9_Recommended patch cluster breaks ssh

Hi all,
I recently installed the latest 9_Recommended patch cluster and it has rendered ssh inoperable. I get the following error when I try to start ssh:
ld.so.1: sshd: fatal: libgss.so.1: open failed: No such file or directory
In doing some research, it seems that there is now a dependency on the SUNWgss packages. I either need to find a place where I can get this package from or remove the patch that has broken ssh.
Any ideas?

I have a case open as well. Here is the current response:
RFE for patch 113273-11
Take a look at this
But he did not give any approved workaround so I pinged him again and
await feedback.
For the "unable to initialize mechanism library
[usr/lib/gss/gl/mech_krb5.so]" (see bug 6392328)
The following workarounds were provided by PTS
and one of my customers confimred that workaound suggestion
number 1 worked for him
1) Turn off GSS-API support in ssh_config(4) and
sshd_config(4)
Add to /etc/ssh/ssh_config and /etc/ssh/sshd_config:
GSSAPIAuthentication=no
GSSAPIKeyExchange=no
2) Configure /etc/krb5/krb5.conf (even if Kerberos V is not in use) so
it contains no syntax errors.
attached krb5.conf was given to me by my backline support for you to
use.
All patches for SSH Solaris9
Solaris 9 Patches:
>>>
113273-10 SunOS 5.9: /usr/lib/ssh/sshd Patch
114356-06 SunOS 5.9: /usr/bin/ssh Patch
Solaris 9 Backport Patches:
Those patches are:
112908-24
117177-02
114356-07
113273-11
Based on previous patch dependencies you will want to confirm that
these previously released patches are also installed:
112908-01
112921-07
112922-02
112923-03
112924-01

Similar Messages

Management console problem after installing recommended patch cluster

Hello everyone.
I just got a new SUN 440 server, it didn't come with the solaris os installed and so I installed solaris 9 12/03 to it. The installation went fine and I could start the management console with no problems at all. Then I went to sunsolve.sun.com to download the latest recommended patch clusters for solaris 9 (this was yesterday Sept.-06-2006), I downloaded the file, checked it with md5sum, unzipped it, went to single user mode and ran the install_patch script. After the patch cluster finished installing I rebooted the server for the patches to take effect, I logged in, tried to run console management 2.1 and it starts OK but I cannot use any services it offers because it says (in the console events):
"Server Not Running: no solaris management console server was available on the specified server. Please ensure there is a Solaris Management Console server available on the specified host and that it is running"
When I click on "See Exception" I get (for a particular service that is... in this case com.sun.admin.fsmgr.client.VFsMgr - for managing file systems):
java.rmi.RemoteException: Server RMI is null:
at sun.com.management.viperimpl.client.ViperClient.lookupServer(ViperClient.java:376)and so on...
I already checked that the service is running on port 898 and it seems that there is no other app. using that same port. I also restarted the service using /etc/init.d.init stop/status/start but I get the same results.
I don't know solaris very much and don't know what else to do, please help.
Thanks in advance.

Well, I guess what you see isn't what you get.
Guess I'm used to the fact USENET just left things formatted the way they were :-O

