fs-hpux1131-5.0MP1P7
Obsolete
The latest patch(es) : fs-hpux1131-5.0MP1P15  fs-hpux1131-5.0MP1P14 

 Basic information
Release type: P-patch
Release date: 2011-09-05
OS update support: None
Technote: None
Documentation: None
Popularity: 2205 viewed    downloaded
Download size: 8.78 MB
Checksum: 1700004086

 Applies to one or more of the following products:
File System 5.0 On HP-UX 11i v3 (11.31)
Storage Foundation 5.0 On HP-UX 11i v3 (11.31)
Storage Foundation Cluster File System 5.0 On HP-UX 11i v3 (11.31)
Storage Foundation for Oracle 5.0 On HP-UX 11i v3 (11.31)
Storage Foundation for Oracle RAC 5.0 On HP-UX 11i v3 (11.31)
Storage Foundation HA 5.0 On HP-UX 11i v3 (11.31)

 Obsolete patches, incompatibilities, superseded patches, or other requirements:

This patch is obsolete. It is superseded by: Release date
fs-hpux1131-5.0MP1P13 (obsolete) 2014-10-28
fs-hpux1131-5.0MP1P12 (obsolete) 2013-10-16
fs-hpux1131-5.0MP1P11 (obsolete) 2013-04-15
fs-hpux1131-5.0MP1P10 (obsolete) 2012-09-12
fs-hpux1131-5.0MP1P9 (obsolete) 2012-05-09

This patch supersedes the following patches: Release date
fs-hpux1131-5.0MP1P3 (obsolete) 2010-11-30
fs-hpux1131-5.0MP1P2 (obsolete) 2010-10-19

This patch requires: Release date
fs-hpux1131-5.0MP1P6 (obsolete) 2011-07-12
sfha-hpux1131-5.0MP1 2010-07-14

 Fixes the following incidents:
2494165

 Patch ID:
PHKL_42496

Readme file
Date: 2011-08-26
OS: HPUX
OS Version:11.31
Etrack Incidents: 2494165
Errors/Problems Fixed:
2494165 Memory leak in internal buffercache after 497 days (lbolt wrap-over)

MP1P6
=====
2245384 Panic in vx_inode_mem_deinit --> mutex_destroy
2245409 1009: HART Vpars:SYM5.1 6/8:non-telnetable hang after 14+ CHO
2245426 System panic in vx_iupdat_clustblks()
2245471 vx_clone_setup can call vx_ierror when clone inodes are free
2275533 write() system call hangs for over 10 seconds on VxFS 3.5 on 11.23
2277469 CFS panic in vx_irtyhold_locked() due to corrupt freelist
2289635 Panic in vx_inull_list after forced unmount of writable clone and unmount of prinmary fileset.
2348422 Threads stuck in vx_rwsleep_rec_lock_em
2351018 EAU delegation timeouts
2376389 vxrestore man page to add -b option details
2405553 Hit ted assertion "f:fdd_getdev:1" when did verification for e1209803
2405565 11.31/5.0.1:OnlineJFS01.VXFS50-AD-RN configure script errors in IC362 coldinstall
2406455 Panic: "pfd_unlock: bad lock state!"
2406457 Performace issue pointing to read flush behind algorithm
2406458 Add functionality of free space defragmentation through fsadm.

MP1P5
======
2207698 QXCR1000550182 issue, not ported to 4.1/11.23
2207702 vxfsstat does not reflect the change of vx_ninode
2232562 LM noise fullfsck NF1 hits assert f:vx_memlock:1a with 5.1SP1A14 build
2238895 VxFS mmap performance degredation on HP-UX 11.31 and 5.0.1 vx_alloc_getpage() suspected
2245420 fsck fails to repair corrupt directory blocks having duplicate directory entries.
2245423 odmstat with vol_vxtrace 0 shows high WTIME values
2245434 Bloomberg requirements: enable rename() and unlink() calls to succeed on the target file currently under execution
2303153 investigation on ilist HOLE
2340745 One second delays introduced with tranflush mountoption
2348339 QXCR1001094046: Unable to tune vxfs_bc_bufhwm on systems > 227 GB memory
2348363 Request to re-word message generated while performing online deactivation of memory cells.
2348513 Access denied on files inheriting default group ACL from parent directory.
2364911 LM stress aborted due to "run_fsck : Failed to full fsck cleanly".

