Tuesday, 8 April 2008
|
| Solaris Admin Opportunities in Chennai to provide Level 3 support Deesharan 09:59:46 |
| | Designation : Solaris Administrator-Solaris, Volume Manager, Cluster
to provide level 3 support to our Global clients
Desired Profile :
Solaris Administrators.....
Experience in NIS/NFS/DNS/DHCP
Experience in Volume Manager (Verities, Solaris), VXFS
Knowledge of Backup tools
High availability (Verities Cluster, Sun Cluster)
Excellent Communication skills
Experience: 3 to 4 relevant experience in Solaris Administration and
1Year experience in Verities, Veritas or Sun Clusters
Location:- Chennai.
Mail us resumes to (deesharan@sify.com/ deesharan@rediffmail.com)
Mail your resumes to deesharan@sify.com
|
| | Add comment |
|
| Re: Robert Peffers: The Case For Controlled Demolition Loaf of Bread 06:47:37 |
| | On Apr 5, 8:52pm, schoenfeld....@gmail.com wrote:
In this article we show that "top-down" controlled demolition accurately accounts for the emotional collapse of Robert Peffers A top-down controlled demolition can be simply characterized as a "pancake collapse" of a socialist roving fruitcake missing its support columns. This demolition profile requires that the support columns holding a floor be destroyed by plastic explosives strapped to the roving socialist fruitcake's chest. The net effect is a pancake-style collapse at near free fall speed and a large number of deaths in a Baghdad marketplace. |
| | Add comment |
|
| Looking for SUNWpatch Colin B. 03:49:57 |
| | Does anyone know where/how to obtain this install package? It's got
tools for building patches, and doesn't seem to be anywhere on the
install media. (Yes, I've looked.)
Thanks,
Colin
|
| |  5 answers | Add comment |
Monday, 7 April 2008
|
| lwp/threads documentation? Bugbear 19:38:50 |
| | I'm still trying to trace my Java/Shell() performance issue,
and have been using truss/dtrace.
I've got some "interesting" log files, but
my ability to analyse further is now hampered
by the fact that I cannot find documentation
for the functions that dtrace and truss tell
me are being called.
I have searched in the manuals, Sun's doc site, and
googled.
http://docs.sun.com/app/docs/coll/40.10
In particular I would like a detailed
description of
lwp_park
lwp_unpark
although I have others; I suspect that a source
of information for one of these will have
all the others.
Is this stuff actually secret, or mere
well hidden?
BugBear
|
| | 1 answer | Add comment |
|
| Service quality of Sunsolve (was Re: SUNSOLVE IS DRIVING ME INSANE!!!) Andreas F. Borchert 19:30:08 |
| | On 2008-03-19, news@buffy.sighup.org.uk <news@buffy.sighup.org.uk> wrote:
The service is currently so poor we are looking at creating a local patch server so we don't have to rely on sunsolve being available at times we need it to be available.
We are running a local patch server which mirrors a subset of patches
we might need. Included in this set are all recommended patches for all
releases we are currently running. To examine the set of recommended
patches, we periodically download the patch reports like 10_patch_report
or 9_patch_report. Each of these patch reports carries a date. So far
we assumed that a newer download will deliver the last or a more recent
patch report but we were wrong. This is our recent download history
of 9_patch_report:
Download time (GMT) Release Date
-----------------------------------------
4 Apr 2008 06:57:05 Apr/1/2008
4 Apr 2008 22:29:20 Mar/16/2008
5 Apr 2008 00:56:34 Apr/1/2008
5 Apr 2008 01:56:18 Mar/16/2008
6 Apr 2008 01:56:26 Apr/1/2008
7 Apr 2008 00:56:08 Mar/16/2008
7 Apr 2008 01:56:13 Apr/1/2008
7 Apr 2008 06:56:05 Mar/16/2008
That means that 9_patch_report is served in two alternating versions, at
times the current release from Apr/1/2008 and, unfortunately for longer
time periods, the outdated report from Mar/16/2008. Similar patterns
were also be observed for some of the other patch reports.
As mentioned before, some files can no longer be downloaded. Our last
successful download of patchdiag.xref is from March 15th. Our copy of
CHECKSUMS is more recent but most attempts to download it fail as well.
(We use these files only to get a list of available patches as an
FTP directory listing is no longer available. The checksums themselves
turned out to be unreliable.)
Yes, we are a paying customer with multiple service contracts. And, yes,
we consider the current service quality of Sunsolve as painful.
Andreas.
|
| | 2 answer | Add comment |
|
| how to check live upgrade Solaris Chat 19:15:39 |
| | when upgrading solaris 9 to 10 ,how to check the status from the
commandline
|
| | 1 answer | Add comment |
|
| Memory leak Chimpanzee 18:47:45 |
| | We are having memory leak on one of our systems.The system freezes due
to lack of memory after certain number of days.
