SUMMARY: 4.1.3 patches (additional useful info)

From: Roy Richter PH/32 (rrichter@ph.gmr.com)
Date: Fri Oct 02 1992 - 01:25:24 CDT


This just in from jim.litchfield@sunwest2.West.Sun.COM
(James Litchfield - NW Area Solaris 2.0 Transition Boss).
It contains the headings from teh README's of available
patches, Some info on the patches, and a shar file
containing the README's from the patches. I deleted the
Open Issues section because it is quite long, and duplicated
in TFM. Anyone wants it, I'll send it to the fist 5 people
who ask and ignore the rest.

If anyone gets these patches and places them in a public
anonymous ftp place, please post that info also.

--
Roy Richter                  Internet: rrichter@ph.gmr.com
Physics Dept, GM Research    UUCP:     rphroy!rrichter
        and Environmental

- The first 4.1.3 patch list. - ----------

4.1.3 Patch List (9/28/92)

100075-09_README: 4.1;4.1.1;4.1.2;4.1.3: rpc.lockd jumbo patch 100170-09_README: SunOS 4.1;4.1.1;4.1.2;4.1.3: jumbo patch to fix various ld problems 100249-04_README: SunOS 4.1;4.1.1;4.1.2,4.1.3: automounter jumbo patch 100283-03_README: SunOS 4.1.1;4.1.2;4.1.3: in.routed mishandles gateways, multiple 100296-04_README: SunOS 4.1.1, 4.1.2, 4.1.3: netgroup exports to world 100338-05_README: SunOS 4.1.1;4.1.2;4.1.3: system crashes with assertion failed panic 100342-03_README: SunOS 4.1 4.1.1 4.1.2 4.1.3: NIS client needs long recovery time if server reboots 100359-06_README: SunOS 4.1.1;4.1.2;4.1.3: streams jumbo patch 100372-02_README: SunOS 4.1.1;4.1.2;4.1.3: tfs and c2 do not work together 100383-05_README: SunOS 4.0.3;4.1;4.1.1;4.1.2;4.1.3: rdist security enhancement 100507-04_README: SunOS 4.1.1, 4.1.2, 4.1.3: tmpfs jumbo patch 100527-03_README: SunOS 4.1.1;4.1.2;4.1.3: rsh uses old-style selects instead of 4.0 selects 100536-02_README: SunOS 4.1.2, 4.1.3: NFS can cause panic: assertion failed crashes 100557-02_README: SunOS 4.1.2, 4.1.3: ftp Jumbo patch 100567-03_README: SunOS 4.1.2, 4.1.3: mfree panic due to mbuf being freed twice 100623-02_README: 4.1.2;4.1.3: UFS jumbo patch 100726-01_README: 4.1.3: large memory (.5G and up) Galaxy/Ross systems panic while booting

This is an accumulation of Solaris 1.1 issues we have encountered.

Many of the support contract customers did not receive a printed version of the first eight chapters of the Solaris 1.1 Release Manual. Instead, they can use the AnswerBook to view these chapters.

Chapter 6 of the Release Manual provides an accurate but incomplete list of the CTE patches integrated into SunOS 4.1.3. Below is the most up-to-date list. These lists differ, because occasionally a SunOS 4.1.3 fix is back-ported to an earlier release and assigned a patchid.

Please discourage everyone from installing a patch from an earlier release after installing SunOS 4.1.3; this may cause their system to crash. CTE is currently porting and testing a number of patches, which were not integrated into SunOS 4.1.3. Below is a list of patch available now and a list of ones to be available soon. Should a customer require a patch from an earlier release, which is not on one of these lists, please use the standard escalation process to request a port.

Attached is an e-mail version of the Open Issues Section from the SunOS 4.1.3 Release Manual. If you have not already done so, you should read it. In particular, read the sections on USRSTACK Changes on Sun-4m Machines, Twisted-Pair Ethernet on SPARCstation 10, and Patches Included in Patch Zone, which includes 13 OpenWindows patches. Following the "Open Issues Section" is some additional discussion on the USRSTACK changes, what motivated them and how they might impact customers.