fsman
=====
2376389 vxrestore man page to add -b option details

MP1P4
=====
2223387 QXCR1001081989 - FCL License not provided as part of Online JFS licensing for VxFS 5.0 on 11.31
2233241 CFS noise.replay hung due to busy loop in vx_statvfs
2207725 QXCR1001072570: pread() system call returns EINVAL
2252142 Panic at getblk() when growing a full filesystem with fsadm

MP1P3
=====
2168326 O_SYNC isn't sync enough. also, missing timestamp updates.
2168341 vx_rdwr should handle the EIO returned from vx_tranidflush
2185390 VxFS5.0 quota information gets messed up on large files

MP1P2
=====
2111616 Modification time is not updated when O_SYNC and nodatainlog mount option used.
2129456 [VxFS][412-240-657][hp-oem] vxfsd is taking a lot of CPU time after deleting some directories
2149410 [VxFS][412-456-910][hp-oem] mmap and msync not updating modification/access times on 5.0/11.31
2162976 5.0.1RP1 LM and CFS kernel conformance->funmount->testvops hits assert "Fault when executing in kernel mode"

MP1P1
=====
2068826 Hang due to disowned beta semaphore
2070455 getacl shows no permission for user but user can still access file
2079589 reduce contention on vx_worklist_lk using bulk enqueue/dequeue
2080725 dm_punch_hole request does not invalidate pages 

i18n(PHCO_41297)
=====
2077422 vx_maxlink can not be changed to the smaller than 32767

MP1
====
1791070 4.1 readahead performance issues on 11.31
1845016 State Map corruption reported, followed by a CFS hang.
1956127 fcache_vn_destroy() called on a vnode with dirty pages
1973793 VxFS doesnt specify limits on tuning on vxfs_bc_bufhwm
1981688 Data key miss fault during hazard testing.
1984818 LM noise replay(UxN2) on build5.0MP3debug-2007-11-7b on SLES9 sp3_rc4is not progressing
2012567 System hang with many processes sleeping in vx_recsmp_range_lock
2012585 VxFS 5.0GA & MP1 - fsadm -b - ioctl(VX_IOCTL | 128) returns EBUSY
2012588 Vxfs log replay fails at double fault for 2TB lun.
2012597 Adding quota support for user 'nobody' on HP-UX
2013376 CFS-Conformance "mount" / "asym_remount" test hit assert "f:vx_msgprint:ndebug" via "vx_olt_iasinit"
2013378 HP-UX 11.31 5.1 SP1 LM :: Version 5 vxfs remount with different logiosizes causes structural file corruption
2014715 ENOTBLK returned via async driver due to fdd not being a block device on 11.31
2016605 quotaon can loop forever
2017775 do not call VX_VN_CACHE_DEINIT() inside VX_INODE_DEINIT() if there are mappings, instead call it in vx_inactive()
2020109 Make the LDH tunable vx_dexh_sz as private .
2028110 open() hangs while vxfsd does vx_fcache_vn_flush()
2029930 Behavior of DST (Access age-based file placement policy with preffered files)
2034240 HPUX11.31/5.0v3MP1 setup running test cfs-stress.reconfig.S5 got exit issue with run_fsck error message.
2038415 system panic during db2 fast pre- allocation with VxFS
2040691 HPUX11.31 5.0v3MP1 running LM stress stanadard test hits an assert f:vx_getimap:1a
2040897 HXRT50MP1V3 fsmap return incorrect Volume name

fspro
=====
2029932 Behavior of DST (Access age-based file placement policy with preffered files)
2029934 incorrect usage threshold calculation while setting threshold to volumes in multi-volume file system.