How to backout a Recommended Patch cluster deployment in Solstice disksuite

Hello Admin's,
I am planning to use the following Plan of action for deploying the latest Solaris 8 Recommended Patch cluster on the production server's i support.My
concern is if the patching activity fails or the application and Oracle databases dont come up after deploying the patch cluster.How do i revert back to system to its original state using the submirror's which i have detached prior to patching the system :
1) Will shutdown the applications and the databases on the server:
2) Will capture the output of the following commands :
df -k
ifconfig -a
contents of the files /etc/passwd /etc/shadow /etc/vfstab /etc/system
metastat -p
netstat -rn
prtvtoc /dev/rdsk/c1t0d0s0
prtvtoc /dev/rdsk/c1t1d0s0
3) We bring the system to the ok prompt
4) We will try to boot the system from both the disk's which are part of the d10 metadevice for root filesystem
=======================================================================================
user1@myserver>pwd ; df -k / ; ls -lt | egrep '(c1t0d0s0|c1t1d0s0)' ; prtconf -vp | grep bootpath ; metastat d10
/dev/dsk
Filesystem kbytes used avail capacity Mounted on
/dev/md/dsk/d10 8258597 3435895 4740117 43% /
lrwxrwxrwx 1 root root 43 Jul 28 2003 c1t0d0s0 -> ../../devices/pci@1c,600000/scsi@2/sd@0,0:a
lrwxrwxrwx 1 root root 43 Jul 28 2003 c1t1d0s0 -> ../../devices/pci@1c,600000/scsi@2/sd@1,0:a
bootpath: '/pci@1c,600000/scsi@2/disk@0,0:a'
d10: Mirror
Submirror 0: d11
State: Okay
Submirror 1: d12
State: Okay
Pass: 1
Read option: roundrobin (default)
Write option: parallel (default)
Size: 16779312 blocks
d11: Submirror of d10
State: Okay
Size: 16779312 blocks
Stripe 0:
Device Start Block Dbase State Hot Spare
c1t0d0s0 0 No Okay
d12: Submirror of d10
State: Okay
Size: 16779312 blocks
Stripe 0:
Device Start Block Dbase State Hot Spare
c1t1d0s0 0 No Okay
user1@myserver>
===================================================================================
ok nvalias backup_root <disk path>
Redefine the boot-device variable to reference both the primary and secondary submirrors, in the order in which you want to access them. For example:
ok printenv boot-device
boot-device= disk net
ok setenv boot-device disk backup_root net
boot-device= disk backup_root net
In the event of primary root disk failure, the system automatically boots from the secondary submirror. To test the secondary submirror, boot the system manually, as follows:
ok boot backup_root
user1@myserver>metadb -i
flags first blk block count
a m p luo 16 1034 /dev/dsk/c1t0d0s7
a p luo 1050 1034 /dev/dsk/c1t0d0s7
a p luo 2084 1034 /dev/dsk/c1t0d0s7
a p luo 16 1034 /dev/dsk/c1t1d0s7
a p luo 1050 1034 /dev/dsk/c1t1d0s7
a p luo 2084 1034 /dev/dsk/c1t1d0s7
o - replica active prior to last mddb configuration change
u - replica is up to date
l - locator for this replica was read successfully
c - replica's location was in /etc/lvm/mddb.cf
p - replica's location was patched in kernel
m - replica is master, this is replica selected as input
W - replica has device write errors
a - replica is active, commits are occurring to this replica
M - replica had problem with master blocks
D - replica had problem with data blocks
F - replica had format problems
S - replica is too small to hold current data base
R - replica had device read errors
user1@myserver>df -k
Filesystem kbytes used avail capacity Mounted on
/dev/md/dsk/d10 8258597 3435896 4740116 43% /
/dev/md/dsk/d40 2053605 929873 1062124 47% /usr
/proc 0 0 0 0% /proc
fd 0 0 0 0% /dev/fd
mnttab 0 0 0 0% /etc/mnttab
/dev/md/dsk/d30 2053605 937231 1054766 48% /var
swap 2606008 24 2605984 1% /var/run
swap 6102504 3496520 2605984 58% /tmp
/dev/md/dsk/d60 13318206 8936244 4248780 68% /u01
/dev/md/dsk/d50 5161437 2916925 2192898 58% /opt
user1@myserver>metastat d40
d40: Mirror
Submirror 0: d41
State: Okay
Submirror 1: d42
State: Okay
Pass: 1
Read option: roundrobin (default)
Write option: parallel (default)
Size: 4194828 blocks
d41: Submirror of d40
State: Okay
Size: 4194828 blocks
Stripe 0:
Device Start Block Dbase State Hot Spare
c1t0d0s4 0 No Okay
d42: Submirror of d40
State: Okay
Size: 4194828 blocks
Stripe 0:
Device Start Block Dbase State Hot Spare
c1t1d0s4 0 No Okay
user1@myserver>metastat d30
d30: Mirror
Submirror 0: d31
State: Okay
Submirror 1: d32
State: Okay
Pass: 1
Read option: roundrobin (default)
Write option: parallel (default)
Size: 4194828 blocks
d31: Submirror of d30
State: Okay
Size: 4194828 blocks
Stripe 0:
Device Start Block Dbase State Hot Spare
c1t0d0s3 0 No Okay
d32: Submirror of d30
State: Okay
Size: 4194828 blocks
Stripe 0:
Device Start Block Dbase State Hot Spare
c1t1d0s3 0 No Okay
user1@myserver>metastat d50
d50: Mirror
Submirror 0: d51
State: Okay
Submirror 1: d52
State: Okay
Pass: 1
Read option: roundrobin (default)
Write option: parallel (default)
Size: 10487070 blocks
d51: Submirror of d50
State: Okay
Size: 10487070 blocks
Stripe 0:
Device Start Block Dbase State Hot Spare
c1t0d0s5 0 No Okay
d52: Submirror of d50
State: Okay
Size: 10487070 blocks
Stripe 0:
Device Start Block Dbase State Hot Spare
c1t1d0s5 0 No Okay
user1@myserver>metastat -p
d10 -m d11 d12 1
d11 1 1 c1t0d0s0
d12 1 1 c1t1d0s0
d20 -m d21 d22 1
d21 1 1 c1t0d0s1
d22 1 1 c1t1d0s1
d30 -m d31 d32 1
d31 1 1 c1t0d0s3
d32 1 1 c1t1d0s3
d40 -m d41 d42 1
d41 1 1 c1t0d0s4
d42 1 1 c1t1d0s4
d50 -m d51 d52 1
d51 1 1 c1t0d0s5
d52 1 1 c1t1d0s5
d60 -m d61 d62 1
d61 1 1 c1t0d0s6
d62 1 1 c1t1d0s6
user1@myserver>pkginfo -l SUNWmdg
PKGINST: SUNWmdg
NAME: Solstice DiskSuite Tool
CATEGORY: system
ARCH: sparc
VERSION: 4.2.1,REV=1999.11.04.18.29
BASEDIR: /
VENDOR: Sun Microsystems, Inc.
DESC: Solstice DiskSuite Tool
PSTAMP: 11/04/99-18:32:06
INSTDATE: Apr 16 2004 11:10
VSTOCK: 258-6252-11
HOTLINE: Please contact your local service provider
STATUS: completely installed
FILES: 150 installed pathnames
6 shared pathnames
19 directories
1 executables
7327 blocks used (approx)
user1@myserver>
=======================================================================================
5) After successfully testing the above we will bring the system to the single user mode
# reboot -- -s
6) Detach the following sub-mirrors :
# metadetach -f d10 d12
#metadetach -f d30 d32
#metadetach -f d40 d42
#metadetach -f d50 d52
# metastat =====> (to check if the submirror�s are successfully detached)
7) Applying patch on the server
After patch installation is complete will be rebooting the server to single user mode
# reboot -- -s
confirming if patch installation was successful (uname �a) .
8) Will be booting the server to the multiuser mode (init 3) and confirming with the database and the application teams if the
Applications/databases are working fine .Once confirmed successful will be reattaching the d12 submirror.
# metattach d12 d10
# metattach d30 d32
#metattach d40 d42
# metattach d50 d52
# metastat d10 =====> (to check the submirror is successfully reattached)
user1@myserver>uname -a ; cat /etc/release ; date
SunOS myserver 5.8 Generic_117350-04 sun4u sparc SUNW,Sun-Fire-V210
Solaris 8 HW 12/02 s28s_hw1wos_06a SPARC
Copyright 2002 Sun Microsystems, Inc. All Rights Reserved.
Assembled 12 December 2002
Mon Apr 14 17:10:09 BST 2008
-----------------------------------------------------------------------------------------------------------------------------