CTE Patches Integrated Into SunOS 4.1.3 --------------------------------------- 100075-08 1040406 1 3 KLM protocol was changed but the RPC version numb 100075-08 1046373 2 1 a shared or exclusive lock does not work under Su 100075-08 1063692 2 2 deadlock is not detected in nfs lockin 100075-08 1064183 3 3 downgrading of a lock does not grant other read l 100075-08 1071435 1 1 panics: corrupt file lock lis 100075-08 1074757 1 1 rpc.lockd in 4.1.2 grows to few MB ( a memory lea 100075-08 1077402 1 1 A file can get left locked by an exited proces 100075-08 1080359 1 1 Read lock gets stuck in the kernel loc 100188-02 1070495 1 1 programs using pty can get output from previous a 100224-03 1051832 2 3 rmail dumps core on certain (unknown) error 100274-02 1044478 1 1 This bug causes dup biodone related panics with S 100371-01 1044726 2 2 showfh timeouts before accessing nfs file handl 100381-01 1068938 3 2 du does not work with VMS(NFS) filesystem due con 100384-01 1063432 2 2 tip to RFS mounted serial port causes panic: data 100399-01 1066556 1 1 csh malloc memory leak results in Memory Fault er 100402-01 1069046 2 1 cron dumps core on what appears to be correct cro 100408-01 1047916 3 2 keypad() misinterprets function keys when remotel 100408-01 1070312 3 2 terms ins/del line and scroll region capability c 100425-01 1055736 4 3 whois is getting host unknown when it uses the ha 100431-04 1063683 3 3 back facing hollow polygon inconsistent when usin 100431-04 1068120 2 2 Undrawing antialiased lines in QUM mode has no ef 100431-04 1073137 2 2 Back-faceed QUADS are slow on GT when FACE CULLIN 100431-04 1073181 3 2 Performance degrades for polygons following a vie 100431-04 1074981 4 3 The color is sometimes incorrect for clipped, dep 100431-04 1080766 3 3 Non-dot markers are slo 100431-04 1086669 2 2 Performance problem - short triangle strips on th 100453-01 1073672 4 2 Increase the 1.5MB DVMA window available to the G 100456-04 1075220 2 1 mcldup does not block interrupts causing mbuf clu 100469-01 1039388 4 4 cdrom mount error messag 100474-01 1074337 2 2 format: assertion failure on 1.3G elites on sun 100475-01 1073234 2 2 mmap system call fails on galaxy causing unexpect 100484-01 1078167 2 2 SCSI driver for micropolis drive produces message 100490-01 1080928 2 2 nl_langinfor(YESSTR) of libxpg.a caused core dump 100494-01 1074704 1 1 WIDs are freed when cg12 contexts are destroyed e 100494-01 1074822 1 1 munmap'ing ctl register page clobbers shmem buffe 100495-01 1074424 1 1 sybase dataserver crashes 4m 4.1.2_ER machine - x 100502-01 1078920 1 1 Transformation matrix in one static context influ 100505-01 1078521 3 3 Zero length directories can be left when a system 100507-01 1038651 1 2 Using tmpfs ie ram /tmp can cause data initializa 100511-01 1082046 3 3 SunDiag cg6test failed on segmentation violatio 100516-01 1077539 2 1 increase HEAPBYTE 100520-01 1031907 1 1 cron dies when daylight time starts or stop 100521-01 1080702 1 1 4/75GT crash when executing screenload with 24bit 100534-01 1083022 2 1 nettest dies with Segmentation Violation on 4.1.2 100536-01 1066287 3 3 nfs hang when looking at large file being changed 100537-01 1069770 2 1 swapper hangs waiting for page of u-area to come 100537-01 1073679 1 1 async i/o peaks can hang syste 100539-01 1085014 1 1 umount of busy hsfs filesystem succeeds causing p 100542-01 1063059 2 1 ipi driver doesn't return error when it didn't ge 100542-01 1077801 1 1 galaxy with cg6+presto crashes on ip 100542-03 1075751 2 1 VME dropped INT_ACK cycle. Plus Pani 100547-01 1079876 1 1 fputest kvm_open faile 100569-01 1072183 3 3 gttest takes too long to report Firmware failur 100569-01 1083997 1 1 mouse freezes when running OW 3.0 applications on 100569-02 1086614 1 1 X Server window ops become very slow when running 100575-02 1080798 1 1 IPI disk runs slower on a 4/670 then a 4/47 100575-02 1086143 1 1 Galaxy system crashes with a watchdog reset - INS 100577-02 1087264 2 2 atomic swap command fails only on IPC running 4.1 100591-01 1089381 1 1 unimp_ins trap trashes %g[123] on MP system 100597-01 1075751 2 1 VME dropped INT_ACK cycle. Plus Pani 100618-01 1074668 1 1 Kernel panics when running sunvision 1.1 as super 100635-01 1090435 1 1 nice'd processes hang in run state and aren't kil 100636-01 1082229 1 1 fork creates duplicate PID 100644-01 1068571 1 1 Double buffered, depth cued markers and lines inc 100644-01 1069371 1 1 Lights in one context modified by changes to othe 100644-01 1078920 1 1 Transformation matrix in one static context influ 100731-01 1077035 1 1 itrunc: /mnt/123598 new size = 0, blocks = -29376

CTE Patches For SunOS 4.1.3 Available Now ----------------------------------------- 100075-09 rpc.lockd jumbo patch 100249-04 automounter jumbo patch 100283-03 in.routed mishandles gateways 100296-04 netgroup exports to world 100338-05 SunOS 4.1.1;4.1.2: system crashes with assertion fa 100342-03 NIS client needs long recove 100359-06 streams jumbo patch 100372-02 tfs and c2 do not work together 100383-05 security: rdist security 100507-04 SunOS 4.1.2: tmpfs fix 100527-03 rsh uses old-style selects inste 100536-02 NFS can cause panic: assertion failed 100557-02 ftp Jumbo patch 100567-03 mfree panic due to mbuf bein 100623-02 UFS jumbo patch 100661-01 panic ifree inode bmap iget 100662-01 libc replacement with all 4.1.3 CTE

CTE Patches For SunOS 4.1.3 To Be Ported ---------------------------------------- 100170-08 jumbo patch to fix various l 100173-08 NFS Jumbo Patch 100257-04 jumbo patch for ld.so, ldd, and 100305-07 lpd can be used to 100347-02 panic's when using fifo's (named 100377-04 security problems with "LD_" 100407-05 accounting files are corrupted 100458-02 Setitimer sometimes fails to 100496-01 ptrace panic with a memory faul 100531-02 domestic libc replacement 100532-02 international libc replacement 100556-02 cpio data corruption when 2 100564-01 C2 Jumbo patch 100581-03 st driver not reading past a double EO 100598-01 LPR-INCORRECT HEADER BANNER 100600-01 fsck -p won't check over 31 par 100606-01 ccom More than 15000 symbols ca 100620-01 Compiler generates incorrect cod 100630-01 environment variables can be used 100631-01 environment variables can be used 100633-01 environment variables can be used 100634-01 Select system call hangs when linked with lnbio & l 100645-01 swapon with very large swap files hangs, sys 100651-01 Cron dumps core & Cron dies 100666-01 bad rmfree 100672-01 static libc.a missing asynchronous I- 100677-01 NFS server not responding when using 100684-01 RFS server patch 100689-01 large processes hang in "disk wait" when the

#! /bin/sh # This is a shell archive, meaning: # 1. Remove everything above the #! /bin/sh line. # 2. Save the resulting text in a file. # 3. Execute the file with /bin/sh (not csh) to create: # 100075-09_README # 100170-09_README # 100249-04_README # 100283-03_README # 100296-04_README # 100338-05_README # 100342-03_README # 100359-06_README # 100372-02_README # 100383-05_README # 100507-04_README # 100527-03_README # 100536-02_README # 100557-02_README # 100567-03_README # 100623-02_README # 100726-01_README # This archive created: Sun Sep 27 17:15:46 1992

export PATH; PATH=/bin:/usr/bin:$PATH if test -f '100075-09_README' then echo shar: "will not over_write existing file '100075-09_README'" else cat << \SHAR_EOF > '100075-09_README' Patch-ID# 100075-09 Keywords: lockd, rpc.lockd, file locking Synopsis: 4.1;4.1.1;4.1.2;4.1.3: rpc.lockd jumbo patch Date: 10/Sep/92

