From: Sree (sreep@qnetstaff.com)
Date: Mon Aug 26 2002 - 06:01:22 EDT
Dear Managers,
I need to stop oracle users + few of the dba users from directly connecting
(login) to the server, instead they should login using their personal
accounts and then su to user oracle
for example administrators can only su to root (only if the option in
/etc/default/login Console is unhashed) , i need to implement the same for
oracle and dba user
can any body help me to acheive this task.
Thanks & B.Rgds
Sree
----- Original Message -----
From: <sunmanagers-request@sunmanagers.org>
To: <sunmanagers@sunmanagers.org>
Sent: Monday, August 26, 2002 10:20 AM
Subject: sunmanagers digest, Vol 1 #1566 - 8 msgs
> Send sunmanagers mailing list submissions to
> sunmanagers@sunmanagers.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
> or, via email, send a message with subject or body 'help' to
> sunmanagers-request@sunmanagers.org
>
> You can reach the person managing the list at
> sunmanagers-admin@sunmanagers.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of sunmanagers digest..."
>
>
> Today's Topics:
>
> 1. SUMMARY: After your opinion: VxFs v ufs+logging & Disk-suite
(Miller, Anthony, A, Tech Dev, VF UK)
> 2. rootdisk mirroring with Disksuit. help. (Mt Dew)
> 3. need help with semaphore (Trinh, Linh T. [C])
> 4. question: need urgent help (raoul s)
> 5. FTP server (sridhara pk)
> 6. SUMMARY (UPDATED): "The IDPROM contents are invalid" (Slamet Ahmad
Fadilah)
> 7. SUNPCI w/ Solaris 9 (Very Pinguin)
> 8. SUMMERY: NIS port number (Moshe Leder)
>
> --__--__--
>
> Message: 1
> Subject: SUMMARY: After your opinion: VxFs v ufs+logging & Disk-suite
> Date: Thu, 22 Aug 2002 08:25:43 +0100
> From: "Miller, Anthony, A, Tech Dev, VF UK"
> <Anthony.Miller@gb.vodafone.co.uk>
> To: <sunmanagers@sunmanagers.org>
> Cc: "Miller, Anthony, A, Tech Dev, VF UK"
> <Anthony.Miller@gb.vodafone.co.uk>, "Craig Henry"
> <craig.henry@vf.vodafone.co.uk>, "Matthew Roberts"
> <Matthew.Roberts@vf.vodafone.co.uk>, "Hardial Lakhpuri"
> <hardial.lakhpuri@vf.vodafone.co.uk>, "Dave Campbell"
> <dave.campbell@vf.vodafone.co.uk>, "Mark Jones"
> <mark.jones@vf.vodafone.co.uk>
>
> All...
>
> Some weeks ago I circulated a request for your opinions on
VxVm/VxFs/UFS/SDA
> etc.
>
> Thanks to everybody who replied including:
>
> Rachel Polanskis [mailto:grove@zeta.org.au]
> Todd M. Wilkinson [tmwilkin@mail.com]
> Simon-Bernard Drolet [sxi@sympatico.ca]
> Kevin Buterbaugh [Kevin.Buterbaugh@lifeway.com]
> Jeff Kennedy [jlkennedy@amcc.com]
> BAUMLER Julie L [julie.x.baumler@co.multnomah.or.us]
> Lee, Elizabeth [elizabeth.lee.contractor@fnmoc.navy.mil]
> King, Brooke [Brooke.King@bankofamerica.com]
> Karl Vogel [vogelke@dnaco.net]
> Broun, Bevan [brounb@adi-limited.com]
> EXT / STERIA TORRES Jean-Yves [ext.steria.torres@sncf.fr]
> John Eisenschmidt [jweisen@eisenschmidt.org]
> Bob Smith [b_smith44@hotmail.com]
>
> I must also thank Karl for some general tuning/performance tips.
>
> Its kinda hard to summarise actually because everybody said loads of
> interesting stuff. basically I have hacked in peoples comments roughly
into
> some relevant sections below. Probably best just to read it all if you
are
> interested and make up your own mind.
>
>
> Thanks everybody.
>
> best regards - T
>
> +----------------------------------------+
> | TONY MILLER
> | Team Leader : Technical Projects,
> | VODAFONE LTD,
> | Derby House,
> | Newbury Business Park,
> | Newbury, Berkshire.
