APAR: IY14660 COMPID: 5765D6600 REL: 201
ABSTRACT: MEMORY CORRUPTION ON AIX USING DB2 VIA TCP/IP WITH NIS ENABLED

PROBLEM DESCRIPTION:
A memory corruption problem has been observed during testing on
AIX. This may result in brokers crashing. This will only occur
on AIX using DB2 connected via TCP/IP, with NIS enabled on the
system, and using PubSub message flows with temporary dynamic
queues.

LOCAL FIX:
The temporary workarounds are either to disabled NIS or to avoid
using temporary dynamic queues.

------

APAR: IY15158 COMPID: 5765C3407 REL: 210
ABSTRACT: V2R1 PTF-10

PROBLEM DESCRIPTION:
Added filesets for AIX v2r1 NSM PTF10

------

APAR: IY15269 COMPID: 5765D6600 REL: 201
ABSTRACT: CUMULATIVE MAINT #02 FOR MQSERIES INTEGRATOR FOR AIX V201

PROBLEM DESCRIPTION:
CUMULATIVE MAINT #02 FOR MQSERIES INTEGRATOR FOR AIX V201

PROBLEM SUMMARY:
Cumulative maintenance #2 for MQSeries Integrator for AIX V2.0.1

PROBLEM CONCLUSION:
Cumulative maintenance #02 contains fixes to the following
problems:-
IC27301 - SETTING MQMD.EXPIRY FIELD TO AN ABSOLUTE VALUE IN COMP
IC28312 - FLOAT DATA TYPES OF LENGTH 8 ARE NOT HANDLED CORRECTLY
IC28324 - COBOL COPYBOOK IMPORT PROBLEM, MESSAGE BIP1812S RECEIV
IC28443 - BIP1812S RECEIVED USING THE CONTROL CENTER GENERATE FU
IC28446 - INSTALL GUIDE DOES NOT DISCUSS ABOUT MQSI_PARAMETERS_F
IC28447 - INSTALL GUIDE PAGE 129/130 HAVE INCORRECT NAMES FOR NT
IC28518 - 1. CREATING A BITSTREAM THAT CONTAINS A REPEATING SEGM
IC28520 - LOOP ON COMPUTE NODE - REPEATING ELEMENT USES A REPEAT
IC28605 - MESSAGE FLOW LOOP WITH MSGBIP2608I MSGBIP2623I AND MSG
IC28724 - IMPORT OF COBOL COPY BOOK THAT DOES NOT HAVE A COMPLET
IC28810 - SUBSCRIPTIONS WON T SHOW UP FOR A SINGLE USERID
IC28926 - SIGNED PACKED DECIMALS WILL ACCEPT F AS WELL AS C
IC29022 - BIP1812E CAUSED DURING A DEPLOY IN STRUCTURES WITH MAN
IC29033 - MQSI V201 USING DUPLICATE FLOW FEATURE TO CREATE MESSA
IC29184 - XML-PARSE ENVIRONMENT INCORRECT
IC29200 - XML MESSAGE GETTING TRUNCATED AFTER PROCESSING THROUGH
IY14540 - AIX BROKER CONFIGURATION LIBPATH MODIFICATION MISSING
IY14660 - MEMORY CORRUPTION ON AIX USING DB2 VIA TCP/IP WITH NIS
IY14720 - MSGBIP2301E WHEN A BROKER TRIES TO LOAD THE MQSI SAMPL
IY15310 - MEMORY LEAK AND THROUGHPUT DEGREDATION
IY16620 - MQSI CORE DUMP CAUSED BY ADDRESS ALIGHNMENT ISSUE WHEN
12057 - Access Violation in MtiImbParser.lil
13319.1 - Node properties dialog change size everytime it opens
13440 - STRESS: Repeated BIP2132 on mixed stress test
13440.1 - ImbPubSubThreads redesign
13590 - ibm-948 missing two chars
13643 - Scrolling and expand tree over than 1 second with 400
13643.1 - CC opening can be slowed due to caches initialisation
13650 - storage leak in compute with CWF format
13767 - Memory leaks and Performance
14043 - Memory leak in WritePackedDecimal
14150 - Leakage XML OUT for tags at Structure Start (Perhaps)
14573 - The Parent name is wrong for CWF error messages
15311 - DB Insert into Oracle 8.1.6 can cause DFE core dump

------

APAR: IY15310 COMPID: 5765D6600 REL: 201
ABSTRACT: MEMORY LEAK AND THROUGHPUT DEGREDATION

PROBLEM DESCRIPTION:
Memory leaks when reading the end field of a CWF structure.
Performance degradation when loading and unloading message sets.

PROBLEM SUMMARY:
Memory leaks when the end field of a CWF structure is read.
There is also a performance issue with the parser having to
load a dictionary every time it does a parse

PROBLEM CONCLUSION:
Fixed the following:
1. Fixed Len Strings not getting freed up if the string is at
   the end of a message.
2. Memory getting allocated using new and then deallocated
   using free and vice versa.
3. Memory leak in writed Packed Decimal

------

APAR: IY17178 COMPID: 5765C3403 REL: 430
ABSTRACT: ISI IN HD PIN

PROBLEM DESCRIPTION:
When running a concurrent reducevg or syncvg there is
a window where we could isi within hd_pin (the top half
of the lvm device driver).

PROBLEM SUMMARY:
We now briefly pin the top half while we
go down this code path. That way we
won't page fault so we won't isi.

PROBLEM CONCLUSION:
We now briefly pin the top half while we
go down this code path. That way we
won't page fault so we won't isi.