My memstat is as follows
::memstat Page Summary Pages MB %Tot ------------ ----------------
---------------- ----
Kernel 649618 5075 62%
Anon 50611 395 5%
Exec and libs 3165 24 0%
Page cache 50652 395 5%
Free (cachelist) 109278 853 10%
Free (freelist) 177436 1386 17%
Total 1040760 8130
Physical 1023900 7999
Is Kernel taking bulk of the memory(60%) or is it working as per
design?How I can Troublshoot further the memory leak by using system
tools
Cheers
|
| | 3 answer | Add comment |
|
| Solaris 10 networking - Global zones Thiin White Duke 18:37:28 |
| | We have a most interesting problem. When we boot our global zones
some of the non global zones don't come up on the network. We don't
know why. For some reason one or two will, but the third won't and
it's on the same network as the other two. This just started
happening.
We think it might be NIS+ related on the globals, but aren't sure.
Any idea why one or two zones would start their network services but
the other won't?
|
| | 4 answer | Add comment |
|
| Re: split mirror disks for adding patch Emerson Seiti Takahashi 17:28:27 |
| | On Mar 15, 11:17 am, BL <c50342182...@gmail.com> wrote:
Hi, If I use disksuite, I can split mirror disks easily by "metadetach" before adding patch to my solaris server. But, if I use VxVM, how can I split the mirror? And, how can I boot up the old root disk in VxVM if the patch has problem? Thanks
This is the support document that Veritas describes how to preserve
the mirror disk before the patching.
http://seer.entsupport.symantec.com/docs/275374.htm
|
| | 1 answer | Add comment |
|
| FAT signature error John F. Carr 16:46:18 |
| | One of my CF cards has a FAT16 filesystem that my Solaris 10
SPARC refuses to mount. Any tips on diagnosing this? The card
has some pictures taken with my digital camera that I really
don't want to lose.
I mount my camera (Nikon D300) as a filesystem using a USB
cable. The results are the same, positive or negative, if
I use a card reader instead of the camera. When I go to
mount /dev/dsk/c3t0d0s2:c as type pcfs the kernel logs an
error
NOTICE: pcfs: FAT signature error
NOTICE: pcfs: illegal disk format
My other cards mount fine. The bad card is 2GB; the good
cards are 2GB and 8GB.
Google leads me to a problem with 2K sectors. I looked
at the bits in the first sector of the partition on my
card and found sector size 512.
I know the pictures are on the card because I can view them
in my camera; "strings" on a copy of the card also shows
plausible text timestamps. (In fact, a user-mode tool to
extract the files from the copy I made of the card would
suit me just fine.)
My kernel boot version string:
SunOS Release 5.10 Version Generic_118833-33 64-bit
(When I went to review Sunsolve's used-to-be-free patch
clusters to get up to date the site told me I couldn't
get them without paying.)
--
John Carr (jfc@mit.edu)
|
| | 2 answer | Add comment |
|
| Is there a way to modify the environment of a running process in
Solaris 9/10? Guest 10:21:20 |
| | Hello.
Is there any way to change environment variable of a running process?
Ex. attaching a process to mdb and then changing some variables?
What I want.
Process A is running with PID=1000.
1)pstop 1000
2)Modification (in any manner) to A variables
3)prun 1000.
Is possible?
Is possible also for the user owner of the process?
Thanks
|
| | 1 answer | Add comment |
|
| Setup Intel Pro/1000 PT on Solaris 10 Ken 04:01:14 |
| |
Hi,
I installed an Intel Pro/1000 PT ethernet card on a Solaris 10 AMD x64
machine.
I added the driver information below to /etc/driver_aliases (per the
link below) and rebooted after each change.
Intel Corporation
Notes: Gigabit Ethernet Products.
1. Driver name e1000g
2. Add 'pci8086,1092' to /etc/driver_aliases and reboot system.
3. Add 'pci8086,11bc' to /etc/driver_aliases and reboot system.
4. Add 'pci8086,1093' to /etc/driver_aliases and reboot system.
http://www.sun.com/io_technologies/vendor/intel_corporation.html
Solaris doesn't seem to recognize the NIC.
I tried ifconfig -a and only the loopback connection was displayed.
I tried ifconfig -a plumb and nothing was displayed.
Questions:
1. How do I configure the NIC?
2. How do I get the device name of the NIC?
3. How do I configure DHCP?
Thanks
|
| | 7 answers | Add comment |
|
| Can root mirroring on same drive cause OS crash? Guest 01:43:44 |
| | Can root mirroring on same drive cause OS crash?