> |
> | Phone +44 (0)1635-677687(local)
> | Email anthony.miller@vf.vodafone.co.uk
> | FAX +44 (0)1635-233517
> +-------------------------------------------
>
>
>
> VxVm/SDA
> ========
> I would not have Veritas controlling my boot disks. I would be wanting to
> stick with something the Solaris install CD can understand.
>
>
>
> * UFS+ logging for Boot system disk
> * VxVM for application file systems
> Why ? for the first case, it's rather easy to un-mirror and un-encapsulate
a
> UFS + logging a boot system disks
> and the action should be performed more often on a system disks than on
> application disks
> UFS is free and allows very fast and simple system management
>
> VxVm is application oriented and its GUI is very powerful and provides
> pertinent informations.
> Also, for application operation, for example, if I want to migrate datas
> from an EMCbay to another one,
> I can simply mirror one bay to another one with VxVm, very simply and
after
> cut the link between the two bays. This is the easiest way and VxVm fits
tho
> our needs
>
>
>
> Recommend using Disksuite on the boot volumes
> due to it's simpler recovery process. VxVm used on data volumes
> or where there are large numbers of disks involved. Veritas give better
> support for multi disk set-ups with multiple controllers as well.
> Disksuite is better for simple set-ups where only a few disks are
involved.
>
>
>
> VxVm - don't use vxupgrade scripts as these are unreliable. Unencapsulate
by
> hand etc and re-encapsulate/mirror after upgrade.
>
>
>
> We use VxVm everywhere - The new release of SVM in Sol9 and the patched
> version in the last release of Sol 8 they keep private regions on the
disk
> rather than on a database making the use of SVM a better option in the
future.
> We are looking at using SVM for mirroring the root disk for appliance like
> boxes with little system data ( DNS servers, load balanced web server
etc. ).
>
>
>
> we've (I) decided that it is easier to have a standard root/boot disk
set-up,
> so we always use SVM+UFS+logging to mirror the boot disks. On top of
that, if
> we have more than 20 disks, or we use a cluster software or we need
support
> 24/7, then we add VxVM+VXFS on the server. The point here is that the
system
> disks on a simple server like a netra T1 with two disks or a SF3800 with
two
> disks, will have the same layout.
>
>
>
> I know a lot of sites that use SVM for mirroring the OS and VxVM for
> everything else. That's because encapsulating the root disk is a pain
> compared to mirroring with SVM (especially dealing with a failed disk / OS
> upgrade). we've found that SVM, especially in conjunction with UFS
logging,
> gives us an adequate level of performance. Management is simpler, since
> you're only dealing with one volume manager instead of two. And, as
you've
> already noted, SVM is free while VxVM is big bucks.
>
>
>
> My answer would be this, use Solaris 9 and scratch Veritas where possible.
> The new volume manager in Solaris 9 looks great, does 90% of what VxVM
will
> do, and for free.
>
>
>
> This is similar to the problems we are wrestling with. I am a huge fan of
> VxVM but the lack of integration into Solaris definitely causes problems.
> More than the upgrade issues, the fact that you have to do special work to
> safely install a kernel patch on a system that has a VxVM encapsulated
root
> file system is miserable. The fact that you need 2 spare partitions to
> install VxVM on your root disk is also a problem for me, now that
management
> is buying us 20-36GB root drives.