------

APAR: IY17603 COMPID: 5765B9501 REL: 330
ABSTRACT: USESINGLENODE QUORUM SYNTAX ERROR

PROBLEM DESCRIPTION:
usesinglenode quorum syntax error

PROBLEM SUMMARY:
Correct logic error found by development

------

APAR: IY17608 COMPID: 5765B9501 REL: 330
ABSTRACT: ASSERT FAILED: IN_CPY->COPYSET...

PROBLEM DESCRIPTION:
assert failed:in_cpy->...

PROBLEM SUMMARY:
GPFS self check logic failed in HandleReq.C

PROBLEM CONCLUSION:
Correct logic error in the token manager

------

APAR: IY17610 COMPID: 5765B9501 REL: 330
ABSTRACT: BAD ALLOCMGR SERVER STATE AFTER INIT FAILURE

PROBLEM DESCRIPTION:
bad allocmgr server state after init failure

PROBLEM SUMMARY:
Correct logic error discovered in development

------

APAR: IY17612 COMPID: 5765B9501 REL: 330
ABSTRACT: NEED DOCUMENTATION ADDED REGARDING REINSTALL OF A GPFS SERVER

PROBLEM DESCRIPTION:
This APAR was opened with regard to the steps required following
reinstall of a GPFS server since the image doesn't backup the
/dev/ entries for mountpoints. You can either mv cluster.nodes
.nodes or mmfs.cfg and restart GPFS to rebuild the /dev entries,
but the documentation doesn't say anything about that.
Documentations should be modified/enhanced to reflect above
steps.

LOCAL FIX:
Modify/enhance the documentation to show the steps required
following the reinstall of a GPFS server.

PROBLEM SUMMARY:
Using network install of a node, the gpfs configuration
files are refreshed; but the entries in the aix
configuration files are not. This results in an inability to
mount file systems until the AIX /dev and /etc/filesystems
entries are refreshed.

PROBLEM CONCLUSION:
On GPFS startup; check to see that the AIX configuration
files contain the needed data.

------

APAR: IY17677 COMPID: 5765B9501 REL: 330
ABSTRACT: MMLSQUOTA -G <GROUP> RUN AS NONE ROOT USER RETURNS THE ERROR

PROBLEM DESCRIPTION:
mmlsquota -g <group> run as none root user returns the error
"operation not permitted" (in this case user loadl).
GPFS documentation has no limits listed on mmlsquota like 'to
run this you need system group permission'.

LOCAL FIX:
The code internally checks the uid of the process issuing the
command and allows non-root users to see only their own user
quotas. This restriction should be mentionded in the man page
in the next level of documents

PROBLEM SUMMARY:
mmlsquota requires root access

PROBLEM CONCLUSION:
Allow mmlsquota to display the quotas for a group which the
issuing user is a member of.

------

APAR: IY17720 COMPID: 5765B9501 REL: 330
ABSTRACT: MTIME CHANGE NOT VISIBLE AFTER OPEN WITH CREATE FLAG

PROBLEM DESCRIPTION:
mtime change not visible after open with create flag

PROBLEM SUMMARY:
mtime not correctly set in certain timing situations

PROBLEM CONCLUSION:
get stronger lock on updating mtime.

------

APAR: IY17725 COMPID: 5765B9501 REL: 330
ABSTRACT: GETMSG LOSES ERRNO SETTING

PROBLEM DESCRIPTION:
getmsg loses errno setting

PROBLEM SUMMARY:
Errno gets lost if the message file becomes inaccessible

PROBLEM CONCLUSION:
Correct handling of errno,

------

APAR: IY17727 COMPID: 5765B9501 REL: 330
ABSTRACT: LOGS NOT MIGRATED IN DELDISK

PROBLEM DESCRIPTION:
logs not migrated in deldisk

PROBLEM SUMMARY:
Log migration failed after deleting all of the original
disks in a file system and then trying immediately to
restripe the file system

PROBLEM CONCLUSION:
Correct an error in the creation of spare logs.

------

APAR: IY17728 COMPID: 5765B9501 REL: 330
ABSTRACT: LX:GPFS DEADLOCK DUE TO THE HEADER HANDLER

PROBLEM DESCRIPTION:
lx:gpfs deadlock due to the header handler

PROBLEM SUMMARY:
Deadlock discovered in development

PROBLEM CONCLUSION:
Correct locking error

------

APAR: IY17730 COMPID: 5765B9501 REL: 330
ABSTRACT: DISK IO SUMMARY STATS, CLOSE ERROR

PROBLEM DESCRIPTION:
disk io summary stats, close error

PROBLEM SUMMARY:
add additional information to service commands

------

APAR: IY17731 COMPID: 5765B9501 REL: 330
ABSTRACT: ASSERT: SUBBLOCKS== 0!!SUBBLOCKS==OFP->GETSUBBLOCKSPERFILEBLOCK

PROBLEM DESCRIPTION:
assert: subblocks==0!!subblocks==ofp->getsubblocksperfileblock

PROBLEM SUMMARY:
GPFS self check logic terminated GPFS when running
applications using datashipping

PROBLEM CONCLUSION:
Corrected logic involving datashipping buffer management

------

APAR: IY17786 COMPID: 5765C3403 REL: 430
ABSTRACT: IOCP CAN DEADLOCK

PROBLEM DESCRIPTION:
The iocp code calls xmattach while holding the socket
lock. xmattach can go to sleep which implies that a sleep
can occur while holding the lock. This releases the cpu
for other tasks. If the same cpu is called to service
the socket due to an interrupt, the interrupt context will
spin waiting for the lock to be released which will never
happen since the context that has it is on the same CPU.
Plus, xmattach should not be called while disabled since
it can page fault.

PROBLEM CONCLUSION:
The solution is to release the lock, do the xmattach, regain
the lock, and then check to see if the request can be
processed. If it can, go ahead and process the request
immediately. Otherwise, enqueue the request to be processed
when the data is received.