Recommended patch sets are and to the best of my knowledge have always been regenerated twice a month.
I think your thinking of maintenance releases. When they generate a new CD image which can be used to do an upgrade install.
They try to generate those every 6 months, but the schedule often slips.
The two most recent were sol10 11/06 and sol10 8/07.

Patch Install Breaks TCP/IP Connectivity

Today (8/29/02) I installed the latest patch cluster for Solaris 2.8 (Sparc). Ever since I installed it, I have been unable to telnet or establish a socket connection from my Solaris system (port 10014) to a particular TCP/IP device. This is not a computer but a satellite reciever developed to allow a TCP/IP socket connections that receive data from the satellite feed.
I have a second Solaris 2.8 (Sparc) box which I have not applied the latest pactch cluster to yet, and it can connect fine.
When I telnet to the device, Telnet gets hung at the "Trying..." prompt. I checked netstat -a and the state of the connection is SYN_SENT.
I used "snoop" to monitor the connection and the patched Solaris box is sending connection requests but receiving no responses from the device. The unpatched Solaris box is getting a response from the device. Both boxes appear to be sending the same information to the device.
The both Solaris boxes can ping the device.
The use to work on the pached Solaris box, before I applied the patches.
Both Solaris systems and the device are on the same network. No routers or external firewalls separate them (in fact they are connected to the same switch).
1. Does anybody have any idea why this is happening? Is there a configuration that I am missing? (I search through the new files on the patched Solaris box looking for one).
2. Does anybody have a way to uninstall the entire patche set? There does not appear to be an "uninstall_cluster" script in the zip file.

