Date: 2010-08-30 OS: HP-UX OS Version: 11.23 Packages: Etrack Incidents: 1857897, 2110468, 2111239 Errors/Problems Fixed: Incidents fixed in MP2RP10 ================================ VxVM 857897 (1819777) Panic in voldiosio_start() as race window exists while handling duplicate DA records VxVM 2110468 (2097320) vxconfigd not posting DMP_PATH_DISABLE_EVENT to vxnotify clients VxVM 2111239 (2078209) resize of vxvm volume resulting in vxconfigd hang -- Need to increase the default value for volpagemod_max_memsz to 64MB ==============================- Incidents fixed in MP2RP9 ================================ 1857901 (1729558) multiple vxplex attach cmds running in parallel on a volume lead to clearing DCO map and subsequently lead to corruption in FMR2 1904290 (1528932) vxconfigd dumps core in cvmcert run when master tries to destroy dg, slave node dumps vxconfigd core. 1927989 (1927982) vxpfto returns error arbitrarily even though the command sets the values correctly. 1937730 (1933970) Restrict the max_specialio tunable value to the permissible limit. 1978836 (339187) CVM activation tag in vxprint -m output breaks vxprint 2008397 (1138518) vx commands (such as vxdg, vxdctl, vxdisk) hanging 2008403 (2008402) IO hang in A/A arrays due to incorrect setting of quiesced flag when paths fail and come back . 2012574 (1744224) FMR3: multiple vxplex attach cmds running in parallel on a volume lead to clearing DCO map and subsequently lead to corruption ================================ Incidents fixed in MP2RP8 ================================ 1676061 System panic'd after 2 out of 4 paths to disk were removed. 1114178 vxconfigd dumped a core in req_dg_get_info_common() 1059720 HP -- Switching to EBN does not imediately show EFI disks correctly 1742702 vxvmconvert fails, probably due to wrong disk capacity calculation 1192166 vxdg -n [newdg] deport [origdg] causes a sort of memory leak 1677416 CVM join & takeover issues in shared A/P storage config due to not breaking more than 64K size kmsgs ================================ Incidents fixed in MP2RP7 ================================ 1205584 (918352) vxassist fails to relayout striped volume 1427459 (1211302) EVA6k/HP-UX : Slave node panics after disable primary paths. 1535212 (1542605) vxconfigd dumps core while setting PFTO on slave due to garbage in dgc_dm_recs 1555903 (1068626) Full resync occurred in remaining nodes after SFORAC panic rebooted 1555927 (1321475) [QXCR1000760447]: Join Failure Panic Loop on axe76 cluster 1555951 (1133089) HP:11.31:WILLOW:CVM: In consecutive master takeovers, slave state is not reset appropriately 1555954 (1213239) HP:11.31: WILLOW: CVM: Recovery for subvolumes(of a layered volume) does not happen due to missing -f option 1555957 (853822) AxRT5.0MP1: CVM: master stuck in 'master selection' during random shutdown - r (updown) on 16 node cluster 1555959 (1210957) [QXCR1000745273] Vxio hung IO's and uncorrectable write errors 1555962 (1260756) [QXCR1000765820]:EDC testing: vxconfigd core dumps after fix for vxcommands hanging is applied 1555966 (1090155) Pri 3/ HP/ VxVM 3.5/290-682-873/ During vxevac, the vxsd command can absorb all nfile resources in kernel 1555974 (1453694) [Amadeus] System panic in scsi_strategy_real() when an extra paths are added to an existing LUN on the fly. 1555976 (1098233) vxio NULL ptr panic, voldiosio_start fails to cope with NULL pointer returned 1587615 (1207898) Upon stop all nodes in the RAC cluster, some stale fencing keys are left from Master node with HDS USP V array 1587634 (1543908) While running vxevac command, Oracle process thread stuck into ogetblk() which leads to i/o hang. 1587660 (1189432) (DS6000)vxdmpadm disalbe/enable ctlr will hang all VX command 1592617 (1435470) Cluster nodes panicked in voldco_or_pvmbuf_to_pvmbuf code after installing 5.0MP3 1592632 (1463547) Persistent vxconfigd core dump on dynamic LUN reconfiguration. 1592670 (147037) vxconfigd cores at start up 1592688 (494347) CVM:vxconfigd core dumps during slave join 1592703 (973132) vxconfigd sending ioctl to partially open device 1594609 (1471581) vxconfigd may hang if no SCSI timeout value is set while issuing various SCSI commands 1595470 (1224886) SFCFS(0212ag): vxbench proc's hang on all cluster nodes after induced site storage failure. 