RP8
===
1882044 Panic'ing/TOC'ing a CFS 5.0.1 node displays ODM error on booting after the panic/TOC
1888709 secondaries ias_ilist not updated fully.
1901644 IAUs are getting corrupted and fsck is marking for full fsck, on a 6TB fs causing outage
1907932 AIX-5.0MP1_61 CFS-Stress+reconfig test hits assert "vx_olt_ilistadd:2f".
1907990 CFS hang while expanding AUs
1907993 vxfs spends too much CPU looking for odd-sized extents (vxi_alloc_fail)
1907997 clone removal can block resive ops
1922445 HPUX11.23 5.0MP2RP1 running test cfs.stress-enterprise hit assert f:vx_imap_process_inode:4a
1907993 vxfs spends too much CPU looking for odd-sized extents (vxi_alloc_fail)
1957034 fsck/basic test failed with "mountit.fatalerror".
1957550 CFS-noise.replay aborted due to run_fsck : Failed to full fsck cleanly on Sles9-SP6 with drop 03-25a
1960658 fsck coredump in cfcomp()
1960660 fsck log replay extremely slow
1960663 CFS: rm performance could be affected due to VX_DIRLOCK() contention
1965703 Enhancement work to allow 100% full filesystems to be resized
1965705 Panic in vx_sum_upd_efree1()
1965712 On HPUX-11.31 Cluster, Test cfs conformance with 5.0.1MPA12 build got failures in test script "fsadm"==>resize
1975557  	sar(1M) -v does not report VxFS inode table overflows

RP7 (b_5_0_hp1131_RP7_maint)
====
1922820 IAUs are getting corrupted and fsck is marking for full fsck, on a 6TB fs causing outage

RP6
====
1375225 : An assert f:vx_fsvm_init:1 is hit when    
1540393 : LM odm stress test hits assert f:vx_mkimtran:1c with  
1587209 : HPUX-11.31 cfs-odmqa-conformance 501A37 build setup test hit  
1591316 : CFS looping in vx_dele_get_freespace() due to smap marked     
1592829 : Panic in fcache_as_uiomove()called with FAUI_NO_GPAGE when    
1593062 : vx_tflush_map() should not disable fs if just map marked bad  
1593480 : AxRT5.0MP3: CFS filesystem needs fsck but is mounted on one   
1602560 : quotacheck coredumps with more than 30 quota- enabled 
1665095 : turn DRB lock into a no-op on non-AIX platforms       
1710912 : subtype command not returning     
1710913 : Heavy rwlock contention on file caused by vx_read_flush().    
1710918 : DPF in vx_dnlc_pathname_realloc       
1710926 : HP LM stress hit an assert "Fault when executing in kernel    
1711827 : vxupgrade 5->6 still fails with ENFILE
1711828 : Need to release CPU in vx_multi_bufinval () for local mount large extent
1711829 : ncheck command dumps core
1711831 : VxFS does not release vfs in the presence of nested mounts.
1781088 : VX_SETEXT ioctl doesn't require full license on 11.31
1787930 : VX_SETEXT with VX_GROWFILE could fail to grow the file or expose uninitialized data.
1790345 : CIO returned ENOTSUP (124) and caused DB2 to crash
1791062 : Panic in fcache_vn_flush() due to race in vx_vget()
1791916 : HP-UX 11.31 cfs Test cfs-stress-enterprise hit the same assert :f:vx_cwfrz_wait:2 is with 501A38 build setup.
1791925 : filesystem performance degradation
1795150 : Panic occurs while doing nested mount when the base cluster mounted base fs gets force unmounted
1799462 : HPUX11.31 test LM cmds-> fsvoladm-> devreorg-> fsvoladm.devreorg.26 is failing with build setup 501A32
1805138 : HPUX11.31 LM-conformance->fcl -> fcl_inode is hit an assert f:vx_msgprint:ndebug on machine with Build :: 501A39 set up
1838743 : Data page fault at vx_qiostats_update() due to fiostats structure already free'd
1839047 : file system disabled.
1839269 : HPUX11.23 SFCFS 5.0MP1RP3HF6 - Request fix Etrack 1122868 - vxupgrade to 7 fields uninitialised
1843818 : System panic in inctext: VTEXT not set and tcount > 0
1853878 : 'mv' hung on CFS

i18n (PHCO_40290)
1705724 : fsppadm.1m manual page needs clarification the analysis keyword's limitations

RP6
====
1603304 : Remove the dependency from VXFS patch PHKL_40059 on ODM patch PHKL_39471