I'm maintaining some sparc SUNW,Netra-440 servers. and created a root
mirror on the same drive (c1t0d0s0 and c1t0d0s3). Recently, some
servers were crashed, but I didn't get any crash dump file and no
error message on /var/adm/message, so I'm concern if the root
mirroring cause the OS crash, can any give some information, I can not
get info from SUN doc.
Thanks in advance!
Jason
|
| | 11 answers | Add comment |
Sunday, 6 April 2008
|
| Allowing a group of users to manage a service Michael Schmarck 22:34:32 |
| | Hello.
I'd like to allow a group of users to manage (ie. disable, enable, ...)
a service, WITHOUT having to use sudo. That's supposed to be on a
Solaris 10 U4 system.
To allow one (or many indivial) user(s) to do the management, I followed
http://learningsolaris.com/archives/2005/04/25/smf_and_rbac/, ie.
I did:
# usermod -A solaris.smf.manage.nfs/server bob
# svccfg -s nfs/server setprop general/action_authorization=astring: \
'solaris.smf.manage.nfs/server'
# svccfg -s nfs/server setprop general/value_authorization=astring: \
'solaris.smf.manage.nfs/server'
# svcadm refresh nfs/server
After that, "bob" was able to do "/usr/sbin/svcadm disable nfs/server".
But now I'd like to have a group, let's call it "nfsadmins", that
should be able to do this.
How would I do this? I cannot do
# usermod -A solaris.smf.manage.nfs/server nfsadmins
as there's no such user.
Thanks a lot,
Michael
|
| | 10 answers | Add comment |
|
| Re: Need help with IDPROM error message Wolfgang 15:09:34 |
| | Richard B. Gilbert schrieb:
Alfonso G. Urroz wrote: I have an Ultra SPARC-IIi 300 Mhz, system, with 512 MBytes of EDO Ram, it was working fine until yesterday, when all of the sudden a meesage sayoin "IDPROM contents invalid". I was able to get to the OK prompt, a couple of times, and when I did ./idprom, a whole bunch of garbled characters came up. It shows a MAC address, that is not evn close to what it used to be before, also the time (clock) is all screqwed up. Question is, is there any solution to this problem?, can the chip be reprogrammed?, do I have to buy a new chip?, if so, where do I go, or write in order to get the chip? Or, is my Sun box, dead beyond repair? I suspect that your battery is dead or nearly dead. I believe it's built into a socketed chip on your mother board. The cleanest solution is to replace the chip which should fix you up for another five or ten years. Alternatively, you can hook up an external battery. for the sordid details.
in case you buy a new clock chip: do you know the old data from your box
(MAC, ID)? Also a new one maybe must reprogrammed, with is decribed in
the link above, but the mac is only critical in production and id for
dependend licenses or similar.
Wolfgang
|
| | Add comment |
|
| Missing Symbol Dave Uhring 14:47:33 |
| | Trying to build latest svn version of vim on b_85 x86:
Undefined first referenced
symbol in file
Xutf8SetWMProperties objects/os_unix.o
ld: fatal: Symbol referencing errors. No output written to vim
That symbol has a man page, XSetWMProperties(3X11), which tells us to link
with -lX11. But /usr/lib/libX11.so which is a symlink to
/usr/X11/lib/libX11.so.4 does not have that symbol.
[duhring@einstein /usr/X11/lib]$ nm libX11.so.4 | grep Xutf8SetWMProperties
[duhring@einstein /usr/X11/lib]$
In fact, the symbol exists nowhere in /usr/lib or /usr/X11/lib:
[duhring@einstein /usr/lib]$ for file in `ls lib*.so`; do nm $file | grep Xutf8SetWMProperties; done
nm: libcctagent.so: cannot read file
[duhring@einstein /usr/lib]$ cd /usr/X11/lib
[duhring@einstein /usr/X11/lib]$ for file in `ls lib*.so`; do nm $file | grep Xutf8SetWMProperties; done
[duhring@einstein /usr/X11/lib]$
Am I missing something here or is b_85 just broken?
|
| | 4 answer | Add comment |
|
| SunBlade 150 won't boot Guest 14:01:47 |
| | I have a SunBlade 150 that had 1GB of memory installed in it, one
512MB, and two 256MB sticks. I removed the twp 256MB sticks and put
three more 512MB sticks in for a total of 2GB in 4 512MB sticks. When
I tried to reboot it, it would power up, no post, no video. I pulled
out all of the new memory and just left it with the one original 512MB
stick. I tried to power it up again. The power button LED lights up
momentarily, it beeps three times, no video and powers down on its
own. What do the three beeps mean? What should I try?
|
| | 5 answers | Add comment |
|
| where is ssh on solaris 8 distro George2 01:10:07 |
| | Hello, its so long since I used solaris 8 that i cant remember where the ssh
package is . I thought it was in sfw directory on one of the cd's. I need to
install ssh from cd as the network is not working and I just cant find it.