>
> Last year, I spoke with some of the Veritas engineers about the kernel
> patching issue and was told that in the next major release of VxVM, you
> wouldn't be required to have a rootdg. This would mean that you could
> unencapsulate root without removing VxVM, which solves both the patch and
> the upgrade problems. It also means that you will be able to use
Disksuite
> for mirroring your root partitions (which is well integrated into the OS)
> and use VxVM for the data partitions and still be able to move your data
> volume groups to another system (some people work around the current
> situation by putting their data, not their root, in rootvg). My plan was
to
> transition to a situation where this is our default configuration. In the
> meantime, I am continuing to run VxVM on all the systems that had it
> already. I have started using Disksuite on new systems, as long as I can
> meet my customer's needs. Apparently there are tools to allow you to
> convert SDS partitions to VxVM partitions, but this is really only an
issue
> if you are using software RAIDS, so I haven't fully investigated it. I am
> by no means an expert in DiskSuite, but knowing what a volume manager can
do
> has made it very easy for me to find the commands I need to run to meet my
> needs. I've been impressed with DiskSuite over the last year, and our
> already project/customer oriented environment has become more so in that
> time period; we are talking about moving to an environment where customers
> (projects) can choose not to pay for the cost of an extra disk to mirror
> their root disks (some people have already doing this against our will by
> going above us), but then agree to non-priority restores (i.e.. We only
> restore during our regular working day and if a RAIDed customer loses
data,
> the later gets priority.); I suspect that we will start offering the
choice
> of SDS/VxVm/both on a cost basis. We also support AIX (and therefore
LVM);
> I find that my co-workers who are comfortable with the concepts of volume
> management have no problem administering multiple volume manager software,
> those who are confused by the concepts need a cookbook approach to most
> tasks, since the DiskSuite documentation provides this, adding DiskSuite
to
> the mix hasn't been at all painful.
>
>
>
> Most of the installations where I have worked use VxVm for all file
systems
> except the OS; those are usually Solstice Disk Suite or UFS (the logic
being
> that the OS can be rebuilt in a relatively short time, and the VxVm
volumes
> re-imported.) I agree that LSM does have a much more seamless way of
> dealing with the systems/OS disks.
>
>
>
> This message is more of a commentary than an answer to your query. One has
> to do one's own cost/risk/benefit analysis. A detailed answer to your
> questions would reveal we use SVM, VxVM, UFS, and VxFS in many situations
> where the other could well be used. We have tended to examine the reasons
> for choosing and made the decisions for classes of servers, not a blanket
> decision for all servers. Obviously, the smaller the support staff, the
> better the argument to standardise on one or the other.
>
> First, do not leave root disks encapsulated. It's a hassle. Instead follow
> the Sun Blueprint for best VxVM practices, and, after mirroring,
> reinitialise the root disk so that it is easily manageable as an ordinary
> VxVM disk. This doesn't make upgrading any easier, but it does make
> replacement easier when that unfortunate need arises.
>
> Second, I have used VxVM and SVM (SDS, ODS, whatever) for years and been
> pretty happy with both. However, I note that with Live Upgrade from
Solaris
> 8 to 9 one still cannot leave the new 9 slices in SVM: they must be plain
> slices until one is done. However, one can access non-system slices from
the
> Live Upgraded system. This is an advantage over VxVM because the SVM is
> upgraded with Solaris.
>
> Third, I have found it to be quicker and easier to replace a failed disk
> quickly with VxVM than with SVM.
>
>
>
> We use both Disk Suite and Volume Manager, more VM than DS. Disk Suite is
a
> nice little product, but is limited in its application. Volume Manager is
> infinitely extensible, so you can add disk array after disk array to your
> running system and subdivide it to the Nth degree.
>
> Whichever product we're using for that particular server, we use it across
the
> board. We do not mix Volume Manager and Disk Suite on a single system.
>
> For our production systems we use Volume Manager, mostly for the minimerge
> capabilities when a mirror gets out of sync. We had some troubling
problems
> with mirror mismatches with Disk Suite, which we fine except they required
a
> full disk remirror - not something I want to do on a production box in the
> middle of the day.
>
> We also have Alpha systems running OpenVMS. On those we use VMS Volume
> Shadowing, which is just like Disk Suite, but that's because Volume
Manager
> isn't available for VMS, and Volume Shadowing is relatively cheap.
>
>
>
> they both have appropriate places in an enterprise. ODS/SDS is a perfectly
> fine choice for simple things such as mirroring and/or striping but lacks
> some nice bells and whistles like soft partitions and shrink online. ok,
> well, the newest SDS does support soft partitions but its really just a
> hack.... basically I use VxVM exclusively if I already need it on a
machine,
> and I
> use ODS/SDS if I do not need VxVM.
>
>
>
> UFS+Logging v VxFs
> ==================
> Opinions vary here but where UFS is used ensure logging is turned on for
all
> file systems that can support it.