RP6
====
1382183 : Enhancement to vx_getattr() taking IGLOCK in share mode 
1448064 : Fails to create 11g database with ODM on VxFS filesystem 
1473912 : Need to change VX_GET_NINODE value back to 57 in 4.x and later 
1511140 : LxRT 5.0MP3 CFS fsqa fcl/misc fcl_write.2 test hits assert f:vx_getmap:4 
1512755 : Panic in vx_unlockmap due to null ml_tranp pointer 
1514803 : Infinite loop in bmap/transaction processing during clone removal truncate/push operation. 
1518677 : bdf commands hung in VX_CFS_GLOCK_GRANT_WAIT when CVM 
1518685 : SDET performance regression on 128 cpu machines 
1522884 : Checkpoint clonedb fails giving error ORA 600 for online, offline, instant checkpoints 
1523978 : EastWest Neo Medical Center: HPUX11v2 IA64 4.1 + Willow + UNOF_4_1p1cpp1/kpp1 2 node RAC. Suspect panics in fdd_siodone 
1527171 : mkfs fails on EVFS volume 
1531603 : f:vx_extentalloc:1d during policy enforcement on CFS secondary 
1533832 : JAGag22272 write reutrns error, as dio fails due to wrong max iosize estimate 
1478188 : fsclustadm cfsdeinit failed with "device busy". PHKL_37113 installed. 
1534324 : HPUX11.31 running LM-stress.odm-S13 got an assert panic: Data page fault.with build 501A27.
1541914 ; Panic at volkio_to_kio_copy() - ODM async I/O buffer with invalid b_spaddr 
1544434 : CFS MTS inode leaks - inode allocations do not seem to use partial IAU's 

RP3
=====
1385482 : [HPUX11.23 IA SFCFS 5.0MP1] [Samsung] takes time for du/bdf to report the same after files are removed 
1443128 : CFS 5.x primary umount hang when full fsck is set 
1448064 : Fails to create 11g database with ODM on VxFS filesystem 
1448113 : HPUX11.31:fdd:Hitting an assert f:vx_dev_strategy:2a while async reads on clone. 
1452036 : invalid ACLs can be set on HP 
1453978 : odmmkfile/qiomkfile should fail for non-root users 

PHCO_40290( b_5_0_hp1131_ga-rp_maint)
1449931 : odmmkfile/qiomkfile should fail for non-root users 

RP2
=====
1289133 : JAGag24745: Oracle text not getting large pages 
1303571 : Difference in VX_FREEZE behavior in hpux 11.31 as compared to hpux 11.23 
1303572 : vx_lookup returning ENOENT due to stale negative DNLC entry 
1315116 : Assertion panic during hazard C8 run 
1363788 : getacl() fails with Invalid argument 
1364185 : Kernel panics with bad_kern_reference during SFRAC testing on 5.0/11.31 

RP1
====
1234527	VxFS gives EINVAL in write() with Oracle 10.2.0.3.0 (without ODM), on query such as parallel index creation.
1249412	[QXCR1000760213] - UNIX 2003 FIFO modifed times not being updated
1252119	[willow5.0] DPF in vx_cbuf_revoke_blk
1262435	HP 5.0 11.31RP6 LM stress test hit an assert "f:vx_dio_chain_start:1x"

Install and Uninstall Instructions:
-----------------------------------

1.Installing VxFS 5.0 MP1P7 patch:

a)If you install this patch on a CVM cluster, install it one
 system at a time so that all the nodes are not brought down
 simultaneously.

b)VxFS 5.0(GA)  must be installed before applying these
  patches.

c)To verify the VERITAS file system level, enter:

     # swlist -l product | egrep -i 'VRTSvxfs'

  VRTSvxfs     5.0.31.0        VERITAS File System

Note : VRTSfsman is a corequisite for VRTSvxfs.Hence VRTSfsman also
needs to be installed alongwith VRTSvxfs.

    # swlist -l product | egrep -i 'VRTS'

  VRTSvxfs      5.0.31.0      Veritas File System
  VRTSfsman     5.0.31.0      Veritas File System Manuals

d)All prerequisite/corequisite patches have to be installed.The Kernel patch
  requires a system reboot for both installation and removal.

e)To install the patch, enter the following command:

# swinstall -x autoreboot=true -s <patch_directory> PHKL_42496

Incase the patch is not registered, the patch can be registered
using the following command:

# swreg -l depot <patch_directory>,

where  <patch_directory> is the absolute path where the patch resides.

2.Removing VxFS 5.0 MP1P7 patches:

a)To remove the patch, enter the following command:

# swremove -x autoreboot=true PHKL_42496