SunOS release: 4.1, 4.1.1, 4.1.2, 4.1.3 Unbundled Product: Topic: rpc.lockd jumbo patch BugId's fixed with this patch: 1077402 1080359 1040406 1046373 1063692 1064183 1071435 1074757 1097368

Changes incorporated in this version: 1097368

Architectures for which this patch is available: sun4 sun4c sun4m sun3 sun3x

Problem Description:

PROBLEMS FIXED BY 100075-09 PATCH

i) When /tmp was mounted as "tmpfs" file system lock recovery failed when clients rebooted.

ii) When locking with svr4 (or 4.x) machines and one of the machines was rebooted, locking to that machine would not work until both machines were rebooted as the client handle had wrong port of rpc.lockd.

iii) When a client systems recovers after crash, not all locks belonging to that system get released.

PROBLEMS FIXED BY 100075-08 PATCH

i) klm protocol was changed but the rpc number was not.

ii) Shared or exclusive lock releative to EOF does not work correctly.

iii) Deadlock is not detected when locking over NFS.

iv) Downgrade of a lock does not grant other read locks.

v) Panic "corrupt file lock list" might occur.

vi) rpc.lockd does not release memory allocated and grows in size.

vii) When doing "region locking" some regions do not get properly unlocked.

viii) Granting of locks not working correctly when local and remote locks on same file are used at the same time.

ix) Unlock that is interrupted by signal could leave the file as being locked by a process that does not exist.

x) Core dump of rpc.lockd when unlocking in lm_unlock_region.

xi) When server is rebooted locks are not reclaimed correctly.

PROBLEMS FIXED BY 100075-07 PATCH

WARNING: ------- Due to fixes for iv) v) and vii) listed below, there is a potential interoperabilty problem when 07 patch is the client and 06 patch is the server; ie. lock requests that are being queued in the 06 server may not get granted. There is no side-effect expected if the 07 patch is the server.

i) Check for sanity of lock requests and return EINVAL if lock ranges is abnormal (ie. start < end).

ii) Fix problem where locking to EOF request results in the last byte of the file not being locked.

iii) Fix problems where lock requests tend to get lost when interruptions to the daemon hosed its internal data structures; this may also cause the daemon to core dump in certain cases.

iv) Fix problems where the server lock daemon not sending back NLM_GRANTED_MSG to remote blocked clients.

v) Fix protocol implementation problem with NLM_GRANTED_MSG.

vi) Fixed problems where NLM_CANCEL_MSGS and KLM_CANCEL_MSG result in applications waiting forever.

vii) Fix error recovery mechanism to allow clients to reclaim locks after servers crash & recover.

viii) Fix the krpc call timeout to 5 sec instead of 45 sec for better response. This also fixed the potential problem where results for lock request is missed by the kernel.

ix) Fix "panic iinactive" on the server.

x) Fix problem where the temporary file /tmp/foo is not being removed after the client lock mgr crashes & recovers. xi) Fix panic trap data fault when more than 135 processes concurrently try an do file locking.

PROBLEMS FIXED BY 100075-06 PATCH

i) Fixed problems where locks were getting lost on a heavily loaded system, particularly when using shared locks or test lock calls.

ii) Fixed problem where pc-nfs applications were failing with "rpc.lockd: unable to unlock a lock" and "rpc.lockd: unable to set a lock."

iii) Fixed problem with automatic upgrade and downgrade of locks.

iv) Fixed problem with client reboot (L1 A) and locks not being recovered after that.

v) Fixed problem with signal interrupting lock calls and wrong error code returned.

vi) Fixed problem with infinite retry of lock on unlinked files.

vii) Fixed problem with local blocking shared locks not being granted a lock when one is available.

viii) Fixed problem with u-area overwrite when doing test lock.

ix) Fixed problem where messages like "klm_lockmgr: unlock denied?!" and "lock-manager: RPC error: .." no longer appear under normal operations.

x) Additionally this patch allows the kernel lockf debugging code to be turned on and off dynamically by setting/unsetting the variable lock_debug_on using the following command:-

#adb -w -k /vmunix /dev/mem physmem XXX lock_debug_on/W 1 ^D

PROBLEMS FIXED BY 100075-05 PATCH

i) Fixed problem with running out of file descriptors and getting RPC TIMEOUT errors, seen when running large number of diskless clients.

ii) Fixed problem where fd and fd structure was not released when doing test lock, eventually running out of file descriptors. This problem would occur when running WP and quitting out of the window.

iii) Fixed problem where rpc.lockd core dumps after a large amount of time when running "fame" application.

iv) Fixed problem where restarting lockd on a client results in the server not being able to communicate with the new lockd as it has an old client handle that is associated with a now invalid port number.

v) Fixed problem where upgrade from a read lock to a write lock is allowed when remote read locks are outstanding.

INSTALL: After extracting fix tape contents into /tmp, as root install the appopriate sun4 sun4c sun4m sun3 or sun3x patches as follows:

(NOTE: For 4.1.3 ONLY ufs_lockf.o and rpc.lockd need to be patched since all the patches upto -08 has been integrated in the 4.1.3 source.) cd /sys/{sun4,sun4m,sun4c,sun3,sun3x}/OBJ mv kern_descrip.o kern_descrip.o.FCS mv klm_lockmgr.o klm_lockmgr.o.FCS mv ufs_lockf.o ufs_lockf.o.FCS mv /usr/etc/rpc.lockd /usr/etc/rpc.lockd.FCS cp /tmp/{sun4,sun4m,sun4c,sun3,sun3x}/kern_descrip.o kern_descrip.o cp /tmp/{sun4,sun4m,sun4c,sun3,sun3x}/klm_lockmgr.o klm_lockmgr.o cp /tmp/{sun4,sun4m,sun4c,sun3,sun3x}/ufs_lockf.o ufs_lockf.o cp /tmp/{sun4,sun4m,sun4c,sun3,sun3x}/rpc.lockd /usr/etc/rpc.lockd chmod 444 kern_descrip.o klm_lockmgr.o ufs_lockf.o chmod 755 /usr/etc/rpc.lockd Then a new kernel will need to be made Please refer to the System and Network Administration manual on building and installing a custom kernel.

