PACKAGE: Update Release 3.1.2.21
IOSLEVEL: 3.1.2.21
VIOS level is |
NIM Master level must be equal to or higher than |
Update Release 3.1.2.21 |
AIX 7200-05-02 |
Please refer to the VIOS Maintenance Strategy here for more details regarding the change to the VIOS release numbering scheme.
Be sure to heed all minimum space requirements before installing.
Review the list of fixes included in Update Release 3.1.2.21
To take full advantage of all the functions available in the VIOS, it may be necessary to be at the latest system firmware level. If a system firmware update is necessary, it is recommended that the firmware be updated before you update the VIOS to Update Release 3.1.2.21.
Microcode or system firmware downloads for Power Systems
Update Release 3.1.2.21 updates your VIOS partition to ioslevel 3.1.2.21. To determine if Update Release 3.1.2.21 is already installed, run the following command from the VIOS command line.
$ ioslevel
If Update Release 3.1.2.21 is installed, the command output is 3.1.2.21.
A database manager change has occurred for Shared Storage Pool (SSP). This change will have no direct impacts on behavior, however it does mean that non-disruptive upgrades of an SSP cluster to 3.1.X.X requires that SSP nodes first are updated to version 2.2.6.31 or higher before upgrading to a version of 3.1.0.0 or higher.
Once all VIOS nodes in the cluster have been updated to the latest 2.2.6.X version, double check that rolling upgrade has completed. This can be done by checking the output of “cluster -status -verbose” while logged in as padmin on one of the VIOS nodes in the cluster. Then, check the output for each node, and check for this field:
Node Upgrade Status: 2.2.6.31 ON_LEVEL
If all nodes have 2.2.6.31 or newer, and all say that they are “ON_LEVEL,” then upgrades to the VIOS to 3.1.0.00 or newer can occur without disruption to the SSP cluster.
Additionally, backup and restore can be used to restore older versions of the SSP cluster to 3.1.X.X versions of the VIOS.
Note: 3.1.X.X nodes are unable to join SSP clusters that contain nodes below level 2.2.6.31 or clusters with exclusively 2.2.6.31 or higher nodes prior to the nodes completing a rolling upgrade and saying that they are ON_LEVEL.
Lastly,
customers that do not use Shared Storage Pool will be unaffected by this
change.
This only
applies to customers who both use third party Java based software and have run updateios -remove_outdated_filesets
to remove Java 7 from their system.
To prevent errant behavior when editing customer’s /etc/environment
file, updateios does not make changes to that file
when run. If a customer is using software that depends on using Java and having
the path to it in your PATH environment variable, the following edit should be
made to allow programs that use the PATH environment variable to locate Java 8.
In the /etc/environment file, customers should see:
PATH=[various directories]:/usr/java7_64/jre/bin:/usr/java7_64/bin
To address a potential issue with Java-dependent third party software, this should be converted to:
PATH=[various directories]:/usr/java8_64/jre/bin:/usr/java8_64/bin
Please ensure that your rootvg contains at least 30 GB and that there is at least 4GB free space before you attempt to update to Update Release 3.1.2.21. Run the lsvg rootvg command, and then ensure there is enough free space.
Example:
$ lsvg rootvg
VOLUME GROUP: rootvg VG IDENTIFIER: 00f6004600004c000000014306a3db3d
VG STATE: active PP SIZE: 64 megabyte(s)
VG PERMISSION: read/write TOTAL PPs: 511 (32704 megabytes)
MAX LVs: 256 FREE PPs: 64 (4096 megabytes)
LVs: 14 USED PPs: 447 (28608 megabytes)
OPEN LVs: 12 QUORUM: 2 (Enabled)
TOTAL PVs: 1 VG DESCRIPTORS: 2
STALE PVs: 0 STALE PPs: 0
ACTIVE PVs: 1 AUTO ON: yes
MAX PPs per VG: 32512
MAX PPs per PV: 1016 MAX PVs: 32
LTG size (Dynamic): 256 kilobyte(s) AUTO SYNC: no
HOT SPARE: no BB POLICY: relocatable
PV RESTRICTION: none INFINITE RETRY: no
Virtual I/O Server support for Power Systems
VIOS Update Release 3.1.2.21 may be applied directly to any VIOS that is at either level 3.1.2.00 or above.
The VIOS must be upgraded to version 3.1.2.00 or 3.1.2.10 before Update Release 3.1.2.21 may be applied.