I think I have found what appears to be the problem (although I am not sure how to resolve it).
I compiled and installed tcpdump on both Solaris boxes. I then wrote a little perl script which does the following:
#!/usr/bin/perl
use Socket;
socket( SH, PF_INET, SOCK_STREAM, getprotobyname( 'tcp')) || die $!;
my $dest = sockaddr_in(10014, inet_aton( "147.81.85.155"));
connect( SH, $dest) || die $!;
send( SH, "\r\n", 0);
while(1)
recv( SH, $buffer, 128, 0 );
print "$buffer";
(simply connects to socket and writes out bytes read)
I then ran "tcpdump -vvv -l host 147.81.85.155" and got the following output:
11:50:30.496927 nattyboh.32942 > hcsn-monitor6.wsicorp.com.10014: S [bad tcp cksum 2175!] 2943092708:2943092708(0) win 24820 <nop,nop,sackOK,mss 460> (DF) (ttl 64, id 41884, len 48)
11:50:33.860142 nattyboh.32942 > hcsn-monitor6.wsicorp.com.10014: S [bad tcp cksum 2175!] 2943092708:2943092708(0) win 24820 <nop,nop,sackOK,mss 460> (DF) (ttl 64, id 41885, len 48)
11:50:40.610137 nattyboh.32942 > hcsn-monitor6.wsicorp.com.10014: S [bad tcp cksum 2175!] 2943092708:2943092708(0) win 24820 <nop,nop,sackOK,mss 460> (DF) (ttl 64, id 41886, len 48)
11:50:54.110248 arp who-has hcsn-monitor6.wsicorp.com (Broadcast) tell nattyboh
11:50:54.110661 arp reply hcsn-monitor6.wsicorp.com is-at 0:a0:e1:98:19:e3
11:50:54.110691 nattyboh.32942 > hcsn-monitor6.wsicorp.com.10014: S [bad tcp cksum 2175!] 2943092708:2943092708(0) win 24820 <nop,nop,sackOK,mss 460> (DF) (ttl 64, id 54229, len 48)
11:51:21.110163 nattyboh.32942 > hcsn-monitor6.wsicorp.com.10014: S [bad tcp cksum 2175!] 2943092708:2943092708(0) win 24820 <nop,nop,sackOK,mss 460> (DF) (ttl 64, id 54230, len 48)
Note the "bad tcp cksum" message. Also that the device is responding to an arp request from the Solaris box. When I ran the test script and tcpdump on the unpached solaris system, the tcpdump looks like the following:
11:51:57.435625 arp who-has hcsn-monitor6.wsicorp.com (Broadcast) tell tsingtao
11:51:57.437165 arp reply hcsn-monitor6.wsicorp.com is-at 0:a0:e1:98:19:e3
11:51:57.437205 tsingtao.51997 > hcsn-monitor6.wsicorp.com.10014: S [tcp sum ok] 3544242549:3544242549(0) win 24820 <nop,nop,sackOK,mss 1460> (DF) (ttl 64, id 10901, len 48)
11:51:57.464533 hcsn-monitor6.wsicorp.com.10014 > tsingtao.51997: S [tcp sum ok] 772727198:772727198(0) ack 3544242550 win 4096 <mss 1460> (DF) (ttl 64, id 10904, len 44)
11:51:57.464577 tsingtao.51997 > hcsn-monitor6.wsicorp.com.10014: . [tcp sum ok] 1:1(0) ack 1 win 24820 (DF) (ttl 64, id 10902, len 40)
11:51:57.464926 tsingtao.51997 > hcsn-monitor6.wsicorp.com.10014: P [tcp sum ok] 1:3(2) ack 1 win 24820 (DF) (ttl 64, id 10903, len 42)
11:51:57.466583 hcsn-monitor6.wsicorp.com.10014 > tsingtao.51997: . [tcp sum ok] 1:1(0) ack 3 win 4096 (DF) (ttl 64, id 10906, len 40)
Which is what I would expect.
Any suggestions on how to resolve this problem?

Non priviledged commands fail after patch cluster install

For several years, to allow users to access system functions for higher clock resolutions in their
applications, we have made the following mods to two files (per Sun recommendations):
In the /etc/system file, add lines:
* 10 ms clock resolution*
set higher_tick = 1*
In /etc/user_attr, add:
dssuser::::type=normal;defaultpriv=proc_clock_highres,file_link_any,proc_info,proc_session,proc_fork,proc_exec
Everything was fine until I installed the latest patch cluster (from 2-14-2011) on our Netra 210. No issues installing the cluster. Upon reboot, the dssuser cannot do some simple things, like use ftp or ping. Messages like this:
ping: unknown host
ftp: socket: Permission denied.
The root user has no problems using any of these commands. As soon as I comment out the dssuser line from the /etc/user_attr file, everything works as expected. But we need the highres clock settings for our applications. I am unfamiliar with the workings and intent of the entries in user_attr so any help would be most appreciated.
Mark