SHAR_EOF chmod +x '100075-09_README' fi if test -f '100170-09_README' then echo shar: "will not over_write existing file '100170-09_README'" else cat << \SHAR_EOF > '100170-09_README' Patch-ID# 100170-09 Keywords: jumbo-patch ld-1.144 shared LD_LIBRARY_PATH -Bstatic SPARCworks Synopsis: SunOS 4.1;4.1.1;4.1.2;4.1.3: jumbo patch to fix various ld problems Date: 23-Sep-92 SunOS release: 4.1 4.1.1 4.1.2 4.1.3 Topic: jumbo ld patch with 4.1.3 compatibility BugId's fixed with this patch: 1019004 1032739 1034788 1034833 1037879 1041946 1042261 1044524 1045272 1046462 1050594 1064820 1065275 1019004 1083430 1080802 1038809 1088382

Architectures for which this patch is available: sun4(all) sun3(all)

Patches which may conflict with this patch:

Obsoleted by:

Problem Description:

Jumbo patch for all known ld bug fixes. The same patch can be used for 4.1, 4.1.1., 4.1.2, and 4.1.3. All SPARC machines should use the sun4 version; all Sun3 machines should use the sun3 version. Note that installing this patch under 4.1.1 also requires the installation of the 4.1.1 libc jumbo patch, 100266 or 100267.

1.132 includes fixes for these bugs:

1032739: ld core dumps with many libs in multiple directories 1034788: -r and -pic do not mix bad secondary magic number on... (also includes 1034844 as a side effect.) ** NOTE: the above were previously supplied as the XEROX patch. 1034833: ld: can't mixed -r with -Bstatic or -A flag in 4.1prefcs... 1037879: Cannot create executable with shared object which points... (This includes the "fix to the fix" for the new problem that Brown U. reported.) 1042261: ld only recognized first directory in LD_LIBRARY_PATH 1044524: multiply defined symbols and seg. fault caused by 4.1's ld 1045272: ld -u & -r do not seem to work properly

1.135 includes all of the above and in addition includes:

1019004: -assert definitions can fail to report undefined symbols, as well as all follow-on problems relating to previous alleged fixes for this bug.

1.136 includes all of the above and in addition includes:

1046462: ld fails with segmentation fault when processing nested #include files

1.136 got updated to include fixes for:

1019004: (do not account for slots created by "definition checking.") 1050594: fix uninitialized structure entries.

1.137 includes a fix for:

1064820: ld can produce bogus diagnostics when processing cascaded .so dependencies

1.138 includes fixes for:

1065275: program dumps core with shared objects having only statics. 1041946: same problem.

1.141 includes fixes for:

1019004: leave output file executable when complaining about undefines. 1083430: fix to 1019004 that eliminates a side effect that confused the loader about the number of allocated vs. number of used output relocations. 1080802: fix -r and -PIC combination

1.142 includes fixes for:

1083430: update to fix potential null pointer dereference xxxxxxx: performance improvement for programs with very large numbers of symbols.

1.143 includes fixes for:

1088382: SPARCworks 2.0 unusable.

1.144 includes fixes for:

1080802: update for -r and -PIC processing, be sure STABs symbol entries are correctly counted when producing symbol table indices for relocation records. xxxxxxx: implement table expansion for larger "ld" capacity.

*** This ld patch is 1.144 ***

NOTE: As of 9/19/91 patchid 100266 (domestic) or 100267 (international) is the new libc needed to repair bug 1045471. for 4.1.1. Bug 1045471 is a libc bug revealed when applying this ld patch. Many applications may be linked without necessarily being affected by bug 1045471.

INSTALL:

As root:

mv /bin/ld to something else (e.g. /bin/ld.FCS) cp `arch`/ld /bin/ld chmod 755 /bin/ld

SHAR_EOF chmod +x '100170-09_README' fi if test -f '100249-04_README' then echo shar: "will not over_write existing file '100249-04_README'" else cat << \SHAR_EOF > '100249-04_README' Patch-ID# 100249-04 Keywords: automount symbolic link timeout, jumbo Synopsis: SunOS 4.1;4.1.1;4.1.2,4.1.3: automounter jumbo patch Date: 18-June-92

SunOS release: 4.1, 4.1.1, 4.1.2, 4.1.3

Bug ID's fixed with this patch: 1038308, 1048875, 1044048, 1083225

Patches which conflict with this patch:

Architectures for which this patch is available: sun4, sun4c, sun4m

Problem description:

Bug 1083225 (March 3, 1992, -04) - -------------------------------- This version of automount is based on SunOS4.1.3(same as 4.1.2) Source.

If an entry is removed from /etc/mtab while the automounter daemon is reading it, it is possible for the automounter to get into a state where it will loop forever, printing

"WARNING: /etc/mtab: line <number>: bad entry"

on the console and hang the server.

Bug 1038308:

In 4.1, the automounter sets up a different symbolic link 5 min. after the first access. This happens if there are multiple locations for the same mount in the auto mount map entry, and those locations don't share the same subdir.

Bug 1048875

The automounter forks due to rpc.rexd. This problem is usually seen when using the on command. The automounter will complain it cannot find /etc/mtab.

Bug 1044048

Automounter consumes a steady, increasing percentage of memory. Coinciding with this is the consumption of swap space as well.