>
>
>
> Always VxFS for oracle file systems because performance is so much better,
and
> we evaluate
> the needs for VxFS on systems that may have high IO needs. Basically if
disk
> performance is
> an issue it gets VxFS. UFS + logging hasn't proven itself to be a speedy
> system as of yet.
>
>
>
> VxFS is by far, a better file system in terms of proven reliability and
> performance. Plus it looks to work with large 500GB+ sizes much better
than
> UFS has shown itself in the past. If you are going to spend the bucks to
> first purchase VxVM on a large system ( tier 3 and sometimes even tier
2 ),
> you probably have the need for VxFS on it as well for a least
> one application. So once you have a license to run VxFS on a box you
might as
> well stick with it over UFS+logging just to keep things similar, not to
> mention performance.
>
>
>
> Veritas file system has really been slower than UFS since Solaris 8 came
out
> unless you get QuickIO, then it makes up the difference and then some (but
it
> costs money of course). Now that Solaris 9 UFS supports snapshots as well
I
> can see no reason to use Veritas for anything except very specific
application
> servers (like high transaction Oracle servers). Even then you may be
better
> off with straight Solaris for the simplicity and integration.
>
>
>
> Fourth, I have had at least two occasions where I had to turn off UFS
> logging in order to mount a formerly logged UFS. It simply quit working. I
> found this odd because the UFS logging code is the same as what was in SDS
> at the time and may be the same as what's in SVM now. Yet, it happened. I
> have never had this problem with VxFS.
>
>
>
> UFS + logging. Some suggestions are below. VxFS is too expensive, and
> if you spread stuff over several drives and get a media problem, it's
> tough to find out what drive is actually failing.
>
>
>
> for most things UFS+logging is just fine but for environments with heavy
> loads VxFS can really extend the performance of the hardware. e.g. places
> like busy: databases, mail, news servers. basically anywhere with small
> individual or random i/o will see an improvement.
>
>
>
> Sun's new file system and volume manager (formerly developed by LSC)
perform
> very well against VxFS for high end database needs. the down side is that
> the product is still fairly young, management is rough and
> administration/tuning can be tedious.
>
>
>
>
> Snapshotting capabilities of Verita
> ===================================
> Most didn't comment about this at all. The message from those that did
was
> basically:
> Solaris 8 & above now also has "fssnap" provided native which will do
> snapshots.
> I was actually aware of this because we use it internally - I was
interested
> to see how many others use it.
>
>
>
> Fifth, one can now take snapshots with UFS, but I haven't tried this
> feature, yet. This is one feature where VxFS used to be a clear winner
> because only VxFS had it, but no more.
>
>
>
> HP-UX user, do you have any intentions to migrate to VxVm
> =========================================================
> Not many responded here.
>
> We did use VxVM on HP, but the support folks ( hp support folks ), through
> such a hissy fit about it that finally discontinued its use and went back
to
> HP LVM. Recover procedures although never tested seemed a little complex
on
> the HP side with VxVM. Please note our HP environment and VxVM on HP is
very
> small.
>
> --__--__--
>
> Message: 2
> Date: Thu, 22 Aug 2002 17:12:53 -0700 (PDT)
> From: Mt Dew <devconf@yahoo.com>
> Subject: rootdisk mirroring with Disksuit. help.
> To: sunmanagers@sunmanagers.org
>
> Hi,
> I have a server that can't come back from reboot, its in Single user
mode..
> Can anyone point me to the right direction?