------

APAR: IY17817 COMPID: 5765C3403 REL: 430
ABSTRACT: GETATTR SLOW DURING HEAVY NFSV3 WRITE LOAD

PROBLEM DESCRIPTION:
If there is heavy NFSv3 write load (lots of pending I/O),
getattr calls for that file are delayed until all outstanding
data has been written to the server.
For example, if a "dd" from /dev/zero to a file on an NFSv3
mounted filesystem is started (with, for example bs=64k), an
"ls -l" of that file will not return until the "dd" has
finished, since new written data fills up faster than it is
written to the NFS server.

LOCAL FIX:
NFSv2 doesn't have this problem, since it doesn't use
asyncronous writes.

PROBLEM SUMMARY:
If dirty pages exist for a file when getattr is called (and
the modified time is to be read), getattr will sync the file
to flush all changes to the server. If the writing process
fills up with more dirty pages faster than they can be flushed,
it might take a long time for getattr to return.

PROBLEM CONCLUSION:
Add a mount option (fastattr) that can be specified if the
above behaviour is not desirable. Since getattr might not
return the same on the server and the client due to
outstanding dirty pages to be written, and thus changes the
behaviour of getattr, the option is NOT enabled by default.

------

APAR: IY17858 COMPID: 5765C3403 REL: 430
ABSTRACT: RETURN CODE FROM AIXTERM -E <COMMAND> SOMETIMES INCORRECT

PROBLEM DESCRIPTION:
When you issue aixterm -e command the return code should be
the return code from command. However, sometimes these return
codes are not correct

PROBLEM CONCLUSION:
A call to signal(SIGCLD, SIG_IGN) will prevent a terminated
child process from becoming a zombie. So a call of waitpid
by the parent could not get the exit status of this child.
Since we use signal(SIGCLD, xgetsig) to handle SIGCLD signal,
we'll delete all calls that try to ignore SIGCLD

------

APAR: IY17871 COMPID: 5765C3403 REL: 430
ABSTRACT: SYSTEM DSI CRASH IN GXENTDD

PROBLEM DESCRIPTION:
System DSI crash in gxentdd. Stack trace appears
as follows:
MST STACK TRACE:
(excpt=0000000a:0a000000:00000000:0000000a:00000106)
(intpri=3)
IAR: . gxentdd:undo_rx +f8 (051bb0a0): sthx r0,r5,r4
LR: . gxentdd:start_adapter +550 (051b8ea0)
00436bc0: . gxentdd:start_adapter +550 (051b8ea0)
00436c30: . gxentdd:reset_timeout +15c (051becb4)
00436c90: .watchdog+d4 (0008e274)
00436cf0: .sys_timer+474 (0002def4)
00436d80: .clock+134 (0002e448)
00436de0: .i_softmod+2a8 (0001c3b0)
00436e70: flih_603_patch+cc (0002863c)

------

APAR: IY17882 COMPID: 5765C3403 REL: 430
ABSTRACT: SYNCLVODM RETURNS ERROR WHEN VGID IS PASSED INSTEAD OF VGNAME

PROBLEM DESCRIPTION:
config operation will fail on passive node if it involves
invocation of synclvodm and passive node will get varied off.

PROBLEM CONCLUSION:
vgid length has changed from 16 chars to 32 chars. So add
additional check in check_desc function.

------

APAR: IY17983 COMPID: 5765C3403 REL: 430
ABSTRACT: LSDEV -C SHOWED DESCRIPTION N/A FOR GXT130P

PROBLEM DESCRIPTION:
The adapter description index was missing from the ODM PdDv
for GXT130P.

PROBLEM CONCLUSION:
Add the message index for the adapter description to ODM.

------

APAR: IY18009 COMPID: 5765B9501 REL: 330
ABSTRACT: GPFS QUOTA OPERATION LOSES TOKEN AND ASSERTS

PROBLEM DESCRIPTION:
The problem seems to be that the filesystem managerwanted to
do a quota operation, but had lost the token for the quota file
to a particular node. It seems the quota code forgot about
reacquiring the token before doing the operation and therefore
asserted when some of its data was not in a "valid" state.

PROBLEM SUMMARY:
GPFS self check logic failed in a stress load with quotas
enabled

PROBLEM CONCLUSION:
Correct locking error on quota file

------

APAR: IY18024 COMPID: 5765B9501 REL: 330
ABSTRACT: INDIRECT BLOCKS LEFT AFTER TRUNCATING TO SMALL FILE

PROBLEM DESCRIPTION:
indirect blocks left after truncating to small file

PROBLEM SUMMARY:
Correct an error in maintaining the indirect level of a file
when truncated from a very large file to a very small
non-zero length

PROBLEM CONCLUSION:
Correctly handle the indirection level when truncating files
to a non-zero small size.

------

APAR: IY18026 COMPID: 5765B9501 REL: 330
ABSTRACT: DEAMON ASSET REPDISKADDR::GETFROMARRAY + 0X9C

PROBLEM DESCRIPTION:
daemon asset repdiskaddr::getfromarray + 0x9c

PROBLEM SUMMARY:
Service tool caused a node panic when used.

PROBLEM CONCLUSION:
Correct a logic in a data collection service tool

------

APAR: IY18027 COMPID: 5765B9501 REL: 330
ABSTRACT: FORMAT SEGMENTATION FAULT

PROBLEM DESCRIPTION:
format segmentation fault

PROBLEM SUMMARY:
Fix potential segmentation fault during file system
creation.

