_eventdefs.h missing from ClEventDefs.java
15822034 zone halt not triggering failover
15863300 clrs command hangs - deadlock between remote invocation code and prctl command
16079574 reduce window where 15807165, shutdown timer expired can occur
16362048 no minor number conflict checks done when registering a local VxVM disk group
16627499 problem with zone clusters
16762581 problem with zone clusters
7002212 UFS fragmentation can lead to ENOSPC and panic in PxFS while being in GREENZONE
7189211 SUNW.Proxy_SMF_failover sc_delegated_restarter file descriptor leak
7189659 with a large number of non-network-aware GDS resources, some fail to restart and remain offline
7203923 tc_mknod_2 test in SCATE PxFS Functional Tests VxFS causes panic mod_rele_dev_by_major: Unheld driver
 
(from 144221-11)
 
7017334 Failfast: Aborting zone "global" (zone ID 0 ) because "odiozcrgmd" died 30 seconds ago
7034550 old ZC president still does not reset R/RG state for other nodes in some cases
7168450 cl_apid leaks memory
7171538 reservation conflict panic when adding a LUN to a 3-nodes cluster having global_fencing=pathcount
7172767 pmfd fails to start a resource with "Could not setup scheduling parameters: Permission denied" error
7173923 rebalance RG if node dies during enable/disable operation
7180717 kernel memory corruption panic on S10U10 if a user belongs to more than 32 groups
 
(from 144221-10)
 
7167480 memory leak in kmem_alloc_73728 while cluster quorum probe is active
 
(from 144221-09)
 
6927249 performance issue with cluster reference count subsystem
7017334 Failfast: Aborting zone "global" (zone ID 0 ) because "odiozcrgmd" died 30 seconds ago
7034550 old ZC president still does not reset R/RG state for other nodes in some cases
7080718 panic: "mod_rele_dev_by_major: Unheld driver: major number <#>"
7096295 fix for CR 6890418 does not fix the problem
7098067 DCS is reporting misleading message when failing to set a node as primary for a given service
7101649 new interface for SVM to get clusterid
7146553 "libscha XDR Buffer Shortfall while encoding arguments API num : 8" continuously reported
 
(from 144221-08)
 
6345201 RFE: for unkillable method, reboot node only if Failover_mode=HARD
6942090 Cluster.CCR: libpnm system error: Failed to resolve pnm proxy pnm_server.2.<zonename>
7061507 assertion failure for ln!=NULL in rgmd while evacuating
7071068 proxied SMF service is in uninitialized state(on non-primary nodes) on node bootup
7093219 qd_userd is leaking in kmem_alloc_73728 if quorum server is configured with hostname
7099037 PxFS thread is blocked in reserve_blocks() while the file system is already in REDZONE
7099278 PxFS deadlock leading to loss of login due to quota commands hanging
7100066 remove the system tunable pxfs_fastwrite_enabled to stop Fastwrites being disabled
 
(from 144221-07)
 
6748118 deadlock between sendfilev() and check_wait_lock_cache_revoke()
6976154 vmem_hash_delete bad free while doing fobj_ii::unreferenced
7026093 ucmm/vucmm reconfiguration can happen even after a failed reconfiguration when it should not
7072258 clnode show/cluster show fail with a core dump when there are a lot of zones configured
7081070 sc_zonesd has a memory leak
7082313 HAStoragePlus ignores zpool export failures and can double import and corrupt ZFS pools
 
(from 144221-06)
 
6484286 SIOCGXARP failed.: No such device or address from PNM when using link-based failure detection IPMP
7001286 need support for Concurrent SRDF
7004911 on some x86 systems, mkfile of 8g is very slow while mkfile of 1g is fast (PxFS/UFS)
7008750 clzonecluster breaks when using pattern matching of zonecfg
7016260 cluster panics in PxFS code with getapage: read token not held!
7018947 rgmd should not panic the node if door_ucred() returns a non-null value
7019158 scinstall does not set global_fencing to nofencing when requested
7025455 cldev combine -t srdf fails with "No such device or address"
7030111 PxFS switch to redzone
7030333 Solaris Cluster 3.3 clresource create fails "bad string"
7031568 PMM should not failfast for any process that already sets its own failfast
7058024 SUNW.Proxy_SMF_failover fails to start a SMF service having a timeout_seconds value of 0
 
(from 144221-05)
 
6798144 rgmd by default should enable system coredump and disable method coredump
7004673 scha_resource_get STATUS_NODE output for proxy rs sometimes varies when executed on different nodes
7008287 check for metadb and mdcomm(SMF) availability prior to ucmm/vucmm framework startup
7012533 _sdssc_convert_ctd_path can benefit from caching did information
 
(from 144221-04)
 
6591873 memory leak in PxFS when issuing a ENOTSUP on files with mandatory locking set
6986324 NTP service is started before bootcluster and ntpdate reports "can't find host clusternode2-priv"
6993425 SC3.2: process qd_userd is leaking in kmem_alloc_73728 when passing back the results of a door call
6997485 unable to configure a crs-framework resource if LANG=ja
7000805 memory leak in conf_get_node_inaddr
7005699 cluster quorum commands hang again after ZFS fix 6881015 was integrated
7013980 clrg create coredumps with Segmentation Fault
 