INSTALL: As root: mv /usr/etc/automount /usr/etc/automount.fcs

Install the new version of automount: cp `arch -k`/{OS Release}/automount /usr/etc/automount chown root /usr/etc/automount chgrp staff /usr/etc/automount chmod 755 /usr/etc/automount SHAR_EOF chmod +x '100249-04_README' fi if test -f '100283-03_README' then echo shar: "will not over_write existing file '100283-03_README'" else cat << \SHAR_EOF > '100283-03_README' Patch-ID# 100283-03 Keywords: gateway, multiple route Synopsis: SunOS 4.1.1;4.1.2;4.1.3: in.routed mishandles gateways, multiple routes Date: 10/Sep/92 SunOS release: 4.1.1 4.1.2 4.1.3 Unbundled Product: Unbundled Release: Topic: Patch version update for 4.1.3 compatibilty BugId's fixed with this patch: 1028094 1085405

Architectures for which this patch is available: sun4 sun4c sun3

Patches which may conflict with this patch:

Obsoleted by:

Problem Description:

Bugid 1085405 fixed in -02 - ------------ The SunOS route daemon (4.1.1 and 4.1.2 and 4.1.3) incorrectly handles a system configuration that has two network interfaces attached to the same IP network.

Bugid 1028094 fixed in -01 - ------------- The routing daemon creates and broadcasts invalid routes for network/gateway pairs found through "active" entries in the /etc/gateways file.

INSTALL:

AS ROOT: mv /usr/etc/in.routed /usr/etc/in.routed.fcs cp `arch -k`/{OS Version}/in.routed /usr/etc

Make sure the permissions of the new in.routed are as follows: -rwxr-xr-x 1 root 32768 Oct 11 1990 /usr/etc/in.routed*

kill old in.routed process and restart with the new one.

SHAR_EOF chmod +x '100283-03_README' fi if test -f '100296-04_README' then echo shar: "will not over_write existing file '100296-04_README'" else cat << \SHAR_EOF > '100296-04_README' Patch-ID# 100296-04 Keywords: nfs rpc mount exportfs Synopsis: SunOS 4.1.1, 4.1.2, 4.1.3: netgroup exports to world Date: 6/18/92 SunOS release: 4.1.1, 4.1.2, 4.1.3 Unbundled Product: Unbundled Release: Topic: Patch revision update for 4.1.3 support BugId's fixed with this patch: 2000680 1044852 1048890 1047410

Architectures for which this patch is available: sun3, sun4, sun4m

Patches which may conflict with this patch:

Obsoleted by:

Problem Description:

Bugid: 2000680(June 18, 1992, -04) This patch version (-04) for 4.1.3 with no source change. If the cached list of netgroups that a client is not a member of exceeds the cache capacity then the mount daemon will acknowledge the client's membership of any netgroup even if it is not a member.

Bugid: 1048890 If the access list of hosts is a string under 256 chars then things work as expected, but that if it is longer everyone can mount the filesystem.

Bugid: 1048524 Additionally this patch also fixes a problem where the cached netgroup entry may contain groups from the previous mount.

Bugid: 1047410 rpc.mountd dies when yp_get_default_domain returns null.

INSTALL:

As root: mv /usr/etc/rpc.mountd /usr/etc/rpc.mountd.FCS cp `arch`/rpc.mountd /usr/etc chown root /usr/etc/rpc.mountd chgrp staff /usr/etc/rpc.mountd chmod 755 /usr/etc/rpc.mountd

You will then need to kill the currently running rpc.mountd and restart it as per /etc/rc.local or simply reboot the system. SHAR_EOF chmod +x '100296-04_README' fi if test -f '100338-05_README' then echo shar: "will not over_write existing file '100338-05_README'" else cat << \SHAR_EOF > '100338-05_README' Patch-ID# 100338-05 Keywords: panic assertion failed Synopsis: SunOS 4.1.1;4.1.2;4.1.3: system crashes with assertion failed panic Date: 4/Sep/92

SunOS release: 4.1.1 4.1.2 4.1.3

Topic: Revision update for 4.1.3 compatibility. Kernel panic assertion failed: vp->v_stream == stp,file: os/str_io.c, line: 609

BugID's fixed with this patch: 1047951

Architectures for which this patch is available: sun4m sun4 sun4c sun3 sun3x

Problem description:

When multiple snodes refer to the same device the above panic can occur when the device is closed.

Note: Version 2 fixes a problem where the sun4/spec_vnodeops.o binaries were corrupt on the patch database.

Version 4 fixes a problem where a panic page_hashout might occur when block devices are used directly. Note for disksuite only the 4.1.2 version is compatible.

Fix description:

The fix includes patched version of:

spec_vnodeops.o

Install instructions:

As root install the appropriate patch as follows:

mv /usr/sys/`arch -k`/OBJ/spec_vnodeops.o /usr/sys/`arch -k`/OBJ/spec_vnodeops.o_orig cp {4.1.1, 4.1.2, 4.1.3}/`arch -k`/spec_vnodeops.o /usr/sys/`arch -k`/OBJ chmod 444 /usr/sys/`arch -k`/OBJ/spec_vnodeops.o Then a new kernel will need to be made.

SHAR_EOF chmod +x '100338-05_README' fi if test -f '100342-03_README' then echo shar: "will not over_write existing file '100342-03_README'" else cat << \SHAR_EOF > '100342-03_README' Patch-ID# 100342-03 Keywords: NIS client server rebind Synopsis: SunOS 4.1 4.1.1 4.1.2 4.1.3: NIS client needs long recovery time if server reboots Date: 18/June/92

SunOS release: 4.1, 4.1.1, 4.1.2, 4.1.3 Unbundled Product: Unbundled Release: Topic: NIS ypbind patch revision update for 4.1.3 compatibility BugID'd fixed for this patch: 1046416

Architectures for which this patch is available: sun4

Patches which may conflict with this patch:

Obsoleted by:

Problem Description:

Bug 1046416:

If you bring a ypserver down into single user and then boot it into multi user by either typing control D or reboot, the yp clients will take a long time to rebind to the server.

********************* WARNING ******************************

This is a new version of ypbind that never uses the NIS binding file to cache the servers binding. This will have the effect of fixing the current symptom. However, it might degrade the overall performance of the system when the server is unavailable.

*************************************************************

INSTALL::

As root and for the correct architecture directory.