PROBLEM CONCLUSION:
Fix bug in serializing multiple worker threads

------

APAR: IY18091 COMPID: 5765C3403 REL: 433
ABSTRACT: DATA STORAGE INTERRUPT - PROC ON ANN

PROBLEM DESCRIPTION:
 02333A6C eMpcProcessReceivedTaggedDataFrame+00051C
          (31BEB000, 703A0080, 70236700)
 023302F8 mpc_receive+0002D4
          (31A33300, 70236700, 0000000E, 703A0080)
 021F3864 atm_dmx_receive+000228 ()
 021ADBF4 chatm_cut_pkt_indi+000788 ()
 021B0B54 chatm_rv_indi_intr+000168 ()
 021C4600 chatm_process_rxque+000070 ()
 021C9ED8 chatm_process_intr+003094
          (31A33000, 02220A08, 48E6C000)
 021CC188 chatm_intr+000A98 ()
 0001C25C i_poll_soft+0000B8 ()
 0001BEC8 i_softmod+0000F8 ()
 0002550C flih_603_patch+0000B4 ()
 00025468 flih_603_patch+000010 ()
Exception (2FF3B400)

PROBLEM CONCLUSION:
remove extraneous trace point

------

APAR: IY18103 COMPID: 5765C3403 REL: 430
ABSTRACT: FLOATING POINT DA DOES NOT INVOKE BINDPROCESSOR

PROBLEM DESCRIPTION:
Processor diagnostics, invoked by diag -ceAd proc<x>
on a multiple processor system is reporting a failure on
different processors, although onlyone processor is
defective.

PROBLEM CONCLUSION:
The invocation of diagnostics with the "e" flag should not
be testing at all, only doing ELA. However, the tests are
run without first doing the bindprocessor, resulting in the
test process executing on any processor. The solution is to
correctly not run the test in ELA mode. As long as ELA mode
is not specified, processor diags on a mp system correctly
binds the processor before testing.

------

APAR: IY18164 COMPID: 5765B9500 REL: 140
ABSTRACT: MMADDCLUSTER CMD HANGS WHEN PASSED INVALID PARAMETER

PROBLEM DESCRIPTION:
mmaddcluster cmd hangs when passed invalid parameter

PROBLEM SUMMARY:
mmcrcluster hangs when passed a node which exists (can be
pinged); but is not part of the HACMP cluster

PROBLEM CONCLUSION:
Correct a logic error in mmcrcluster

------

APAR: IY18166 COMPID: 5765B9501 REL: 330
ABSTRACT: GPFS 1.5 SP /ASSERT IN SYNC.C DIRTYINDBUFS > 0 && IBDP != NULL

PROBLEM DESCRIPTION:
gpfs 1.5 sp /assert in sync.C dirtyindbufs >0 && ibdp != null

PROBLEM SUMMARY:
GPFS self check logic failed in sync.C line 2683

PROBLEM CONCLUSION:
 In writeIndirect, wait until updateLogger mutex is held
before checking whether indirect block is dirty.

------

APAR: IY18169 COMPID: 5765B9501 REL: 330
ABSTRACT: ASSERT FAILED HANDLEREQ.C LINE 2598

PROBLEM DESCRIPTION:
assert failed handlereq.c line 2598

PROBLEM SUMMARY:
GPFS self check logic failed at
HandleReq.C line 2598

PROBLEM CONCLUSION:
Token reclaiming flag needs to be turned off when the
token is put backin STABLE state.

------

APAR: IY18171 COMPID: 5765B9501 REL: 330
ABSTRACT: REMOVE ID AND LOG RECORDS FROM SHIPPED FILES

PROBLEM DESCRIPTION:
remove id and log records from shipped files

PROBLEM SUMMARY:
Minor packaging changes

------

APAR: IY18173 COMPID: 5765B9501 REL: 330
ABSTRACT: IMPROVE PRIMARYSERVER ERROR REPORTING

PROBLEM DESCRIPTION:
improve primaryserver error reporting

PROBLEM SUMMARY:
Minor message cleanup

------

APAR: IY18181 COMPID: 5765C3403 REL: 430
ABSTRACT: MIGRATEPV FAILING WITH MULTIPLE DESTINATION PVS

PROBLEM DESCRIPTION:
The migratepv command is returning Illegal parameter error in
certain cases with multiple destination PVs.
migratepv -l <lvname> hdisk0 hdisk1 hdisk2
 0516-022 lmigratelv: Illegal parameter or structure value.
 0516-812 migratepv: Warning, migratepv did not completely
 succeed.
The following conditions apply, for this error to occurr: 1.
There are multiple destination PVs(hdisk1 hdisk2 ) 2. hdisk1
should have some USED PPs. 3. The USED PPs distribution on
hdisk1 and hdisk2 should
    be complimentary.(corresponding pp nos which are used in
    hdisk1 should be free on hdisk2).

PROBLEM SUMMARY:
migratepv fails with error , as lvm_migratelv()
returns illegal parameter or value.
when there are multiple destination PVs

PROBLEM CONCLUSION:
 Ensure that when ever the newpv_id(dest PV) inside the loop
 changes, the routine get_pvandpp() is called for that PVid.
 Now the savepp2 (which is used to calculate the individual PP
 address) will contain the starting PP address of the changed
 dest PV.(Earlier it contains the starting PP adress of first
 destination PV regardless of the number of destination PVs).

------

APAR: IY18263 COMPID: 5765C3403 REL: 430
ABSTRACT: VARYON MWCC TSTAMP WRITE NOT NEEDED

PROBLEM DESCRIPTION:
In a shared disk environment where nodeA
has read/write access and nodeB has read
only access to the disk, if we do an
extendvg and then 'importvg -L' to teach
nodeB about the change, the 'importvg -L'
might fail.
Varyon will try to update the on disk mwcc
timestamp for the newly added disk. nodeB
has read only access to the disk so the write
would fail and therefore so would the varyon.

PROBLEM CONCLUSION:
Don't update the on disk mwcc timestamp
at varyon time. We do update the on disk mwcc
timestamp at other times which are more
appropriate, and doing it at varyon time is
redundant.