The problem is that after patch implementation, the special privileges for any user defined in the /etc/user_attr file were modified to append the string "!net_privaddr" specifically denying use of common commands ping and ftp for example. The output of ppriv:
$ ppriv $$
14870: -sh
flags = <none>
E: basic,proc_clock_highres,!net_privaddr
I: basic,proc_clock_highres,!net_privaddr
P: basic,proc_clock_highres,!net_privaddr
L: all
I had to add the keyword net_privaddr in the user_attr file like:
dssuser::::type=normal;defaultpriv=proc_clock_highres,file_link_any,proc_info,proc_session,proc_fork,proc_exec,net_privaddr
to get past this problem. A bug or what?

Solaris 2.6 Patch cluster

Is there anyone out there knows where I can get the last Solaris 2.6 patch cluster? I know it's de-supported now but sadly some legacy applications are still out there and need to run on old software/hardware.
Many thanks in advance,
Jonathan

if you run a pstack against the core, I might
recognize the failure. Usually, logging is of no
use, as we don't log a crash. . .Thanks. I'll do that. It just generated a new core file.
>
I don't know of any problems with installing 4.15p7,
nor of installing later hotfixes on top of it, and
I've been supporting Messaging Server since 4.0 came
out. . . .
If you've had file permission issues since 4.05,
that's truly news to me, unless it was that pesky
problem with Admin Server, where it would change the
permissions of the logs when it went to roll them
over. I don't remember when we fixed that one, but
we did..We've had file permission issues with the patches since 4.0. When we'd upgrade some of the files would be owned by root and not exec'able by nobody. Some of the patches would fail when the patch process would try to bind to the directory (I don't know why). We've used NMS since the 3.x version. It has also been a long time since we tried to patch to app because our experience was bad. It was stable for a long time (once we disabled shared folders and resolved the nscd issue).

Solaris 8 patches added to Solaris 9 patch cluster and vice versa

Has anyone noticed this? On the Solaris 8 and 9 patch cluster readmes, it shows sol 9 patches have been added to the sol 8 cluster and sol 8 patches have been added to the sol 9 cluster. what's the deal? I haven't found any information about whether or not this is a mistake.

Desiree,
Solaris 9's kernel patch 112233-12 was the last revision for that particular patch number. The individual zipfile became so large that it was subsequently supplanted by 117191-xx and that has also been supplanted when its zipfile became very large, by 118558-xx.
Consequently you will never see any newer version than 112233-12 on that particular patch.
What does <b>uname -a</b> show you for that system?
Solaris 8 SPARC was similarly effected, for 108528, 117000 and 117350 kernel patches.
If you have login privileges to Sunsolve, find <font color="teal">Infodoc 76028</font>.

Solaris 8 x86 patch cluster installation

hi. i recently installed solaris 8 x86 on my intel p4 machine. i got through the installation fine but i am now running into problems with installing the 8x86 recommended patch cluster.
whenever i run the install_cluster script, it says every patch failed to install due to return code 1. i even tried manually installing the patches manually via patchadd...but i end up getting a message saying the patch directory is not valid.
i basically downloaded the 8x86 recommended patch cluster zip file on a windows 2k machine. i unzip the file and then burn the patches onto a cd. i then copy the patches from the cd onto my solaris machine and try to install the patches that way. so far...this doesn't work..and i dont know what im doing wrong.
does anyone know how to fix this problem? thx.

hi again. i just fixed my problem. apparently when i unzipped the file and then burned it onto cd...the data got corrupted. i fixed it by copying the zip file onto cd and then extracting the patches onto the solaris machine.

Latest patch cluster idated march 14 nstalled on solaris10

I installed the latest patch cluster on my solaris10 system and afterwards the Solaris Management console would not work. I kept getting the error that the server was not running. I followed the limited trouble-shooting guidelines for stopping and starting the service but still did not work. After reviwing the patch cluster I discoved there was a patch that updated smc. There was no special instructions for this patch. I removed the patch and the smc worked find. Has anyone had any problems with patch 121308-12.

Patch 119313-19(SPARC), 119314-20(x86) will resolve your problem.
But these patch are NOT included in the recommanded pathces.
You need to join Sun's SunSolve or Sun's standard support.