Kill the currently running ypbind:

ps aux|grep ypbind kill <processid of ypbind>

Make a backup copy of ypbind:

mv /usr/etc/ypbind /usr/etc/ypbind.FCS

Install the new version of ypbind:

cp `arch`/ypbind /usr/etc

chown root /usr/etc/ypbind chmod 755 /usr/etc/ypbind

Restart ypbind

/usr/etc/ypbind SHAR_EOF chmod +x '100342-03_README' fi if test -f '100359-06_README' then echo shar: "will not over_write existing file '100359-06_README'" else cat << \SHAR_EOF > '100359-06_README' Patch-ID# 100359-06 Keywords: panic assertion failure, machine freeze, console printf hangs, jumbo Synopsis: SunOS 4.1.1;4.1.2;4.1.3: streams jumbo patch Date: 04/Sep/92 SunOS release: 4.1.1 4.1.2 4.1.3 Topic: Streams Jumbo Patch, revision update for 4.1.3 compatibility BugId's fixed with this patch: 1068158 1044100 1044888 1047355 1058097 1061174

Architectures for which this patch is available: Sun3 Sun3x Sun4 Sun4c Sun4m

Patches which may conflict with this patch: 100179 100233 100255 100279 100310 100189

IMPORTANT: This patch can be installed on a system with DBE1.1 or 1.2 installed. Just make sure that the dbe versions of the str_syscalls.o is used. This applies to each sparc architecture.

Obsoleted by:

This patch obsoletes patches: 100179 100233 100255 100279 100310 100189

Problem Description:

2/6/92 This version of the patch contains the additional binaries for sun4m architecture.

Bugid 1068158 System crash with assertion failure vp->v_stream when using clone driver.

Bugid 1044100 This patch incorporates the fix for bug id #1044100 which results in a machine freeze.

Bugid 1044888 This patch also fixes a bug where kernel printf hangs the system.

Bugid 1047355 Polling not working correctly.

Bugid 1058097 System crash with panic data fault, with bad buffer pointer, usually in the login process.

Bugid 1061174 'poll()' does not return -1 when 'poll()' receives a signal when polling fds greater than 30.

Bugid 1058097 (11/11/91 100359-03) The same system crash panic data fault can occur when the ioctl FIONREAD is being processed.

INSTALL:

As root, make a backup copy of the files to be patched:

mv /sys/`arch -k`/OBJ/str_io.o /sys/`arch -k`/OBJ/str_io.o_orig mv /sys/`arch -k`/OBJ/str_syscalls.o /sys/`arch -k`/OBJ/str_syscalls.o_orig Now install the patched files:

cp `arch -k`/str_io.o /sys/`arch -k`/OBJ cp `arch -k`/str_syscalls.o /sys/`arch -k`/OBJ

OR FOR SYSTEMS WITH DBE: cp `arch -k`/str_io.o /sys/`arch -k`/OBJ cp `arch -k`/str_syscalls.dbe.o /sys/`arch -k`/OBJ/str_syscalls.o

chmod 444 /sys/`arch -k`/OBJ/str_io.o chmod 444 /sys/`arch -k`/OBJ/str_syscalls.o

Then a new kernel will need to be made

Please refer to the System and Network Administration manual on building and installing a custom kernel. SHAR_EOF chmod +x '100359-06_README' fi if test -f '100372-02_README' then echo shar: "will not over_write existing file '100372-02_README'" else cat << \SHAR_EOF > '100372-02_README' Patch-ID# 100372-02 Keywords: tfs, security, C2 Synopsis: SunOS 4.1.1;4.1.2;4.1.3: tfs and c2 do not work together Date: 08-Sept-92 SunOS release: 4.1.1 4.1.2 4.1.3 Unbundled Product: Unbundled Release: Topic: Patch update revision for 4.1.2 and 4.1.3 compatibility BugId's fixed with this patch: 1052574

Architectures for which this patch is available: sun3 sun3x sun4 sun4c sun4m

Patches which may conflict with this patch:

Obsoleted by:

Problem Description: tfs doesn't work correctly with c2 security

INSTALL:

As root:

# mv /usr/etc/tfsd /usr/etc/tfsd.FCS # cp `arch -k`/{OS Release}/tfsd /usr/etc/tfsd # chmod 755 /usr/etc/tfsd # chown root.staff /usr/etc/tfsd

Reboot your machine. SHAR_EOF chmod +x '100372-02_README' fi if test -f '100383-05_README' then echo shar: "will not over_write existing file '100383-05_README'" else cat << \SHAR_EOF > '100383-05_README' Patch-ID# 100383-05 Keywords: security rdist setuid client remote distfile Synopsis: SunOS 4.0.3;4.1;4.1.1;4.1.2;4.1.3: rdist security enhancement Date: 10/Sep/92 SunOS release: 4.0.3, 4.1, 4.1.1, 4.1.2 4.1.3 Unbundled Product: Unbundled Release: Topic: Patch version update for 4.1.3 compatibility BugId's fixed with this patch: 1069497 1074961

Architectures for which this patch is available: application arch sun3, sun4

Patches which may conflict with this patch:

Obsoleted by:

Problem Description: /usr/ucb/rdist under some conditions can be forced to create setuid root programs thus causing a security problem

INSTALL: As root:

mv /usr/ucb/rdist /usr/ucb/rdist.FCS chmod 100 /usr/ucb/rdist.FCS cp `arch`/{4.0.3,4.1,4.1.1,4.1.2,4.1.3}/rdist /usr/ucb/rdist chmod 4751 /usr/ucb/rdist chown root.staff /usr/ucb/rdist

SHAR_EOF chmod +x '100383-05_README' fi if test -f '100507-04_README' then echo shar: "will not over_write existing file '100507-04_README'" else cat << \SHAR_EOF > '100507-04_README' Patch-ID# 100507-04 Keywords: tmpfs, crash, fail assertion, leaks, anonymous, tmp_rename, panic Synopsis: SunOS 4.1.1, 4.1.2, 4.1.3: tmpfs jumbo patch Date: 03/Sep/92 SunOS release: 4.1.1 4.1.2 4.1.3 Unbundled Product: Unbundled Release: Topic: fixes for several tmpfs bugs BugId's fixed with this patch: 1038651 1091294 1089447 1083412

Architectures for which this patch is available: sun4 sun4c sun4m

Obsoleted by: SVr4

Changes incorporated in this version: 1083412

Problem Description:

These files contain bug fixes for the following bugs:

1038651: Using tmpfs ie ram /tmp can cause data initialization with Fortran on SunOS 4.1. 1091294: tmpfs "leaks" anonymous memory 1089447: panic: tmp_rename 1083412: copying files from an nfs mounted partition to a tmpfs mount can result in security breach

INSTALL:

mv the new versions of tmp_dir.o, tmp_subr.o, tmp_tnode.o, tmp_vfsops.o, and tmp_vnodeops.o into /sys/($arch -k)/OBJ after first saving the old versions aside as a precaution.