(from 144221-03)
 
6873514 scal_dg_probe exits with returned error (32512) due to incorrect use of putenv()
6927853 PxFS: directio(ON) leads to inconsistent file info between client and server after switchover
6955526 srdf.ksh expect immediate "synchronized" status after "establish"
6955824 scvxinstall/clvxvm don't encapsulate fully (due to scrconf failure) when /usr or /var are separate FS
6979079 rgmd startup options specified in svc_rgm_starter are not passed to rgmd processes started by pmmd
6980648 IPMP validation fails when adding logical hostname with IPv6
6986884 Resource_project_name property incorrectly validated due to nscd caching enhancements
6987225 panic in orb invocation code due to a freed context
6987560 PIPE signal is ignored by a process if started with pmf
6987905 rgmd does not accept a project defined in a zone for resource_project_name property for LH resource
6990306 drain_asyncs and putapage hang
6995923 clsetup fails to create SUNW.scalable_rac_server_proxy resource on Oracle RAC 11gR2
6996074 scha_cluster_get performance issue on Solaris 10 vs Solaris 9
 
(from 144221-02)
 
6845964 PxFS switch to redzone can occur prematurely
6865719 clqourum status commands hang after a quorum server goes offline/online
6907541 one node of cluster hangs while shutting down after executing `scshutdown`
6907892 SUNW.Proxy_SMF_failover does not recognize privileges set in method_credential in manifest file
6941843 "assertion failed: write_token.contains (node_id)" panic on PxFS/VxFS when FS becomes full
6962778 hastorageplus_prenet_start times out due to quotacheck running for too long
6964602 scds_pmf_action_script randomly exit 1
6967500 unexpected console messages with clrg unmanage -- 11gR2 native skgxn
6971060 11gR2 CRS may not start if crs_framework rs was disabled during boot up
 
(from 144221-01)
 
6563744 RFE: change crs_framework agent to start CRS if not running
6757800 Solaris Cluster is blocking port due to Solaris 10 labeling capabilities
6761328 rpc.fed does not release the contracts it creates
6890418 tar extract on a PxFS gives incorrect timestamp on some files
6898391 HDS TrueCopy replication with Solaris Cluster cannot handle a loss of host without effecting replication
6914750 scinstall aborted unexpectedly: Unable to determine if "node" has joined the cluster
6917107 placeholder for FBCs 1822 and 1823, Oracle 11gR2 support
6925468 lock contention when accessing the fidlist during unreference processing
6933172 bad trap panic in cl_flk_state_transition_notify
6934273 PATH environment variable limit of 1024 prevents GDS resource start
6934594 geopg create when using TrueCopy fails with "Internal I/O error" and Java cores
6951865 SUNW.crs_framwork resource incorrectly reports ONLINE when 11gR2 CRS is not fully online
6953117 CRS intances rs won't start in ZC 3.3_10
6953269 Oracle CRS proxy resource creation fails on releases prior to 11gR2 when using the wizards
6953389 scrconf fails to set cluster transport network number when numvirtualclusters=0
6954926 rgm core dump when another node joins or goes away
6958692 crs_framework resource may report offline sooner than all CRS processes exiting

--------------------------------
 
There are three (3) possible procedures for installing patches on Solaris
Cluster.  The proper method to use and any additional instructions
for this patch are specified below in the "Special Install Instructions"
section.
 
Refer to the chapter entitled "Patching Sun Cluster Software and
Firmware" in the "Sun Cluster System Administration Guide" for a
description of the different install processes and instructions on how
to install Solaris Cluster 3.2 patches.
 
Refer to the man pages for instructions on using 'patchadd' and
'patchrm' commands provided with Solaris, to install and remove patches.
 
Read all the instructions in this README before applying or removing
this patch.

-----------------------------
 
NOTE 1:  Before you install this patch, ensure that there are NO pending
         reconfigurations in progress.
 
NOTE 2:  Install this patch by using either of the following patch install
         procedures found in the "Patching Sun Cluster Software and Firmware"
         chapter of the "Sun Cluster System Administration Guide":
 
         * Rebooting Patch (Node)
         * Rebooting Patch (Cluster and Firmware)
 
NOTE 3:  The fix for BugId 7001286 (need support for Concurrent SRDF) allows
         for support of EMC concurrent SRDF in a three-site implementation.
         To use EMC concurrent SRDF, follow the procedure:
 
         "How to Configure an EMC SRDF Replication Group"
 
         in the Oracle Solaris Cluster System Administration Guide with
         the following changes:
 
         1) Before starting the procedure, add the following entry to the
            Solutions Enabler options file on all participating cluster nodes:
 
            SYMAPI_2SITE_CLUSTER_DG=<device-group>:<rdf-group-number>
 
            where <device-group> is the name of the device group, and
            <rdf-group-number> is the RDF group that connects the host's local
            symmetrix to the second site's symmetrix.
 
         2) When performing the above procedure, add the following parameter
            to all symrdf commands:
 
            -rdfg <rdf-group-number>