1603224 (1045980) vxconfigd is looping doing poll() on /dev/null device 1603230 (1272385) Panic in vol_oes_err_bcast() due to NULL volume object pointer dereference 1603289 (148140) CVM: DRL : panic in voldrl_moveto_mru 1603462 (1087073) disk.convert script prints VGs converted list when one or more failed ================================ Incidents fixed in MP2RP6 ================================ 1158574 Other than singleactive, rest of the IO policies allows IOs even when the cpp of dmpnode is NULL in case of nonA/A array 1362441 High kernel CPU usage when DRL is turned on 1458181 New LUN added is not usable if iopolicy is SingleActive. cur_pri_path is null for newly added lun. 1459832 vxclustadm man page needs mcsg instead of hpsg, redundant ifdefs 1466208 Customer requiring fix for e848038 to be included in 4.1MP2RP6 for HP 11.23 1468900 The vxbrk_rootmir script does not complete and is hanging after invoking vxprivutil 1479851 HPUX 11.11: Errors building static vxconfig results in system being unbootable ENVIRONMENT Incidents fixed in MP2RP5 ================================ e514685 VM 4.1 - Invalid Cabinet serial no. returned for DGC ( cx arrays ) when claimed by libvxemc.sl where PowerPath is installed e1134513 vxconfigd hang in transaction commit while acquiring rwsleep write lock e1389596 Bad News Panic when testing 4.1MP2RP4 Kernel Patch 09 PHKL_38428. e1405732 vxpfto settings are not honored on the disk group import e1409527 Manpage for vxdg(1M) needs to cover last shared dg disk detach scenario better ================================ Incidents fixed in MP2RP4 ================================ e1057144 DCO maps get unexpected cleared.leading to data corruption e1086773 vxconfigd semaphore should not be world modifiable e1086980 case 240-588-425 pri [2] false serial split brain detection e1184361 ASL Request HxRT SF-RAC 4.1mp1 HP_UX 11.23 Hitachi DF800 e1227940 VxVM : 311-758-846 : HP : QXCR1000740460 : locker forgot to unlock panic while testing campus cluster. e1254820 what string of VxVM .o modules include meaningless string "@(#)comms:%M% %I%" e1269396 /etc/vx/disk.info permissions are incorrect e1275934 vxbrk_rootmir fails if a volume name = end of another vol name e1280480 vxvm gives same dmpnode for 2 netapp luns e1280481 Allow mixed Naming Schemes, OSN/EBN e1286814 311-536-294 / HP / Several bugs in vxassist relayout option and vxrelayout e1289509 VxVM : 240-755-355 : vxconfigd dumps core in assign_disk_unique_native_name() e1292634 PFTO value is set for only one paths though the DMP node has multiple paths. e1315267 [PRI-3] [281-344-809] [Amadeus] [HPUX 11.23PA CVM 41MP2RP1HF4] incorrect vxdg free output and vxconfigd core dump =============================== Incidents fixed in MP2RP2 =============================== e1180363 vxresize shrink goes ahead even when the volume is in disabled state e1181039 Root mirroring fails for non-system volumes on rootdisk, with Lite/No license e1208034 FMR:DCO volume moves to another disk when volume is grown, does not honour allocation specs. e1210784 Increasing vol_maxio, vol_maxspecialio, and MAXIOSIZE to 1 MB on HP-UX e1213001 'vxdg free' displays wrong information e1214357 Transactions fail after minor no. conflict & temporary remapping of shared vols =============================== Incidents fixed in MP2RP1 =============================== 1122712 1134513 1134516 1139031 1158421 1159216 1170090 =============================== Incidents fixed in MP2 =============================== 1002929 415080 564844 645794 781337 830851 867085 899703 972413 973140 989943 990576 990611 990639 990652 990682 990701 990709 990719 990722 991714 991730 992607 996708 996709 996711 914946 996706 996707 983091 1021717 1023677 1034909 768086 863562 925690 996994 1026855 1038190 1067429 971418 1075637 1079279 1072711 1088800 1093954 1105097 1107132 1108075 1110521 1110528 =============================== Fixes Applied for Products: Volume Manager Install/Uninstall Instructions: 1. Remove all the UnOfficial patches before installing patch. 2. Install this patch using: $ swinstall -x autoreboot=true -s PHCO_41439 PHKL_41440 3. Please do swverify after installing the patches in order to make sure that the patches are installed correctly using: $ swverify PHCO_41439 PHKL_41440 4. Uninstall the patches using: $ swremove -x autoreboot=true PHCO_41439 PHKL_41440 Additional Notes: 1. The Patch 4.1 VxVM MP2RP10 (PHCO_41439 & PHKL_41440) can be installed on VRTSvxvm(4.1.011 or 4.1.010) installed. 2. The Patch PHCO_41439 has COREQUISITE PHCO_40492, the 4.1 11.23 MP2RP8 VMPRO Patch.