A single-step upgrade can be performed, but it requires that the user does the following:
1. Put the 3.1.2.10 install packages into the desired install directory.
2. Place the 3.1.2.21 install packages in the same directory.
3. Use the updateios command to perform the update.
The VIOS must first be upgraded to 3.1.0.00 before the 3.1.2.21 update can be applied. To learn more about how to do that, please read the information provided here.
Warning: The update may fail if
there is a loaded media repository.
To check for a loaded
media repository, and then unload it, follow these steps.
1.
To check
for loaded images, run the following command:
$ lsvopt
The Media column lists any loaded media.
2.
To unload
media images, run the following commands on all Virtual Target Devices that
have loaded images.
$ unloadopt -vtd
<file-backed_virtual_optical_device >
3.
To verify
that all media are unloaded, run the following command again.
$ lsvopt
The command output should show No
Media for all VTDs.
The Virtual I/O Server
(VIOS) Version 2.2.2.1 or later, supports rolling updates for SSP clusters. The
VIOS can be updated to Update Release 3.1.2.21 using rolling updates.
A non-disruptive
rolling update to VIOS 3.1 requires all SSP nodes to be at VIOS 2.2.6.31 or
later. See detailed instructions in the VIOS 3.1 documentation
The rolling updates
enhancement allows the user to apply Update Release 3.1.2.21 to the VIOS logical
partitions in the cluster individually without causing an outage in the entire
cluster. The updated VIOS logical partitions cannot use the new SSP
capabilities until all VIOS logical partitions in the cluster are updated.
To upgrade the VIOS
logical partitions to use the new SSP capabilities, ensure that the following
conditions are met:
· All VIOS logical
partitions must have VIOS Update Release version 2.2.6.31 or later installed.
·
All VIOS
logical partitions must be running. If any VIOS logical partition in the
cluster is not running, the cluster cannot be upgraded to use the new SSP
capabilities.
1. Run the following
command:
$ cluster -status -verbose
2. Check the Node Upgrade
Status field, and you should see one of the following terms:
UP_LEVEL: This means that the software level of the logical partition is higher
than the software level the cluster is running at.
ON_LEVEL: This means the software level of the logical partition and the
cluster are the same.
There is now a method
to verify the VIOS update files before installation. This process requires
access to openssl by the 'padmin'
User, which can be accomplished by creating a link.
Instructions: Verifying VIOS update
files.
To verify the VIOS
update files, follow these steps:
1. $ oem_setup_env
2.
Create
a link to openssl
# ln -s /usr/bin/openssl /usr/ios/utils/openssl
3. Verify the link to openssl was created
# ls -alL /usr/bin/openssl /usr/ios/utils/openssl
4. Verify that both files
display similar owner and size
5.
# exit
Use one of the following
methods to install the latest VIOS Service Release. As with all maintenance,
you should create a VIOS backup before making changes.
If you are running a
Shared Storage Pool configuration, you must follow the steps in Migrate Shared Storage Pool Configuration.
Note: While running 'updateios' in the following steps, you may see accessauth messages, but these messages can safely be
ignored.
Version Specific
Warning: Version 2.2.2.1, 2.2.2.2, 2.2.2.3, or 2.2.3.1
You must run updateios command twice to get bos.alt_disk_install.boot_images fileset
update problem fixed.
Run the following
command after the step of "$ updateios –accept
–install –dev <directory_name >" completes.
$ updateios –accept –dev <directory_name >
Depending
on the VIOS level, one or more of the LPPs below may be reported as
"Missing Requisites", and they may be ignored.
MISSING
REQUISITES:
X11.loc.fr_FR.base.lib 4.3.0.0 # Base Level Fileset
bos.INed 6.1.6.0 # Base Level Fileset
bos.loc.pc.Ja_JP 6.1.0.0 # Base Level Fileset
bos.loc.utf.EN_US 6.1.0.0 # Base Level Fileset
bos.mls.rte x.x.x.x # Base Level Fileset
bos.svprint.rte x.x.x.x # Base Level Fileset
Warning: If VIOS rules have been deployed.
During update, there have been occasional issues with VIOS Rules files getting
overwritten and/or system settings getting reset to their default values.
To ensure that this
doesn’t affect you, we recommend making a backup of the current rules
file. This file is located here:
/home/padmin/rules/vios_current_rules.xml
First, to capture your current system settings, run this command:
$ rules -o
capture
Then, either copy the
file to a backup location, or save off a list of your current rules:
$ rules -o list >
rules_list.txt
After this is complete,
proceed to update as normal. When your update is complete, check your
current rules and ensure that they still match what is desired. If not,
either overwrite the original rules file with your backup, or proceed to use
the ‘rules -o modify’ and/or ‘rules -o add’ commands to change the rules to
match what is in your backup file.