> Thanks
>
>
> vfstab:
> #/dev/dsk/c1d0s2 /dev/rdsk/c1d0s2 /usr ufs 1 yes -
> fd - /dev/fd fd - no -
> /proc - /proc proc - no -
> #/dev/dsk/c0t2d0s1 - - swap - no -
> /dev/md/dsk/d1 - - swap - no -
> /dev/md/dsk/d0 /dev/md/rdsk/d0 / ufs 1 no -
> swap - /tmp tmpfs - yes -
> /swapfile - - swap - no -
>
>
> df -k in single mode:
> Filesystem kbytes used avail capacity Mounted on
> /pci@1f,0/ide@d/disk@2,0:a
> 114360309 2481362 110735344 3% /
> /proc 0 0 0 0% /proc
> fd 0 0 0 0% /dev/fd
> mnttab 0 0 0 0% /etc/mnttab
> swap 1576216 0 1576216 0% /var/run
> /dev/dsk/c0t0d0s0 114360309 2481362 110735344 3% /tmp
>
>
> metastat:
> d0: Mirror
> Submirror 0: d10
> State: Needs maintenance
> Submirror 1: d20
> State: Needs maintenance
> Pass: 1
> Read option: roundrobin (default)
> Write option: parallel (default)
> Size: 232294800 blocks
>
> d10: Submirror of d0
> State: Needs maintenance
> Invoke: metasync d0
> Size: 232294800 blocks
> Stripe 0:
> Device Start Block Dbase State Hot Spare
> c0t2d0s0 0 No Okay
>
>
> d20: Submirror of d0
> State: Needs maintenance
> Invoke: metasync d0
> Size: 232294800 blocks
> Stripe 0:
> Device Start Block Dbase State Hot Spare
> c0t0d0s0 0 No Okay
>
>
> d1: Mirror
> Submirror 0: d11
> State: Needs maintenance
> Submirror 1: d21
> State: Needs maintenance
> Pass: 1
> Read option: roundrobin (default)
> Write option: parallel (default)
> Size: 2101200 blocks
>
> d11: Submirror of d1
> State: Needs maintenance
> Invoke: metasync d1
> Size: 2101200 blocks
> Stripe 0:
> Device Start Block Dbase State Hot Spare
> c0t2d0s1 0 No Okay
>
>
> d21: Submirror of d1
> State: Needs maintenance
> Invoke: metasync d1
> Size: 2101200 blocks
> Stripe 0:
> Device Start Block Dbase State Hot Spare
> c0t0d0s1 0 No Okay
>
>
> metadb -i
>
> flags first blk block count
> a m c luo 16 1034 /dev/dsk/c0t2d0s3
> a c luo 16 1034 /dev/dsk/c0t2d0s4
> a c luo 16 1034 /dev/dsk/c0t2d0s5
> a c luo 16 1034 /dev/dsk/c0t0d0s3
> a c luo 16 1034 /dev/dsk/c0t0d0s4
> a c luo 16 1034 /dev/dsk/c0t0d0s5
>
>
> metastat -p
> d0 -m d10 d20 1
> d10 1 1 c0t2d0s0
> d20 1 1 c0t0d0s0
> d1 -m d11 d21 1
> d11 1 1 c0t2d0s1
> d21 1 1 c0t0d0s1
>
>
>
>
>
> Executing last command: boot
> Boot device: disk File and args:
> SunOS Release 5.8 Version Generic_108528-13 64-bit
> Copyright 1983-2001 Sun Microsystems, Inc. All rights reserved.
> configuring IPv4 interfaces: dmfe0.
> Hostname: doppler
> NOTICE: dmfe0: PHY 1 link up 100 Mbps Full-Duplex
> NOTICE: dmfe1: PHY 1 link down
> mount: /dev/md/dsk/d0 is not this fstype.
> failed to open /etc/coreadm.confsyseventd: Unable to open daemon lock file
> '/etc
> /sysevent/syseventd_lock': 'Read-only file system'mount: warning: line for
> "<nul
> l>" in vfstab has too few entries
> mount: warning: line for "<null>" in vfstab has too few entries
>
> INIT: Cannot create /var/adm/utmpx
>
> INIT: failed write of utmpx entry:" "
>
> INIT: failed write of utmpx entry:" "
>
> INIT: SINGLE USER MODE
> HotJobs - Search Thousands of New Jobs
> http://www.hotjobs.com
>
> --__--__--
>
> Message: 3
> From: "Trinh, Linh T. [C]" <TrinhL@nima.mil>
> To: sunmanagers@sunmanagers.org
> Subject: need help with semaphore
> Date: Fri, 23 Aug 2002 10:51:11 -0400
>
> Hi Alls,
>
> I am having problem with the oracle database server, WebLogic application
> can run only one oracle instance and Veritas NetBackup 3.4 did not run for
> the last 2 night because of this error
>
> 08/21/2002 23:09:00 host1 host1 system call semop() lock failed, No space
> left on device
>
> It is an E5500 with 8 400MHZ CPUS, 6 GB memory, Solaris 8, Oracle 8.1.7.3
>
> /etc/system setting
> set shmsys:shminfo_shmmax=4294967295
> set shmsys:shminfo_shmmin=1
> set shmsys:shminfo_shmmni=300
> set shmsys:shminfo_shmseg=10
> set semsys:seminfo_semmns=3640
> set semsys:seminfo_semmni=1820
> set semsys:seminfo_semmsl=250
> set semsys:seminfo_semopm=100
> set semsys:seminfo_semvmx=32767
>
> Anyone have any ideas about these number.