------

APAR: IY18267 COMPID: 5765C3403 REL: 430
ABSTRACT: CTOK_MEM_ERR MCA TOKENRING WON'T CONFIGURE

PROBLEM DESCRIPTION:
CTOK_MEM_ERR errors in error report. Adapter
remains in the defined state.

LOCAL FIX:
CTOK_MEM_ERR errors in error report. Adapter
remains in the defined state.

PROBLEM SUMMARY:
CTOK_MEM_ERR in error report. MCA TokenRing adapter
will not configure.

------

APAR: IY18268 COMPID: 5765C3403 REL: 430
ABSTRACT: MCA FDDI ADAPTER CRASHES DURING CONFIGURATION

PROBLEM DESCRIPTION:
MST STACK TRACE:
0x2ff3b400 (excpt=d0026000:40000000:00008811:
d0026000:00000106) (intpri=0)
IAR: .xmemdma_pwr+518 (000feff0): teq r14,r14
LR: .xmemdma_pwr+184 (000fec5c)
2ff3b160: . mca_pwr_busdd:d_master_pwr +1f8 (0105b828)
2ff3b1f0: . fddidd:fddi_config +5bc (0119af5c)
2ff3b2e0: .config_kmod+f4 (0015fdf8)
2ff3b370: .sysconfig+144 (001601ac)
2ff3b3c0: .sys_call_ret+0 (00003a4c)

PROBLEM SUMMARY:
MCA FDDI adapter crashes during configuration.
MST STACK TRACE:
0x2ff3b400 (excpt=d0026000:40000000:00008811:
d0026000:00000106) (intpri=0)
IAR: .xmemdma_pwr+518 (000feff0): teq r14,r14
LR: .xmemdma_pwr+184 (000fec5c)
2ff3b160: . mca_pwr_busdd:d_master_pwr +1f8 (0105b828)
2ff3b1f0: . fddidd:fddi_config +5bc (0119af5c)
2ff3b2e0: .config_kmod+f4 (0015fdf8)
2ff3b370: .sysconfig+144 (001601ac)
2ff3b3c0: .sys_call_ret+0 (00003a4c)

------

APAR: IY18270 COMPID: 5765C3403 REL: 430
ABSTRACT: SETGID DIRECTORIES NOT WORKING CORRECTLY OVER NFS

PROBLEM DESCRIPTION:
APAR IY16105 introduced a too stringent limitation for setgid
directories over NFS. With that APAR installed, the user must
be member of the group of which the created files are to be
owned by. This not a valid requirement, causing problems in
some cases.

LOCAL FIX:
NFS version 2 does not show these problems.

PROBLEM SUMMARY:
***************************************************************
*USERS AFFECTED: *
* All users with the following filesets at these levels *
* bos.net.nfs.client 4.3.3.30irement, causing problems in
* bos.net.nfs.client 4.3.3.31
* bos.net.nfs.client 4.3.3.50
***************************************************************
*PROBLEM DESCRIPTION: *
* APAR IY16105 introduced a too stringent limitation for
* setgid directories over NFS. With that APAR installed,
* the user must be member of the group of which the
* created files are to be owned by. This not a valid
* requirement, causing problems in some cases.
***************************************************************
*RECOMMENDATION: *
* Apply apar IY18270
***************************************************************

PROBLEM CONCLUSION:
Remove the check of group membership in the nfs(3)_create
functions that was introduced in 324998

------

APAR: IY18286 COMPID: 5765C3403 REL: 433
ABSTRACT: CPSD COREDUMPS WHEN AUTH METHOD IS RSA SIGNATURES IN IKE

PROBLEM DESCRIPTION:
/usr/sbin/cpsd coredumps when trying to activate an Phase 1
IKE tunnel using RSA signatures as the authentication
method.

PROBLEM CONCLUSION:
Modified the compareIDs routine to use GeneralNames object
and checked the return code of some x509 parser calls.

------

APAR: IY18294 COMPID: 5765C3403 REL: 430
ABSTRACT: LDAP CLIENT DAEMON HANGS IF /ETC/GROUP LINE IS TOO BIG

PROBLEM DESCRIPTION:
The LDAP client daemon secldapclntd hangs when
commands like "lsuser -R LDAP <uname>" or
"lsgroup -R LDAP <groupname>" is entered and the
user <uname> belongs to a group which has a large
number of users (app 1500 users in /etc/group from
which ldap database is created). The lsgroup command
also hangs trying to get information for such a group.
If the daemon is killed, the comands give partial output.

PROBLEM SUMMARY:
The LDAP client daemon secldapclntd hangs when
/etc/group contains large number of users. All the
associated commands like "lsuser -R LDAP" and
"lsgroup -R LDAP" also hang due to this.

PROBLEM CONCLUSION:
The memory allocation required for such large
number of users is insufficient which is now
fixed.

------

APAR: IY18297 COMPID: 5765C3403 REL: 430
ABSTRACT: BINDER COREDUMPS WITH KAI COMPILER

PROBLEM DESCRIPTION:
Customer is facing the problem while compiling and linking.
While linking the application,binder core dumps with SEGV. The
problem only shows up when you use rtl, lapi, and
noobjreorder (but that isn't an unusual combination) with a
KCC compiled routine and one of the KCC libraries (but,
ofcourse, KAI claims that this is a binder problem and not a
problem with their stuff.) The load line needed to show the
problem is as follows: ld /lib/mcrt0.o -bnoobjreorder tttt.o
-L/usr/lpp/ppe.poe/lib -binitfini:poe_remote_main -lmpi_r
-llapi_ above line includes an "r" to finish -llapi_r and the
last line follows libKCC-eh-ts_so.a -brtl -lc I have given you
on snow a tar file with the above information, the tttt.o file,
and the libKCC-eh-ts_so.a library (which I got permission from
KAI to give to IBM for purposes of finding this problem.)

