Solaris 10 headaches

i have 2 brand-new X4140 servers in my office. They came with solaris 10 pre-installed.
When booting the first server it started and i entered the necessary details. When running the GUI update-engine it failed. I could only use the commandline to register my server.
After 1 day downloading and installing all updates the server needed a reboot, it died instantly. The RAID controller searches for the hardisks and can't find the boot-disk. The server searches for a networkboot so i shut it down. I lost 2 days valuable time, since i need to test a MySQL setup this week.
So i fired up the second server today. For some reason the GUI update-engine failed also on the register part and i had to use the commandline.
I tried to use the SMC program to enable some new users. BAM,... SMC hit the floor with some strange errors. and i had to use the commandline to make the new users and groups. I could try to install SMC 4.0, but after trying that on some other X4104 and other severs without succes, i won't waste time on that again.
I'm getting disapointed with this messy serverstuff. I do like the ZFS and zone things in Solaris. I think they are the best in the world. But i hate the fact that Sun build some crappy GUI on their servers which doesn't give you anything. If they disable it no-one will miss it, because you can't go any further then enable a CD-server for music (duh! ) or change the desktoptheme.
If i need to manage the server i can choose to install Nagios or Webmin. doesn't matter they are not from Sun, because SMC 4.0 isn't either.
Why doesn't Sun put an OpenSolaris DVD inside the serverbox? Solaris is already as dead as an American Bank.
Why don't they explain how they sliced the HD's? or give you an option to choose one at server startup?
Why don't they tell which version they installed?

Which version of Solaris and how much memory do you have in the system.

Similar Messages

Solaris 10 Display setting Headache

This is my issue:
Open Xwin Server on windows XP machine (my desktop machine)
Open putty session and set Enable X11 forwarding
Using putty I then connect to db server (Solaris 10) logged in as "oracleuser"
$ set
BASH=/bin/bash
BASH_ARGC=()
BASH_ARGV=()
BASH_LINENO=()
BASH_SOURCE=()
BASH_VERSINFO=([0]="3" [1]="00" [2]="16" [3]="1" [4]="release" [5]="sparc-sun-solaris2.10")
BASH_VERSION='3.00.16(1)-release'
COLUMNS=126
DIRSTACK=()
DISPLAY=localhost:10.0
EDITOR=vi
EUID=101
GROUPS=()
HISTFILE=/export/home/oracle//.bash_history
HISTFILESIZE=500
HISTSIZE=500
HOME=/export/home/oracle/
PATH=/usr/bin:/usr/local/bin:/oracle/product/db/11gR202/bin:/usr/ccs/bin:/usr/bin/X11:/usr/local/bin
TERM=xterm
.........etc
As you can see DISPLAY is set
$/usr/openwin/bin/xclock
Xclock displayed on windows XP machine
Success!
Now connecting to an identical box.
open Xwin Server on windows XP machine
open putty session and set Enable X11 forwarding
using putty open connection to Solaris 10 db server logged in as "oracleuser"
$ set
BASH=/bin/bash
BASH_ARGC=()
BASH_ARGV=()
BASH_LINENO=()
BASH_SOURCE=()
BASH_VERSINFO=([0]="3" [1]="00" [2]="16" [3]="1" [4]="release" [5]="sparc-sun-solaris2.10")
BASH_VERSION='3.00.16(1)-release'
COLUMNS=126
DIRSTACK=()
EDITOR=vi
EUID=101
GROUPS=()
HISTFILE=/export/home/oracle//.bash_history
HISTFILESIZE=500
HISTSIZE=500
HOME=/export/home/oracle/
PATH=/usr/bin:/usr/local/bin:/oracle/product/db/11gR202/bin:/usr/ccs/bin:/usr/bin/X11:/usr/local/bin
TERM=xterm
...etc
DISPLAY is NOT set
So
$export DISPLAY=localhost:10.0
$echo $DISPLAY
localhost:10.0
$/usr/openwin/bin/xclock
Error: Can't open display: localhost:10.0
So
$/usr/openwin/bin/xhost
/usr/openwin/bin/xhost: unable to open display "localhost:10.0"
any and all attempts to run anything X11 related end up with the same message
other checks
$cat /etc/ssh/sshd_config | grep -i x11
# X11 tunneling options
X11Forwarding yes
X11DisplayOffset 10
X11UseLocalhost yes
$more /etc/hosts
# Internet host table
10.100.200.10 name-of-server.dev.orgnameint.net name-of-server loghost
::1 localhost
127.0.0.1 localhost
10.1.2.16 ntp1 ntp1.orgnameint.net
10.1.2.17 ntp2 ntp2.orgnameint.net
$./xauth list
name-of-server/unix:10 MIT-MAGIC-COOKIE-1 7b39a111db9b71932c7742f752815073
name-of-server/unix:11 MIT-MAGIC-COOKIE-1 1abaa8baded63a3beceb34f3b0a4ea89
all settings that are set for the box that works are also set for the box that does not work.
I have tried the above as root - still no Joy.
It feels like I have checked every web page designated to this problem and tried all of the solutions that are relevant to Solaris (and even some that aren't) but still had no joy.
Does anyone know of any logs that I can use to try and identify the issue, or think of something I may be missing.
Regards,
JP

Fantastic!
On the good box I used
/usr/openwin/bin/xterm -display localhost:10
Success Xterminal popped up on Windows machine
from problem box I got
/xterm Xt error: Can't open display: localhost:10
Arrrgh!
But! I noticed something this time.
so I shutdown all putty sessions
deleted my saved entry for the Problem Solaris box
And low and behold - Success
The entry in putty must have been corrupt. When I was connecting to the Solaris box the connection was failing to register in .Xauthority (this is what i noticed!) so, i guess, the key in there was out of date and so... issues.
Thanks for the advice. It opened up a new avenue of investigation which is what I needed.
Regards,
JP