August Patch Cluster Problems

Has anyone had the following issue after installing the latest Patch Cluster?
After a reboot I get
couldn't set locale correctly
To correct this I have to edit /etc/default/init
and remove
LC_COLLATE=en_GB.ISO8859-1
LC_CTYPE=en_GB.ISO8859-1
LC_MESSAGES=C
LC_MONETARY=en_GB.ISO8859-1
LC_NUMERIC=en_GB.ISO8859-1
LC_TIME=en_GB.ISO8859-1
If I then create a flash archive and use this flash archive the jumpstart process then puts the locale info back and the problem appears again.
It's not critical as I don't need to be on the latest Patch Cluster but would wondered if I'm the only one having issues?

If you open the directory in CDE's file manager, right click on the zipped file and select unzip. The cluster will be unzipped to a directory structure called x86_recommended or something of the sort. Change to that directory to run the patch cluster install script. The patch script is looking for that directory structure.
Lee

Patch cluster 3/26/10 will hose X4500 disk management task in multiuser

I just want to warn user about the potential problem that you will encounter if you have a SUN X4500 system.
After applying patch cluster 3/26/10 on a SUN X4500 (thumper), now I can not do many disk management tasks in multiuser mode, such as
format
zpool create
and possibly many more commands (involving disk devices) would just hang.
Drop it to single user mode, however, you CAN still do these management task. I had to create zpool in single user mode
zfs create pool/zfsset still works in multiuser mode.
Thank you for your attention.

I'm having exactly the same problem on an x4500 server running OpenSolaris snv_127. We suffered a disk failure (c7t0d0), so I resilvered one of the pools onto a spare drive, which succeeded:
# zpool history data
<snip>
2010-04-20.14:45:08 zpool offline data c7t0d0
2010-04-20.14:58:19 zpool replace -f data c7t0d0 c7t1d0
# zpool status data
pool: data
state: ONLINE
scrub: resilver completed after 43h58m with 0 errors on Thu Apr 22 10:56:50 2010
config:
NAME STATE READ WRITE CKSUM
data ONLINE 0 0 0
raidz2-0 ONLINE 0 0 0
c11t1d0 ONLINE 0 0 0
c10t1d0 ONLINE 0 0 0
c13t1d0 ONLINE 0 0 0
c12t0d0 ONLINE 0 0 0
c8t1d0 ONLINE 0 0 0
c12t1d0 ONLINE 0 0 0
c10t0d0 ONLINE 0 0 0
c10t7d0 ONLINE 0 0 0
c12t2d0 ONLINE 0 0 0
c8t0d0 ONLINE 0 0 0
c7t1d0 ONLINE 0 0 0 381G resilvered
As you can see, the drive c7t0d0 has been successfully replaced and no longer appears in any zpool:
# zpool status | grep c7t0d0
# iostat -En | grep c7t0d0
c7t0d0 Soft Errors: 2 Hard Errors: 4 Transport Errors: 0
# cfgadm -l | grep c7t0d0
sata1/0::dsk/c7t0d0 disk connected configured ok
However when I attempt to unconfigure the sata port in preperation for replacing the drive, it doesn't work:
# cfgadm -c unconfigure sata1/0
Unconfigure the device at: /devices/pci@0,0/pci1022,7458@1/pci11ab,11ab@1:0
This operation will suspend activity on the SATA device
Continue (yes/no)? yes
cfgadm: Hardware specific failure: Failed to unconfig device at ap_id: /devices/pci@0,0/pci1022,7458@1/pci11ab,11ab@1:0
Also just pulling out the drive doesn't work, you end up with an unusable unconfigured port:
# cfgadm -l | grep unusable
sata6/0 disk connected unconfigured unusable
It's highly frustrating, because it means to replace disks we'll have to shut this server down. Not ideal!
I'm fairly certain this used to work fine.
If anyone has a solution, I'm all ears. I'm going to update this box to snv_130 soon (mainly to get the new zfs resilvering code) but I'm not hopeful about this bug/issue.
Regards,
Alasdair