PROBLEM SUMMARY:
The binder coredumps in some cases when -brtl, -bnoobjreorder
are used

PROBLEM CONCLUSION:
 Incrementing the export counter for bss symbol will resolve.

------

APAR: IY18328 COMPID: 5765C3403 REL: 430
ABSTRACT: NAMED4 VERSION SYNTAX UPDATE

PROBLEM DESCRIPTION:
Issuing "named4 -d 1" produces output in (/usr/tmp/named.run).
Prior to the apply of DNS Security PTF U474686
(APARS IY16182, IY16214), named.run shows date X. After the
apply of PTF U474686, and after the re-run of AIX Command
"named4 -d 1", named.run still shows date/time X instead of
date/time X+1, or X+2, etc.
Prior to the apply of PTF U474686, issueing the AIX Command
"what named4" produces bos.net.tcp.server 4.3.3.28 showing
an older timestamp/date of X. After the Apply of PTF U474686,
issueing the AIX Command "what named4" produces
bos.net.tcp.server 4.3.3.29 showing a newer timestamp/date
of X+1, or X+2, etc.
Concluding:
1st) "what named4" shows the older, appropriate timestamp/date
for 4.3.3.28. Whereas, "what named4" for 4.3.3.29 shows a newer
timestamp/date than 4.3.3.28.
2nd) "named4 -d 1" produces output in (/usr/tmp/named.run).
Here Bind4 shows the same timestamp/date to be the same both
prior to the apply of DNS Security PTF U474686 and after the
apply of DNS Security PTF U474686.

PROBLEM SUMMARY:
Submit named4 -d 1 on the command line. Output contained only
4.9.3 Bind. Information indicating the Cert Update is
necessary.

PROBLEM CONCLUSION:
Changed named4 to contain Cert Update Information.

------

APAR: IY18355 COMPID: 5765B9501 REL: 330
ABSTRACT: MMDELDISK STOPS WHEN IT FINDS BROKEN DISK ADDRESS

PROBLEM DESCRIPTION:
MMDELDISK STOPS WHEN IT FINDS BROKEN DISK ADDRESS

PROBLEM SUMMARY:
mmdeldisk terminiates when it finds a disk
block which can not be read.

PROBLEM CONCLUSION:
mmdeldisk should continue when
encountering a bad disk block and mark the indirect block
pointing at the bad disk block as bad.

------

APAR: IY18385 COMPID: 5765C3403 REL: 430
ABSTRACT: GRAPHICS IMAGES NOT BE GENERATED PROPERLY FOR HIGH NUMBER OF

PROBLEM DESCRIPTION:
The graphics images may not be generated properly. Various
parts of 3d models will drop in and out of the display.

PROBLEM CONCLUSION:
Increase the process soft memory limit to account for the
extra memory used for each thread so that the same amount
of memory is available for data no matter how many threads
are used.

------

APAR: IY18403 COMPID: 5765C3403 REL: 430
ABSTRACT: SU COREDUMPS WHEN 'SU -ABC' IS ISSUED FROM ROOT

PROBLEM DESCRIPTION:
The su command coredumps when 'su -abc' is issued
from the root on some systems. 'abc' can be a
username or some characters. This is always not
recreatable. It occurs only on some systems.

PROBLEM SUMMARY:
The su command coredumps when an option is
given incorrectly as described above when infact
it should have returned an incorrect option error.

PROBLEM CONCLUSION:
The su command coredumps because of uninitialized
local variable 'pw' in getcmd function which is
accessed at a later stage. This is now initialized
to NULL.

------

APAR: IY18413 COMPID: 5765C3403 REL: 430
ABSTRACT: TABLET WILL KILL X AFTER RUNNING TEST CASE 3RD TIME

PROBLEM DESCRIPTION:
X process dies after changing the primary pointer to a tablet
after several iterations.

PROBLEM CONCLUSION:
Make sure device is open in the "device on" case statement.

------

APAR: IY18416 COMPID: 5765C3403 REL: 433
ABSTRACT: CPSD COREDUMPS WHEN THERE IS NO IKE CERTIFICATE DATABASE

PROBLEM DESCRIPTION:
cpsd coredumps as soon as it started.

PROBLEM CONCLUSION:
Modified the code not to double free the password buffer.

------

APAR: IY18427 COMPID: 5765C3403 REL: 430
ABSTRACT: RPC.MOUNTD MEMORY LEAK / LOCKUP

PROBLEM DESCRIPTION:
The changes made in APAR IY10455 (bos.net.nfs.client 4.3.3.31)
had two issues, causing a slow but steady memory leak as well
as a risk of lockups of rpc.mountd, making it stop servicing
new requests.

LOCAL FIX:
Backing out APAR IY10455.

PROBLEM SUMMARY:
***************************************************************
*USERS AFFECTED: *
* All users with the following filesets at these levels *
* bos.net.nfs.client 4.3.3.31
* bos.net.nfs.client 4.3.3.50
***************************************************************
*PROBLEM DESCRIPTION: *
* The changes made in APAR IY10455 (bos.net.nfs.client
* 4.3.3.31) had two issues, causing a slow but steady
* memory leak as well as a risk of lockups of
* rpc.mountd, making it stop servicing new requests.
***************************************************************
*RECOMMENDATION: *
* Apply apar IY18427
***************************************************************

PROBLEM CONCLUSION:
One of the struct hostent fields that are duplicated for each
mountd request never gets deallocated. This causes a slow but
steady memory leak. Also, one unlock operation was accidentally
moved under an if condition, sometimes causing a lockup since
the lock was not released before the thread exited.