As root:

mv /sys/`arch -k`/OBJ/tmp_dir.o /sys/`arch -k`/OBJ/tmp_dir.o.FCS mv /sys/`arch -k`/OBJ/tmp_subr.o /sys/`arch -k`/OBJ/tmp_subr.o.FCS mv /sys/`arch -k`/OBJ/tmp_tnode.o /sys/`arch -k`/OBJ/tmp_tnode.o.FCS mv /sys/`arch -k`/OBJ/tmp_vfsops.o /sys/`arch -k`/OBJ/tmp_vfsops.o.FCS mv /sys/`arch -k`/OBJ/tmp_vnodeops.o /sys/`arch -k`/OBJ/tmp_vnodeops.o.FCS mv `arch -k`/tmp_dir.o /sys/`arch -k`/OBJ/tmp_dir.o mv `arch -k`/tmp_subr.o /sys/`arch -k`/OBJ/tmp_subr.o mv `arch -k`/tmp_tnode.o /sys/`arch -k`/OBJ/tmp_tnode.o mv `arch -k`/tmp_vfsops.o /sys/`arch -k`/OBJ/tmp_vfsops.o mv `arch -k`/tmp_vnodeops.o /sys/`arch -k`/OBJ/tmp_vnodeops.o

Run /etc/config on your kernel and do a make. Install the newly configured vmunix and reboot. Refer to the Systems and Network administrations manual on how to configure a kernel. SHAR_EOF chmod +x '100507-04_README' fi if test -f '100527-03_README' then echo shar: "will not over_write existing file '100527-03_README'" else cat << \SHAR_EOF > '100527-03_README' Patch-ID# 100527-03 Keywords: 4.0, selects, old-style, rsh Synopsis: SunOS 4.1.1;4.1.2;4.1.3: rsh uses old-style selects instead of 4.0 selects Date: 5/6/92

SunOS release: 4.1.1;4.1.2;4.1.3

Unbundled Product:

Unbundled Release:

Topic: Patch revision update for 4.1.3 compatibility

BugId's fixed with this patch: 1017537, 1088518.

Architectures for which this patch is available: sun4, sun4c, sun4m

Patches which may conflict with this patch:

Obsoleted by:

Files included with this patch: /usr/ucb/rsh

Problem Description:

Bugid: 1017537 - -------------- If a process with a lot of open file descriptors exec's rsh, rsh's select() calls can get confused because they use the old 3.x-style select interface.

Bugid: 1088518 - -------------- An incorrect bug fix was made to patch 100527-01. This incorrect bug fix causes rsh to wrongly ignore standard input. This makes it impossible to do things like: $ cat /etc/termcap | rsh host wc -l

Install Instructions:

As root: cd to the appropriate release subdirectory of the patch:

# cd 4.1.[123]

save the existing /usr/ucb/rsh executable # mv /usr/ucb/rsh /usr/ucb/rsh.FCS

copy the new rsh for the appropriate architecture to /usr/ucb # cp `arch`/rsh /usr/ucb

set the permissions and setuid bit of the new rsh program # chmod 4755 /usr/ucb/rsh

SHAR_EOF chmod +x '100527-03_README' fi if test -f '100536-02_README' then echo shar: "will not over_write existing file '100536-02_README'" else cat << \SHAR_EOF > '100536-02_README' Patch-ID# 100536-02 Keywords: panic, assertion, failed, NFS, vm Synopsis: SunOS 4.1.2, 4.1.3: NFS can cause panic: assertion failed crashes Date: 18/Jun/92

SunOS release: 4.1.2 4.1.3

Unbundled Product:

Unbundled Release:

Topic: NFS patch revision update for 4.1.3 compatibility

BugId's fixed with this patch: 1066287

Architectures for which this patch is available: sun4, sun4c, sun4m

Patches which may conflict with this patch:

Obsoleted by:

Files included with this patch: nfs_vnodeops.o seg_vn.o

Problem Description:

The fix for bugid 1066287 was incorrect and could result in the above panic, this patch contains the correct fix for the above bugid and prevents the panic.

Install Instructions:

As root, make a backup copy of the files to be patched:

mv /sys/`arch -k`/OBJ/nfs_vnodeops.o /sys/`arch -k`/OBJ/nfs_vnodeops.o.orig mv /sys/`arch -k`/OBJ/seg_vn.o /sys/`arch -k`/OBJ/seg_vn.o.orig

Now install the patched files and set permissions:

cp `arch -k`/nfs_vnodeops.o /sys/`arch -k`/OBJ cp `arch -k`/seg_vn.o /sys/`arch -k`/OBJ chmod 444 /sys/`arch -k`/OBJ/seg_vn.o chmod 444 /sys/`arch -k`/OBJ/nfs_vnodeops.o

Config, make and install a new kernel.

Please refer to the system and networking administration manual for details on building and installing a new kernel SHAR_EOF chmod +x '100536-02_README' fi if test -f '100557-02_README' then echo shar: "will not over_write existing file '100557-02_README'" else cat << \SHAR_EOF > '100557-02_README' Patch-ID# 100557-02 Keywords: *, wildcard, mput, ftp Synopsis: SunOS 4.1.2, 4.1.3: ftp Jumbo patch Date: 18/Jun/92

SunOS release: 4.1.2;4.1.3

Unbundled Product:

Unbundled Release:

Topic: ftp Jumbo patch. Patch revision update for 4.1.3 compatibility

BugId's fixed with this patch: 1080677 1043669

Architectures for which this patch is available: sun4, sun4c

Patches which may conflict with this patch:

Obsoleted by:

Files included with this patch: ftp

Problem Description: - --------------------

Bugid: 1043669 - -------------- if you supply an account entry in the .netrc file it still prompts for the account.