> Thanks, Linh
>
> --__--__--
>
> Message: 4
> Date: Sat, 24 Aug 2002 11:47:36 -0700 (PDT)
> From: raoul s <raoul_s2001@yahoo.com>
> Subject: question: need urgent help
> To: sunmanagers@sunmanagers.org
>
> on ultra10/solaris2.8, i did:
> ok> boot disk kernel/unix to boot from 32bit.
> then later i did:
> ok>setenv boot-file kernel/unix
>
> now, i can not get it into 64 bit boot.
> i do:
> ok>setenv boot-file kernel/unix/spacrv9
> it goes into can not open ...
>
> how can i boot into 64 buit mode?.
> raoul
> Yahoo! Finance - Get real-time stock quotes
> http://finance.yahoo.com
>
> --__--__--
>
> Message: 5
> Date: 26 Aug 2002 01:51:43 -0000
> From: "sridhara pk" <sridharapk@rediffmail.com>
> Reply-To: "sridhara pk" <sridharapk@rediffmail.com>
> To: sunmanagers@sunmanagers.org
> Subject: FTP server
>
> Hi
> I have an ftp anonymous server with sublogins and they have access
> to one particular folder
> My problem is if a user logs in as anonymous he is able to browse
> all the directories under that folder I want to restrict the same
> to his home folder only how to do that
>
> Thanks in advance
>
> Sridhara
>
> --__--__--
>
> Message: 6
> From: Slamet Ahmad Fadilah <Slamet.Fadilah@ptsi.siemens.co.id>
> To: "'sunmanagers@sunmanagers.org'" <sunmanagers@sunmanagers.org>
> Subject: SUMMARY (UPDATED): "The IDPROM contents are invalid"
> Date: Mon, 26 Aug 2002 09:29:36 +0700
>
> Hi,
>
> Found that the way to clear the errormessage: "The IDPROM contents are
> invalid" with script mkp or mkpl did not work properly for E4500 machine
> type 0x83..... with Ethernet address 0:3:ba:xxxxxx Host ID 83xxxxxx.
>
> It needs an upgrade of OBP to 3.2.29. (Mine was 3.2.22)
>
>
> BR, slamet
>
>
>
> > -----Original Message-----
> > From: Slamet Ahmad Fadilah [SMTP:Slamet.Fadilah@ptsi.siemens.co.id]
> > Sent: Thursday, August 08, 2002 7:21 AM
> > To: 'sunmanagers@sunmanagers.org'
> > Subject: SUMMARY: "The IDPROM contents are invalid"
> >
> > Yihaaa!!!
> >
> > God for shake, the chip replacement is not the only way to do.
> > However, to have a spare of nvram chip is highly recommended.
> >
> > Many many thanks to
> > Christophe Dupre
> > Jeff Markham
> > Hichael Morton
> > Sysadmin@astro.su.se
> > and especially to Buddy Lumpkin who exactly pointed me out the other
great
> > way.
> >
> > ----- Buddy Lumpkin:
> > write down the ethernet address and hostid (if they are valid, all f's
> > means
> > it's erased).
> > Let's assume the ethernet address is 8:0:20:af:b1:cc and the hostid is
> > 80afb1cc
> >
> > Then do this:
> >
> > ok 0 0 mkp
> > ok 8 0 20 af b1 cc 80afb1cc mkpl
> > ^d^r
> >
> > (that's control+d followed by control+r)
> >
> > if you still have troubles, try 1 1 mkp instead. (0 0 mkp did work on my
> > case)
> >
> >
> > Your getting the message about IDPROM contents being invalid because
your
> > checksum is hosed up.
> > The mkpl command above writes the ethernet address and hostid and
> > recreates
> > the checksum.
> > You may need a new nvram soon.