------

APAR: IY18452 COMPID: 5765B9501 REL: 330
ABSTRACT: DEADLOCK RESTRIPING A METAFILE.

PROBLEM DESCRIPTION:
deadlock restriping metadata.

PROBLEM SUMMARY:
Deadlock when running restripe command.

PROBLEM CONCLUSION:
Correct locking error in GPFS

------

APAR: IY18460 COMPID: 5765C3403 REL: 430
ABSTRACT: INCORRECT RING_SPEED IN TOKENRING MIB

PROBLEM DESCRIPTION:
Incorrect ring_speed in Tokenring MIB to snmp.

PROBLEM SUMMARY:
The ring_speed in the Token Ring MIB was reported
incorrectly. It is now set correctly.

PROBLEM CONCLUSION:
ring_speed in Tokenring MIB is reported correctly.

------

APAR: IY18461 COMPID: 5765C3403 REL: 430
ABSTRACT: MKSYSB PROGRESS MESSAGES ERRONEOUS

PROBLEM DESCRIPTION:
This is usually not seen because most backups are probably
of a large number of files most of which are not large.
The backup completes without ever showing a final total of
files or percentage it thought it backed up. The problem is
when all/most of the files are backed up are large. This
allows it to display progress almost to the individual number
of files it is backing up. As a result it may show near the
end that it has backed up more than it thought was available.
For example: 18 of 20 120%

PROBLEM SUMMARY:
Under certain circumstances mksysb/savevg can report that it
has backed up more files then it though were available. For
example, it could report 20 of 18 114%. With this apar, it
will never go beyond 100% and will not report beyond the
number of files that it is attempting to back up.

PROBLEM CONCLUSION:
This apar fix will make sure that mksysb/savevg does not
report beyond 100% backup.

TEMPORARY FIX:
It is just erroneous information, the backup functions
properly.

------

APAR: IY18481 COMPID: 5765C3403 REL: 430
ABSTRACT: GXT3000/2000 OPENGL BASED APP. CORES CLOSING WINDOW.

PROBLEM DESCRIPTION:
On GXT3000P or GXT2000P adapters, the customers OpenGL based
application cores when they exit the application.

PROBLEM CONCLUSION:
The application is using the Run-time linker (RTL) and uses
dlopen and dlsym to load and get the address of the OpenGL
and GLX subroutines. This creates a problem because the
OpenGL subroutines have two sets of entry points: one set
that is used when a direct context is current and another set
when an indirect context is current or no context is current.
The method used to swap these subroutine entry points does
work when RTL is used.
To get around the problem, an application that runs into this
situation must define an AIX environment variable before
OpenGL client initialization is performed (most likely done
in the start-up scripts before the application is loaded).
The variable is _OGL_RTL_ACTIVE. It can be defined with any
value. When this variable is defined, then GLX will not
change the set of entry points that will be used for direct or
indirect contexts. The entry points used will check to see
whether the direct or indirect version is required and will
call the appropriate routine.
The drawback to this solution is that there is a potential
for a significant performance degradation when using
a direct context due to the extra overhead required to get
to the direct context version of the subroutines.

------

APAR: IY18489 COMPID: 5765B9501 REL: 330
ABSTRACT: INCREASEINDIRECTIONLEVEL LIVELOCK RESERVING LOGSPACE

PROBLEM DESCRIPTION:
increase indirection level livelock reserving logspace

PROBLEM SUMMARY:
fix potential deadlock discovered in
development.

------

APAR: IY18505 COMPID: 5765C3403 REL: 430
ABSTRACT: GXT4000P OPENGL APP. CORES WHEN COMPILED WITH

PROBLEM DESCRIPTION:
On GXT4000P adapter, customers OpenGL based application cores
when compiled with option -bmaxdata:2000000000.

PROBLEM CONCLUSION:
Modify Makefile such that some debug code is NOT
compiled into the executable.

------

APAR: IY18528 COMPID: 5765B9501 REL: 330
ABSTRACT: ASSERT SUBROUTINE FAILED: !UNPINSOMEBUFFER: NO BUFFERS FOUND

PROBLEM DESCRIPTION:
assert subroutine failed: !unpinSomeBuffer: no buffers found

PROBLEM SUMMARY:
Remove incorrect assert. When other threads are
pinning and
unpinning,
the accounting cannot prevent this assertion from happening.
Just wait a litle while and try again.

------

APAR: IY18529 COMPID: 5765B9501 REL: 330
ABSTRACT: SIGSEGV IN MHWRITEDATA PROLOG WHEN OUT OF DATASHIPPING MODE

PROBLEM DESCRIPTION:
SIGSEGV in mhWriteData prolog when out of datashipping mode
Thread trace back ...
0xD028977C compare_and_swap() + 0x30
0x101E4BBC mhWriteData(TscMsgHeader*,void*,const NodeIncarnation
0x100075C8 tscHandleMsg(TscMsgHeader*,void*,const NodeIncarnatio
0x10023AEC RcvWorker::main() + 0x114
0x1002393C RcvWorker::thread(int) + 0x84
0x1000E99C Thread::callBody(Thread*) + 0x9C
0x10175328 Thread::callBodyWrapper(Thread*) + 0x98
0xD0131358 _pthread_body() + 0xD0
0xFFFFFFFC

PROBLEM SUMMARY:
Incorrect handling of a message received after data shipping
was turned offf caused a segmentation violation

PROBLEM CONCLUSION:
Correct test in data shipping.

------

APAR: IY18530 COMPID: 5765C3403 REL: 430
ABSTRACT: BPF CRASH IN TIMER

PROBLEM DESCRIPTION:
bpf crashes in timer code called with
interrupts disabled.

PROBLEM SUMMARY:
tcpdump could result in crash in bpf if aix_tap was
not immediately available. Situation resolved.