Finally, if you’ve failed
to back up your rules, and are not sure what the rules should be, you can
deploy the recommended VIOS rules by using the following command:
Then, if you wish to copy
these new VIOS recommended rules to your current rules file, just run:
$ rules -o capture
Note: This will overwrite any
customized rules in the current rules file.
Applying Updates
Warning:
If the target node to
be updated is part of a redundant VIOS pair, the VIOS partner node must be
fully operational before beginning to update the target node.
Note:
For VIOS nodes that are
part of an SSP cluster, the partner node must be shown in 'cluster -status ' output as having a cluster status of
OK and a pool status of OK. If the target node is updated before its VIOS
partner is fully operational, client LPARs may crash.
Instructions: Applying updates to a
VIOS.
To apply updates from a remotely mounted file system, and the remote file
system is to be mounted read-only, follow the steps:
$ shutdown -restart
Note: If shutdown
–restart command failed, run swrole –PAdmin in
order for padmin to set authorization and establish
access to the shutdown command properly.
$ clstartstop -start -n <cluster_name >
-m <hostname >
$ ioslevel
Instructions: Checking for an
incomplete installation caused by a loaded media repository.
After installing an
Update Release, you can use this method to determine if you have encountered
the problem of a loaded media library.
Check the Media
Repository by running this command:
$ lsrep
If the command reports:
"Unable to retrieve repository data due to incomplete repository
structure," then you have likely encountered this problem during the
installation. The media images have not been lost and are still present in the
file system of the virtual media library.
Running the lsvopt command
should show the media images.
Instructions: Recovering from an
incomplete installation caused by a loaded media repository.
To recover from this type
of installation failure, unload any media repository images, and then reinstall
the ios.cli.rte package. Follow these steps:
1.
Unload any
media images
$ unloadopt -vtd
<file-backed_virtual_optical_device>
2.
Reinstall
the ios.cli.rte fileset by
running the following commands.
To escape the
restricted shell:
$ oem_setup_env
To install the failed fileset:
# installp –Or –agX ios.cli.rte –d <device/directory >
To return to the restricted
shell:
# exit
3.
Restart
the VIOS.
$ shutdown –restart
4.
Verify
that the Media Repository is operational by running this command:
$ lsrep
For additional details, including known capabilities, limitations, and additional install considerations, as well as some additional instructions, please reference the readme for 3.1.2.10 located here.
This version will include all fixes found in all previous 3.1.2.X releases. The fixes for the previous release can be found here.
APAR |
Description |
IJ29633 |
/etc/mail/aliases
gets overwritten during update |
IJ30061 |
_P_SIGTIMEDWAIT() NOT
HANLDING MALLOC FAILURE |
IJ22238 |
Install images for openssh.msg.zh_TW |
IJ28934 |
add_managed_system failed for CEC |
IJ28937 |
After migration, some
of the VMs discovery might fail. |
IJ29232 |
A truncated TSIG
response can lead to an assert |
IJ29295 |
SYSTEM CAN CRASH IN
CLOSEX DUE TO INCORRECT FILE DESCRIPTOR FLAG |
IJ29591 |
IN 7200-05 NIMADM
FAILS DURING "INITIALIZING NIM CLIENT" |
IJ29603 |
REDUCEVG FAILS WHEN
PVID OF DISK IN 'REMOVED STATE' IS USED |
IJ29604 |
WITH NO MTA, ALIAS
RESOLUTION USING SUBMIT.CF DOES NOT WORK |
IJ29606 |
LVM reclaim may access
invalid IOCINFO disk information |
IJ29608 |
Wrong ODM entries after