Bugid: 1080677 - -------------- ftp with mput * does not always work. ftp indicates transfer is complete, and the file is created on the remote system but it is emtpy. This is an intermittent problem. The files are between 16k-70k, binary and ascii. He did specify binary mode for the binary files. It sometimes fails with a bad file number. However, a partial wildcard like mput l* works ok. This does not always happen in the same directory.

Install Instructions: - ---------------------

AS ROOT:

1. mv /usr/ucb/ftp /usr/ucb/ftp.fcs 2. cp sun4/ftp /usr/ucb

3. Use "chmod" and "chown" to set the file permissions of /usr/ucb/ftp to those below.

- -rwxr-xr-x 1 root staff 81920 Oct 2 15:54 /usr/ucb/ftp*

SHAR_EOF chmod +x '100557-02_README' fi if test -f '100567-03_README' then echo shar: "will not over_write existing file '100567-03_README'" else cat << \SHAR_EOF > '100567-03_README' Patch-ID# 100567-03 Keywords: icmp_error, m_free, panic, mbuf, mfreem Synopsis: SunOS 4.1.2, 4.1.3: mfree panic due to mbuf being freed twice Date: 6/Apr/92

SunOS Release: 4.1.2, 4.1.3

Topic: mfree patch - revision update for 4.1.3 compatibility

BugID's fixed with this patch: 1087460

Architectures for which this patch is available: sun4 sun4c sun4m(only 4.1.2)

Patches which may conflict with this patch:

Obsolete By:

Problem Description:

BUGID: 1087460

We free the same mbuf a second time. This causes the mfree panic.

INSTALL:

As root:

mv /sys/`arch -k`/OBJ/ip_icmp.o /sys/`arch -k`/OBJ/ip_icmp.o.fcs cp sun4/ip_icmp.o /sys/`arch -k`/OBJ

A New kernel will need to be made and installed. Please refer to the system and networking administration manual for details on building and installing a new kernel

SHAR_EOF chmod +x '100567-03_README' fi if test -f '100623-02_README' then echo shar: "will not over_write existing file '100623-02_README'" else cat << \SHAR_EOF > '100623-02_README' Patch-ID# 100623-02 Keywords: ufs_inactive syncip ufs zero length directories panic ifree inode bmap iget Synopsis: 4.1.2;4.1.3: UFS jumbo patch Date: 11/Sep/92

SunOS release: 4.1.2 4.1.3

Unbundled Product:

Unbundled Release:

Topic: Patch revision update for 4.1.3 compatibility

BugID's fixed with this patch: 1078521 1039693 1082206 1071839

Architectures for which this patch is available: sun4, sun4c, sun4m

Patches which may conflict with this patch: 100505-01 100548-01 100575-02

NOTE: ufs_bmap.o is modified in both this patch and 100575-02. This patch ufs_bmap.o patch has the 100575 ufs_bmap.o integrated and should always be the last installed of the two.

Obsoleted by:

Files included with this patch: ufs_bmap.o ufs_inode.o ufs_dir.o

Problem Description:

This patch includes fixes for all known ufs-related bugs, including some that can cause "panic: freeing free inode" and "panic: ialloc: dup alloc"

1078521 Zero length directories can be left when a system is powered off 1039693 panic: ifree: freeing free inode 1082206 bmap references block after calling brelse 1071839 iget shouldn't hammer i_flag when reclaiming an inode

Install Instructions: mv /sys/`arch -k`/OBJ/ufs_bmap.o /sys/`arch -k`/OBJ/ufs_bmap.o_orig mv /sys/`arch -k`/OBJ/ufs_inode.o /sys/`arch -k`/OBJ/ufs_inode.o_orig mv /sys/`arch -k`/OBJ/ufs_dir.o /sys/`arch -k`/OBJ/ufs_dir.o_orig

cp `arch -k`/{OS Release}/ufs_bmap.o /sys/`arch -k`/OBJ/ufs_bmap.o cp `arch -k`/{OS Release}/ufs_inode.o /sys/`arch -k`/OBJ/ufs_inode.o cp `arch -k`/{OS Release}/ufs_dir.o /sys/`arch -k`/OBJ/ufs_dir.o

Then, build, install, and boot a new kernel. Please refer to the System and Network Administration Manual. SHAR_EOF chmod +x '100623-02_README' fi if test -f '100726-01_README' then echo shar: "will not over_write existing file '100726-01_README'" else cat << \SHAR_EOF > '100726-01_README' Patch-ID# 100726-01 Keywords: large memory panic Synopsis: 4.1.3: large memory (.5G and up) Galaxy/Ross systems panic while booting Date: 26/Aug/92

SunOS release: 4.1.3

Unbundled Product:

Unbundled Release:

Topic: OBP context table must be reclaimed correctly back into availmemory

BugId's fixed with this patch: 1099405

Architectures for which this patch is available: sun4m

Patches which may conflict with this patch: 100542, 100743

Obsoleted by:

Files included with this patch: machdep.o

Problem Description:

Bugid: 1099405 - --------------- Galaxy/Ross systems configured with 768MB or 1GB of memory will not boot, but instead panic in various locations, depending on the kernel booted. Increasing maxusers to 225 doesn't seem to help. The configurations were constructed with 512MB of 16MB SIMMs on the system board and 256MB of 4MB SIMMs on the expansion memory boards.

Booting kadb on a Galaxy/Ross with 2.8 PROMs and 4.1.3 panics the system when 512MB of 16MB SIMMs are installed on the system board.

Install Instructions:

For SunOS4.1.3 - --------------- AS ROOT: 1. Save the FCS module aside.

mv /sys/`arch -k`/OBJ/machdep.o /sys/`arch -k`/OBJ/machdep.o.FCS

2. Copy the new module to /sys/ARCH/OBJ/

cp `arch -k`/{os version}/machdep.o /sys/`arch -k`/OBJ

4. Rebuild the kernel and reboot the system with the new kernel.

Refer to the System Instalation Manual for information about building and booting new kernels.

SHAR_EOF chmod +x '100726-01_README' fi exit 0



This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:50 CDT