GLOBAL ZONE stuck in shutting_down after applying latest patch cluster

Hi,
after installing the latest patch cluster, my zones on the system are not accessable
root@typhoon # zoneadm list -civ
ID NAME STATUS PATH BRAND IP
0 global shutting_down / native shared
- typhoon-log installed /local/zone/typhoon-log native shared
- typhoon-ftp installed /local/zone/typhoon-ftp native shared
- typhoon-rndrepos installed /local/zone/typhoon-rndrepos native shared
- typhoon-drpreview installed /local/zone/typhoon-test native shared
- typhoon-ossec installed /local/zone/typhoon-ossec native shared
- typhoon-drjboss installed /local/zone/typhoon-drjboss native shared
- typhoon-webdata installed /datapool/web/zones/webdata native shared
- typhoon-webzone installed /datapool/web/zones/typhoon-webzone native shared
# uname -a
SunOS typhoon 5.10 Generic_139556-08 i86pc i386 i86pc
there is one service reporting in maintenane
# svcs -xv
svc:/system/filesystem/volfs:default (Volume Management filesystem)
State: maintenance since Wed Jul 15 20:04:07 2009
Reason: Restarting too quickly.
See: http://sun.com/msg/SMF-8000-L5
See: man -M /usr/man -s 7FS volfs
See: /var/svc/log/system-filesystem-volfs:default.log
Impact: This service is not running.
I tried stopping and starting the vold, but without any effect
the / is on a zfs pool
# zpool list
NAME SIZE USED AVAIL CAP HEALTH ALTROOT
datapool 576G 105G 471G 18% ONLINE -
datapool2 148G 111G 37.0G 74% ONLINE -
rpool 120G 37.1G 82.9G 30% ONLINE -
root@typhoon # zpool status
pool: datapool
state: ONLINE
scrub: scrub completed after 0h31m with 0 errors on Thu Jul 16 02:31:53 2009
config:
NAME STATE READ WRITE CKSUM
datapool ONLINE 0 0 0
mirror ONLINE 0 0 0
c2d1s3 ONLINE 0 0 0
c1d1s3 ONLINE 0 0 0
errors: No known data errors
pool: datapool2
state: ONLINE
scrub: scrub completed after 1h53m with 0 errors on Wed Jul 15 23:53:06 2009
config:
NAME STATE READ WRITE CKSUM
datapool2 ONLINE 0 0 0
mirror ONLINE 0 0 0
c2d0s0 ONLINE 0 0 0
c1d0s3 ONLINE 0 0 0
errors: No known data errors
pool: rpool
state: ONLINE
scrub: scrub completed after 0h26m with 0 errors on Thu Jul 16 06:20:39 2009
config:
NAME STATE READ WRITE CKSUM
rpool ONLINE 0 0 0
mirror ONLINE 0 0 0
c1d1s0 ONLINE 0 0 0
c2d1s0 ONLINE 0 0 0
errors: No known data errors
root@typhoon #
# tail /var/svc/log/system-filesystem-volfs:default.log
[ Jul 16 08:03:11 Method "start" exited with status 0 ]
Thu Jul 16 08:03:11 2009 fatal: mounting of "/vol" failed
[ Jul 16 08:03:11 Stopping because all processes in service exited. ]
[ Jul 16 08:03:11 Executing stop method (:kill) ]
[ Jul 16 08:03:11 Executing start method ("/lib/svc/method/svc-volfs start") ]
[ Jul 16 08:03:11 Method "start" exited with status 0 ]
Thu Jul 16 08:03:11 2009 fatal: mounting of "/vol" failed
[ Jul 16 08:03:11 Stopping because all processes in service exited. ]
[ Jul 16 08:03:11 Executing stop method (:kill) ]
[ Jul 16 08:03:11 Restarting too quickly, changing state to maintenance ]
please advise, (this was on test run before implemention this patch cluster on the production server).
And I need to get a solution, to get my zone back running and then decide what to do with this patch cluster
thx
Edited by: Denis@SDN on Jul 16, 2009 1:06 AM

Sound familiar?
[http://opensolaris.org/jive/thread.jspa?threadID=105001&tstart=0]
This guy killed a process as workaround
[http://alittlestupid.com/2009/07/04/solaris-zone-stuck-in-shutting_down-state/]
We patched some SPARC systems recently and no issues though that's little consolation to you x86 admins.

SPARC10 not booting after 2.5.1 Recommended Patch Cluster Install

Hi everyone,
Just finished installing the "latest" patch cluster on my ancient SPARC10, and now it doesn't boot. Here's what I see:
ok boot /iommu/sbus/espdma@f,400000/esp@f,800000/sd@0,0
screen not found.
Can't open input device.
Keyboard not present. Using tty for input and output.
SPARCstation 10 (1 X 390Z55), No Keyboard
ROM Rev. 2.25, 112 MB memory installed, Serial #6332710.
Ethernet address 8:0:20:12:7f:99, Host ID: 7260a126.
Rebooting with command: /iommu/sbus/espdma@f,400000/esp@f,800000/sd@0,0
Boot device: /iommu/sbus/espdma@f,400000/esp@f,800000/sd@0,0 File and args:
not found: abort_enable
not found: abort_enable
not found: tod_validate
not found: tod_fault_reset
krtld: error during initial load/link phase
Memory Address not Aligned
Type help for more information
ok
At first I thought I had a RAM problem. I've swapped RAM, but that didn't make it boot. I then thought I had a system board problem. I've swapped that as well, and still no dice. I wish I'd have kept the list of patches that successfully installed like my gut told me to, but I didn't. Searching Google and the Sun websites didn't turn up any fixes to the problem, so I'm hoping I can find help here.
Did the patches clobber something the machine needs to boot? I believe the OS disks are mirrored, and I've tried booting off each submirror, but the result is the same.
If I can avoid booting off of CD-ROM, that would be ideal since I don't have an external CD-ROM available to boot from.
Thanks for any help!