migration to AIX 7.2 |
IJ29609 |
move_login failure on vios could cause crash |
IJ29610 |
CMDB MAY NOT WORK WITH
UMASK 077 FOR ROOT |
IJ29611 |
DR support for DGC
UNITY Storage |
IJ29612 |
Improve ESB migration delay
logic to avoid spurious interrupts |
IJ29615 |
Log PERM error instead
of info error when there is an outstandi |
IJ29618 |
Potential hang during
device removal |
IJ29620 |
Cluster creation may fail
for disk size change on the storage |
IJ29621 |
Unconfig
of non-mpio disks can result in crash. |
IJ29623 |
System get hung during
IPL with ref code Code: 2578 |
IJ29624 |
Added kernel tracing
for disaster recovery |
IJ29629 |
UNABLE TO SEND MAIL
AFTER UPGRADING TO AIX 7.2 TL4 SP2 |
IJ29631 |
SENDMAIL MAY NOT USE
/ETC/MAIL/ALIASES |
IJ29635 |
ROOTVG DISK'S
ALGORITHM CHANGED TO DEFAULT AFTER MKSYSB RESTORE |
IJ29636 |
NIM SAVEVG WITH -R
OPTION FAILED DUE TO WRONG SPACE REQUIREMENT |
IJ29637 |
A TIMING WINDOW CAN
LEAD TO NODE CRASH DUE TO A NULL POINTER |
IJ29638 |
Missing title message
when displaying an error diagmenu 802032 |
IJ29639 |
trcnm
fails when a kernel pathname is specified |
IJ29640 |
AIXIM coredump in zh_CN with some
inputs |
IJ29644 |
iptrace
-T/-B/-Q give wrong PID and unreadable trace file. |
IJ29646 |
Assertion fails
without checking reboot tunable values |
IJ29647 |
Select Input Kuten Panel are not closed by esc key |
IJ29648 |
Coredump
occur when execute xlsfonts -l |
IJ29650 |
iptrace
fails on IPv6-only network interface using BPF |
IJ29651 |
nmon(m)
panel not showing AMS statistics |
IJ29652 |
p11km allows the
overwrite of a certificate file. |
IJ29653 |
p11km object deletion suceeded without selecting one. |
IJ29655 |
Improve trace format
for undispatch hook 4B0 |
IJ29656 |
Modifying a type in alt_rootvg_op command |
IJ29657 |
coredump
by isakmpd at getGSSIdentityName |
IJ29658 |
isakmpd
hits assert in isakmp_buf::~isakmp_buf |
IJ29659 |
core dump by ikev2d in
ParsePayload |
IJ29660 |
Unable to create
cluster with 1023 pool |
IJ29665 |
nimadm
does not set BLV size for 7.1 and 7.2 |
IJ29666 |
Handle chdev of udid attr for SEA |
IJ29671 |
system crash @IPRA.$entcore_dealloc_args+000354 |
IJ29672 |
BCS PCI reset while
device cfg/uncfg can
cause device to go dead |
IJ29746 |
dumpfs
displays incorrect dmap information on large fs |
IJ29799 |
FAIL TO ESTABLISH LACP
AGREEMENT AFTER SWITCH REBOOT |
IJ29817 |
vfcmap
fails with "does not appear to be a valid FC port" |
IJ29871 |
POOL START MAY FAIL ON
VIOS LEVEL3.1.1.X,WHEN USING VIPA FOR S |
IJ29872 |
NIM SAVEVG -R FLAG MAY
NOT GETTING PASSED TO CLIENT |
IJ29874 |
system hung due to IPSEC
SA idle timer handler lock contention |
IJ29875 |
acfstat
command doesn't return usage error without hyphen |
IJ29876 |
Correctly handle
target structure when Move Login fails |
IJ29877 |
A potential security
issue exists |
IJ29878 |
DU COMMAND REPORTS
SIZES INCORRECTLY FOR UDF FILESYSTEMS |
IJ29879 |
Code that locks bmaps may cause operations to fail. |
IJ29896 |
PAM_AUTH does not
display messages about LDAP password status |
IJ29903 |
Client may hang on a small
icrease of num_cmd_elems
attribute |
IJ29904 |
Migration may not
complete, stuck in collecting minisnap data |
IJ29905 |
IPTRACE DOES NOT
CAPTURE ARP WITH FILTERS APPLIED |
IJ29906 |
chcluster
fails adding backup disks with diff name in same site |
IJ29907 |
mksysb
migration may fail with extendlv error of /usr |
IJ29908 |
After LPM on 3.1.2 for
npiv, there may be a slight slowness |
IJ29958 |
LDAPAFILES FAILS IF
USERS ARE MISSING LDAP ATTRS |
IJ30045 |
migration install may fail
to create boot image |
IJ30046 |
ahafs
device name change to /aha post live update |