> >
> > Also there should be some variation of copy nvram to clock or something
> > like
> > that.
> > Just do a sifting copy or sifting clock to find the command.
> > There should be a few variations, none of them will "hurt" your system.
> > The command simply updates other boards in the system with the same
> > information.
> >
> > -----
> >
> >
> > :-)
> > slamet fadilah
> > icm n ms tac2, siemens mobile networks
> >
> >
> >
> >
> > > -----Original Message-----
> > > Hi All,
> > >
> > > I just got a packet of a new SUN E4500 (sun4u machine).
> > > But after its first OS installation with an old OS of 2.5.1:
> > > It sent an error message "The IDPROM contents are invalid" which
appears
> > > after banner (The banner looks as expected).
> > >
> > > Tried to reset the NVRAM (using set-defaults or set-nvram and power
> > cycle
> > > it), didn't work.
> > >
> > > Our sunmanagers archive says that it could be due to something wrong
on
> > > NVRAM chip (or its battery).
> > > I wonder, this machine is just a new one, the NVRAM and its battery
> > should
> > > be okay.
> > > Any idea how to check their condition? Is there something else causing
> > the
> > > problem, like a minimum OBP version to be install? Or a chip
replacement
> > > is
> > > the only way to resolve this kind of problem ?
> > >
> > > Many thanks for your time.
> > > A summary will follow.
> > >
> > > BR, slamet
> > _______________________________________________
> > sunmanagers mailing list
> > sunmanagers@sunmanagers.org
> > http://www.sunmanagers.org/mailman/listinfo/sunmanagers
>
> --__--__--
>
> Message: 7
> Date: Sun, 25 Aug 2002 20:52:42 -0700 (PDT)
> From: Very Pinguin <dt1649651@yahoo.com>
> Subject: SUNPCI w/ Solaris 9
> To: sunmanagers@sunmanagers.org
>
> Hi all,
>
> I just installed a SUNPCI 400Mhz on Ultra 5 with
> Solaris 9.
> Looking at Sun site, I installed the SunPci software
> 1.3.
>
> I have two questions:
> 1. How do I know if OBP recognizes the SUNPCI card ?
> ( I 've had touch /reconfigure and reboot )
>
> 2. When adding the package SUNWspci, I get the
> following error:
> [...]
> /opt/SUNWspci/tables/keytables/dos_kb_us
> [ verifying class <none> ]
> ## Executing postinstall script.
> Warning: /etc/devlink.tab not updated correctly
> Perhaps the system OBP needs to be updated
> Can not open the SunPCi device
> The device driver is not installed.
> pkgadd: ERROR: postinstall script did not complete
> successfully
>
> Installation of <SUNWspci> partially failed.
>
> I tried the software 2.3.1 and got similar error.
>
> Please give me some advice,
>
> Thanks,
>
> DT
> Yahoo! Finance - Get real-time stock quotes
> http://finance.yahoo.com
>
> --__--__--
>
> Message: 8
> From: Moshe Leder <leder@mercury.co.il>
> To: "'sunmanagers@sunmanagers.org'" <sunmanagers@sunmanagers.org>
> Subject: SUMMERY: NIS port number
> Date: Mon, 26 Aug 2002 10:17:28 +0200
>
> It uses SUNRPC port (Numbered 111)
>
> Thanks all for the handful & helpful responses!
>
> Moshe Leder.
>
> > -----Original Message-----
> > From: Moshe Leder
> > Sent: Thursday, August 15, 2002 6:23 PM
> > To: sunmanagers@sunmanagers.org
> > Subject: NIS port number
> >
> > Hi managers,
> >
> > Does any one knows what is the port number ypbind uses when it sends
> > broadcasts to look for the NIS server ?
> >
> > New Cisco OS has the option to configure a certain IP forwarding feature
> > that will turn broadcast to unicast & forward it across routers. This in
> > turn can be enable NIS server to answer broadcast for many subnets.
> >
> > Thanks in advance,
> > - Leder.
>
>
> --__--__--
>
> _______________________________________________
> sunmanagers mailing list
> sunmanagers@sunmanagers.org
> http://www.sunmanagers.org/mailman/listinfo/sunmanagers
>
>
> End of sunmanagers Digest
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers
This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:24:53 EDT