PROBLEM CONCLUSION:
tcpdump no longer crashes in bpf if aix_tap not
immediately available.

------

APAR: IY18553 COMPID: 5765C3403 REL: 430
ABSTRACT: INTERMITTENT SYSTEM CRASH WHEN DOING IPV6 PINGS

PROBLEM DESCRIPTION:
Intermittent system crash when doing IPv6 pings due to a
double-free of an mbuf.

PROBLEM CONCLUSION:
Make sure icmp6_input returns the correct value to indicate
that there is no next header.

------

APAR: IY18556 COMPID: 5765C3403 REL: 430
ABSTRACT: ULTRA2 SCSI PCI ADAPTER DIAGNOSTICS CAUSES AIX ERROR LOG

PROBLEM DESCRIPTION:
Running diagnostics on the Ultra 2 SCSI PCI Adapter(Channel B)
causes an error to be logged against Channel A in the error
log.

PROBLEM CONCLUSION:
Ignore the error message. It does not indicate a failure
on the adapter.

------

APAR: IY18560 COMPID: 5765C3403 REL: 430
ABSTRACT: SHUTDOWN OF LOCKD VERY SLOW

PROBLEM DESCRIPTION:
When there are many NFS clients which do file locking on an
NFS server, it may take a long time for rpc.lockd to exit
after it is asked to shutdown.

PROBLEM CONCLUSION:
Can vfs_search/vn_search only once when determining if
locks exists for the cached sysid's.

------

APAR: IY18583 COMPID: 5765C3403 REL: 430
ABSTRACT: /TMP/CODCRON GETTING CREATED UNNECESSARILY

PROBLEM DESCRIPTION:
A temporary file is being created on all machines when
changing system attributes. This file is used for modifying
the root crontabs file and poses no problem to that crontabs
file other than it is not being removed from /tmp.

LOCAL FIX:
This customer is using a shell script to execute chdev which
results in the temporary file being created. A line added
to the shell script to remove that file should be
a sufficient workaround.

PROBLEM SUMMARY:
A temporary file is being created on all machines when
changing system attributes. This file is used for modifying
the root crontabs file and poses no problem to that crontabs
file other than it is not being removed from /tmp.

PROBLEM CONCLUSION:
added code to remove the temporary file once it's not needed.

------

APAR: IY18597 COMPID: 5765B9501 REL: 330
ABSTRACT: ONLINE MMCHECKQUOTA: DEALLOC ASSERTS IN FIXSHADOWTABLEBLOCKCOUNT

PROBLEM DESCRIPTION:
online mmcheckquota: dealloc asserts in FixShadowTableBlockCount

PROBLEM SUMMARY:
GPFS self check logic terminated while running mmcheckquota

PROBLEM CONCLUSION:
Fix serialization error in mmcheckquota

------

APAR: IY18602 COMPID: 5765B9501 REL: 330
ABSTRACT: INODE PREFETCH LOOPING

PROBLEM DESCRIPTION:
inode prefetch looping

PROBLEM SUMMARY:
Infinite loop under rare conditions found in
development.

PROBLEM CONCLUSION:
InodePretchInstance::WorkerThreadBody
nIdle variable not initialized resulting in prefetch
thread spinning.

------

APAR: IY18608 COMPID: 5765C3403 REL: 430
ABSTRACT: MPOA DROPS THE LAST FRAGMENTED PACKET IF IPDATA SIZE < 8 BYTES

PROBLEM DESCRIPTION:
Last packet of a fragment dropped if IPDATA size < 8 bytes.

PROBLEM CONCLUSION:
Only do a pullup when an ICMP Must Fragment message has be
sent back.

------

APAR: IY18610 COMPID: 5765C3403 REL: 430
ABSTRACT: FDDI PCI FLASH UPDATE UTILITY FAILS

PROBLEM DESCRIPTION:
With the new flash module on the FDDI, the old flash
utility will not be able to flash the chip

PROBLEM CONCLUSION:
A very small change was made to the flash utility to allow
the tool to be compatible with both, the old and new chips

------

APAR: IY18655 COMPID: 5765B9501 REL: 330
ABSTRACT: MMDELDISK FAILS WITH NOT ENOUGH MEMORY

PROBLEM DESCRIPTION:
mmdeldisk fails with not enough memory

PROBLEM SUMMARY:
 E_NOMEM error received when running mmdeldisk

PROBLEM CONCLUSION:
Correct memory leak in the token manager
when running mmdeldisk.

------

APAR: IY18676 COMPID: 5765B9501 REL: 330
ABSTRACT: SLOW NFS FLUSH_RANGE CALLS

PROBLEM DESCRIPTION:
slow nfs flush_range calls

PROBLEM SUMMARY:
NFS performance improvement

PROBLEM CONCLUSION:
Streamline one piece of the data flush when
running NFS
writes.

------

APAR: IY18685 COMPID: 5765C3403 REL: 430
ABSTRACT: CHINESE BIG-5 REFRESHING PROBLEM

PROBLEM DESCRIPTION:
chinese chars of diffrent heights, when cat from a file,
is not refrshed in dtterm.

PROBLEM CONCLUSION:
handled refrshing properly in dtterm.

------

APAR: IY18810 COMPID: 5765C3403 REL: 430
ABSTRACT: X SERVER CORES WHEN DESTROYING GLX RESOURCES

PROBLEM DESCRIPTION:
The X-Server may core dump when closing an application's
window.

PROBLEM CONCLUSION:
Free the clients resources in reverse order.

------

APAR: IY18968 COMPID: 5765B9500 REL: 140
ABSTRACT: README UPDATE RELEATIVE TO IBM ESS STORAGE

PROBLEM DESCRIPTION:
README update relative to IBM ESS Storage

------

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]