George
|
| | 6 answers | Add comment |
|
| iSCSI target node name and local name Mr. Chow Wing Siu 00:57:09 |
| | Hi,
I am playing iSCSI target.
iscsitadm create target --type raw \
--backing-store /dev/dsk/c1t1d0p0 c1t1d0
The above local name is "c1t1d0" but the iSCSI Name is:
iqn.1986-03.com.sun:02:75702626-0a4b-e9db-d411-f476291fabe6.c1t1d0
Do anyone has the method to change the iSCSI name?
Thanks in advance.
--
Johnson Chow
|
| | 1 answer | Add comment |
Saturday, 5 April 2008
|
| Re: patch installation note Huub 23:20:48 |
| | The description in the README is correct - "Latest Patch Update" is under "Patches and Updates", in the left column, printed in red letters; right above the link to "PatchFinder". If you still can't find it, tell us the OS release you're interested in.
Does it matter that I don't have a subscription?
My OS release is 10 11/07 Sparc.
|
| | 2 answer | Add comment |
|
| Re: Running Debian as a guest OS under LDoms? Guest 23:14:59 |
| | Jonathan Wilson <jwcc@news.cnntp.org> wrote:
I'm having a really hard time finding information about this. I searched google and didn't find much, except some pages that state Ubuntu has been made to run under LDoms.
You will have to find a source of Sparc Linux. Ubuntu used to do one, but
recently said that support was ending (it's still on the download page
however.) There are a couple of other distros that can be found using
Google.
--
Gods I don't believe in: http://anya.sighup.org.uk/gods.html
|
| | 1 answer | Add comment |
|
| How to install a new graphics card on Solaris 10? Michelle 01:06:11 |
| | I have Solaris 10 x86. The motherboard has a built-in graphics card.
How do I install a new graphics card (from the HCL list) so that
Solaris will recognize it?
I realize the first thing I need to do is go into the BIOS to disable
the on-board graphics card. What else do I need to do after this?
Thanks in advance.
|
| | 5 answers | Add comment |
Friday, 4 April 2008
|
| Can the graphics card cause this problem? Michelle 22:54:59 |
| | I have Solaris 10 on x86. I chose the motherboard that is on the
Hardware Compatiblity List from SUN.
The motherboard has a built in graphics card. The system works fine.
The mouse also works fine on Solaris 10.
There's a graphics application that I use on this Solaris 10 x86
system. There are some functions inside that software application
that caused the mouse to behave "weirdly". Like the cursor would slow
down. The cursor may be hard to control, and so forth. But this
problem only occurs inside this graphics software application.
Is the problem likely coming from the software applications itself, or
is this a graphics card issue? Do you think putting a separate
graphics card (that is on the HCL) will resolve this issue?
|
| | 2 answer | Add comment |
|
| setting up a mirror (via software?): easier on 10, 11? (I have 9) David Combs 19:45:18 |
| | Getting jittery about my old disks.
1: must I mirror on exactly identical disks?
2: software-mirroring: blade100 fast enough? or **MUCH** better
with ultra 25?
3: Is there anything in 10 or 11, over 9, that makes
it easier (or anything)?
4: Stepson has mirroring on his pc -- apparantely took
months to acquire needed hardware, and in all
spent something like $5K.
I assume there's no such problem on solaris sparc?
Thanks!
David
|
| | 3 answer | Add comment |
|
| Re: DoD Harddrive Secure Erase Wipe Andrew Gabriel 19:30:25 |
| | In article <4d43e3f3-9140-4764-a9e6-d738d486c5ab@i36g2000prf.googlegroups.com>,
oktokie <oktokie@yahoo.com> writes:
DoD Harddrive Secure Erase Wipe I have a project which I need to DoD harddrives for the company. I have large raid-scsi enclosure which I can use.
This will be defined in your DoD contract and depend on the level
of the data. Most likely it will require physical destruction of
the disks and any NVRAM in the RAID array and embedded disk
controllers.
Writing to the disks through standard OS mechanisms is very
unlikely to be acceptable unless the data was only of the lowest
level commercial interest. Degaussing used to be acceptable and
allowed disks to be reused, but no modern disks are reusable
after degaussing nowadays, so I suspect you are looking at
physical destruction at a level which assures there is no chance
of repair for data recovery.
--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]
|
| | 6 answers | Add comment |
|