Ouch !
Haven't seen that one in years !
It's your choice of that ancient Operating Environment, combined with a cpu faster than 419MHz.
I hope you still have the extra OECD disk that was packed with that system when it was first shipped from Sun.
Otherwise you may just be out of luck with using Solaris 2.5.1
Here's a link at a non-Sun 3rd party web site to SRDB document *20576*
http://www.sunshack.org/data/sh/2.1/infoserver.central/data/syshbk/collections/intsrdb/20576.html
Perhaps MAAL and haroldb can contribute their thoughts on this.

MB.ACT light OFF --- after Solaris 9 recommended patch cluster applied

After applying the recommended patch cluster to several SunFire V210s/V240s the "system ready" or "activity" light on the front panel no longer illuminates. The systems come up fine and are fully functional, just no light.
While this has absolutely no impact on usability it is a bit disconcerting for the folks in the server room when glancing at the rack and not seeing the typical green light.
I first noticed the problem in mid 2007 and have tried patch clusters as recent as last month, but the light remains out.
One interesting discovery is that if I reset (hard) the ALOM module the light comes back on. However, if the system is subsequently rebooted for any reason the light does not come back on...
If anyone can shed some light on the issue I would greatly appreciate it!
-Sap

continued.
I removed patch 138217-01 (svccfg & svcprop patch) and re-applied it.
Then, after booting into S-mode, I tried to continue with the broken patch 139555-08 (kernel).
It ends with
! root@jumpstart:/var/spool/patch # patchadd 139555-08
Validating patches...
Loading patches installed on the system...
Done!
Loading patches requested to install.
Preparing checklist for non-global zone check...
Checking non-global zones...
This patch passes the non-global zone check.
139555-08
Summary for zones:
Zone oem
Rejected patches:
None.
Patches that passed the dependency check:
139555-08
Patching global zone
Adding patches...
Checking installed patches...
Executing prepatch script...
Installing patch packages...
Pkgadd failed. See /var/tmp/139555-08.log.21103 for details
Patchadd is terminating.
! root@jumpstart:/var/spool/patch #the logfile shows
! root@jumpstart:/var/spool/patch # vi /var/tmp/139555-08.log.21103
Dryrun complete.
No changes were made to the system.
This appears to be an attempt to install the same architecture and
version of a package which is already installed. This installation
will attempt to overwrite this package.
pkgadd: ERROR: couldn't lock in /var/run/.patchSafeMode/root/var/sadm/install (s
erver running?): Resource temporarily unavailable
Installation of <SUNWarc> failed (internal error).
No changes were made to the system.Help!
-- Nick

Solaris reboot loop after recommended patch cluster

Hi,
I have a problem with my solaris 10 x86 install on my intel notebook. after installing the latest recommended patch cluster and the necessary reboot my latop stays in a loop.
one of the last lines I was able to see:
panic[cpu0]/thread=fec1d660: boot_mapin():No pp for pfnum=20fff
fec33b4c genunix:main+1b ()
skipping system dump - ....
rebooting...
any ideas ?
cu+thx

This horrible patch caught me out too. I found this and it fixed my problem.
http://weblog.erenkrantz.com/weblog/software/solaris
<snip>
[...at the Solaris boot prompt; enable kmdb, debugging, and single-user
so that you can remove the patch and reboot...]
boot -kds
[...wait for it to boot...]
physmax:w
:c
[...you'll see 'stop on write of'...]
physmax/X
[...you'll see something like the following line:
physmax: bff7f
this is a hex number; add one; so if you see bff7f,
your next line will need to be bff80...]
physxmax/W bff80
:c
[...system will boot and go into single user mode...
now, go toss those patches...]
patchrm 118844-19 120662-03 118345-12 118376-04 \
118565-03 118886-01 119076-10 118813-01 \
118881-02 120082-07 119851-02
shutdown -i6 -y -g0 "sun should test their patches"
</snip>

9_Recommended patch cluster breaks ssh

Similar Messages

Maybe you are looking for