IJ30090 |
Tape backup or restore
may fail or be slow with VIOS 3.1.2 |
IJ30120 |
ISDS ADVANCED PASSWORD
POLICY ERRORS NOT SHOWN |
IJ30124 |
MEMORY LEAKS IN
GETUSERATTRS AND LOGINRESTRICTIONSX |
IJ30146 |
NIM AT 7200-05 NIMADM
MAY FAIL WITH KSH: C_CH_NFSEXP: NOT FOUND |
IJ30153 |
send data transferred infor to protocol driver |
IJ30197 |
PERFSTAT_DISK_TOTAL
CAN LEAK MEMORY |
IJ30198 |
Merged output for
"lparstat -mpw"
due to large values |
IJ30199 |
sockets iocp do not work well with live update |
IJ30203 |
Lun-level validation
should copy correct vfcFrame data out |
IJ30204 |
LPM of IBMi client with double mapped adapters could fail |
IJ30205 |
HSmon
status is inoperative on few VIOS |
IJ30214 |
User OFED test case
crashed AIX partition |
IJ30225 |
NIMADM FAILS WITH:
0042-130 C_CH_NFSEXP: "NFS_VERS" CANNOT BE... |
IJ30283 |
rmdev
hang on client vNIC device |
IJ30284 |
REMOVE THE VNIC_LOG AS
IT CAN CAUSE LOWLEVEL KDB TO CRASH |
IJ30285 |
SU USES PAM EVEN IF
AIX.SECURITY.SU AUTHORISATION APPLIES |
IJ30286 |
Error seen while
adding tsd entry for libraries |
IJ30435 |
Improve stuck-in-ENXIO
state handling |
IJ30438 |
Trace timestamp for
I/O INTERRUPT FLIH Hook D2 parameter |
IJ30443 |
Options missing from
Ethernet adapter SMIT menus |
IJ30444 |
MKVDEV WITH
MIRRORED=TRUE CAN FAIL IN VIOS 3.1 |
IJ30445 |
Allow varyoffvg operation when all local disks are unavailable |
IJ30446 |
A potential security
issue exists |
IJ30447 |
A potential security
issue exists |
IJ30452 |
LKU TUNABLES TRANSFER
FAILURE DUE TO MISSING BOS.NET.NIS.CLIEN |
IJ30454 |
assembler does not
handle ptesync or xxspltd
correctly |
IJ30499 |
VKE LOGS EALREADY WHEN
CAA RETURNS EMPTY IP LIST WITH VIPA |
IJ30500 |
Junk chars after rmdev cmd during migration
abort case |
IJ30592 |
INCORRECT ODM
ATTRIBUTE DESCRIPTIONS FOR SOME INTERFACE TYPES |
IJ30593 |
ksys_hsmon daemon is not up, in VMRM HA solution. |
IJ30595 |
Remove VIOS by Ksys should fail on fail to connect to database. |
IJ30596 |
PAM LOGS INCORRECT
SYSLOG ERROR MESSAGE WHEN RUNNING SU. |
IJ30627 |
nim savevg
-r with size preview should not do savevg &
just exit |
IJ30628 |
WHEN /AUDIT IS FULL,
AUDITSELECT RETURNS THE WRONG RETURN CODE |
IJ30676 |
LSPV COREDUMP WITH
MALLOCDEBUG DUE TO A DOUBLE FREE IN LIBSM.A |
IJ30687 |
VFC4_ERR19 ERROR
LOGGED ON NPIV ADAPTERS AFTER UPGRADE TO 7.2.5 |
IJ30705 |
LPM of clients using
NPIV may fail under rare circumstances. |
IJ30715 |
System crash caused by
nfs4 remount. |
IJ30718 |
Device FFC missing
from Diagnostics menu number |
IJ30748 |
vfc_host
need to set SUPPRESS_ABTS bit in login resp |
IJ30765 |
ssp_postgres_backup fails if hostname aliases are defined |
IJ30808 |
Multithreaded
applications core dump or hang |
IJ30841 |
Memory leak fix in
ikev2d |
IJ30845 |
FCA_ERR2 WITH SENSE
DATA 0DB8 AFTER FABRIC EVENT |
IJ30846 |
preserving host_nqn after change on cfgmgr |
IJ30847 |
Send process_login to the correct channel |
IJ30848 |
Conversion robustness
enhancements |
IJ31181 |
aynchronous events when the switch is pulled |
IJ31191 |
PATH FAILURES AFTER
NPIV CLIENT UPDATE TO AIX 7.2 TL5 SP1 |
IJ31195 |
suma_swinv does not function due to syntax error |
IJ31198 |
Updating timezone data files to latest available for 2021 |
IJ31231 |
Potential crash when
flashing FW that is unoperational afterward |
IJ31596 |
pwd behavioural
change moving from SP3 to SP7 with a |
IJ31681 |
dsmserv
hang in on AIX 7.2 TL5 level |
IJ31807 |
AIX crash due to
memory corruption |
IJ31810 |
A potential security
issue exists |
IJ31845 |
Variables may be
typeset outside of install |
IJ31881 |
NTP fileset migration may change links to NTP3 |