Using getRuntime().exec() with Serlvet to run applications in Solaris

I am currently doing a project which requires the execution of Solaris applications through the use of Java Servlet with this Runtime method - Runtime.getRuntime.exec()
This means that if the client PC tries to access the servlet in the server PC, an application is supposed to be executed and launched on the server PC itself. Actually, the servlet part is not an issue as the main problem is the executing of applications in different platforms which is a big headache.
When I tried running this program on a Windows 2000 machine, it works perfectly fine. However, when I tried it on a Solaris machine, nothing happens. And I mean nothing... no errors, no nothing. I really don't know what's wrong.
Here's the code.
public void doPost(HttpServletRequest request, HttpServletResponse response)
throws ServletException, IOException
response.setContentType("text/html");
PrintWriter out = response.getWriter();
Process process; Runtime runtime = Runtime.getRuntime();
String com= "sh /opt/home/acrobat/";
String program = request.getParameter("program");
try
process = runtime.exec(com);
catch (Exception e)
out.println(e);
It works under Windows when com = "c:\winnt\system32\notepad.exe"
When under Solaris, I have tried everything possible. For example, the launching of the application acrobat.
com = "/opt/home/acrobat"
com = "sh /opt/home/acrobat"I have also tried reading in the process and then printing it out. It doesn't work either. It only works when excuting commands like 'ls'
BufferedReader read = new BufferedReader(new InputStreamReader(process.getInputStream()));Why is it such a breeze to execute prgrams under Windows while there are so many problems under Solaris.
Can some experts please please PLEASE point out the errors and give me some advice and help to make this program work under Solaris! Please... I really need to do this!!
My email address - [email protected]
I appreciate it.
By the way, I'm coding and compiling in a Windows 2000 machine before ftp'ing the .class file to the Solaris server machine to run.

Hi!
I'm no expert, but I think I remember doing something similar a few months ago, and I think the problem was that the UNIX server does not know which shell to use. The solution was this:
Create a script that starts the program on the server, and be sure to include a first row saying:
#!/bin/ksh
to specify which shell to use (in this case Korn shell). Then call the script from Runtime.getRuntime().exec()

How to execute an application in Solaris using Runtime.getRuntime.exec() ?

I am currently doing a project which requires the execution of Solaris applications through the use of Java Servlet with this Runtime method - Runtime.getRuntime.exec()
This means that if the client PC tries to access the servlet in the server PC, an application is supposed to be executed and launched on the server PC itself. Actually, the servlet part is not an issue as the main problem is the executing of applications in different platforms which is a big headache.
When I tried running this program on a Windows 2000 machine, it works perfectly fine. However, when I tried it on a Solaris machine, nothing happens. And I mean nothing... no errors, no nothing. I really don't know what's wrong.
Here's the code.
public void doPost(HttpServletRequest request, HttpServletResponse response)
     throws ServletException, IOException
     response.setContentType("text/html");
     PrintWriter out = response.getWriter();
          Process process;                                                       Runtime runtime = Runtime.getRuntime();
          String com= "sh /opt/home/acrobat/";
          String program = request.getParameter("program");
          try
                    process = runtime.exec(com);
          catch (Exception e)
               out.println(e);
It works under Windows when com = "c:\winnt\system32\notepad.exe"
When under Solaris, I have tried everything possible. For example, the launching of the application acrobat.
com = "/opt/home/acrobat"
com = "sh /opt/home/acrobat"I have also tried reading in the process and then printing it out. It doesn't work either. It only works when excuting commands like 'ls'
BufferedReader read = new BufferedReader(new InputStreamReader(process.getInputStream()));Why is it such a breeze to execute prgrams under Windows while there are so many problems under Solaris.
Can some experts please please PLEASE point out the errors and give me some advice and help to make this program work under Solaris! Please... I really need to do this!!
My email address - [email protected]
I appreciate it.
By the way, I'm coding and compiling in a Windows 2000 machine before ftp'ing the .class file to the Solaris server machine to run.

it is possible that you are trying to run a program that is going to display a window on an X server, but you have not specified a display. You specifiy a display by setting the DISPLAY environment variable eg.
setenv DISPLAY 10.0.0.1:0
To check that runtime.exec is working you should try to run a program that does not reqire access to an X Server. Try something like
cmd = "sh -c 'ls -l > ~/testlist.lst'";
alternatively try this
cmd = "sh -c 'export DISPLAY=127.0.0.1:0;xterm '"
you will also need to permit access to the X server using the xhost + command

Issue:Oracle 8.1.6 running on solaris 9

Hi,
We are planning to upgrade OS version to 9 and DB Version from 8.1.6 to 10.2.0.2 on our Production Box using exp / imp method. In this effort, we will be doing OS upgrade first and will be doing db upgrade after a couple of days.
Are there any issues if the Database version 8.1.6 running on Solaris 9.
Thanx,
KarthicK

Possibly that it can't be done so you need to check that out. My recollection was that you had to be at 8.1.7.4.
Personally I'd install the o/s on a different server, create a nice clean database, and then move the data. Servers are inexpensive and doing this you will likely inherit a boat load of legacy stuff that will cause many many headaches.
Also do not expect any code written last millenium for 8.1.6 to perform well. You should do this first on a test box, several times, and then use it to tune your code.
If you are planning to do this, no testing, on your production box make sure you have a pillow and blanket to bring into work. Oh and a really solid cold backup of the entire system that you have proven will work. You'll need that too.

Solaris 7 to 8 Upgrade with Oracle 8.1.5 (opinions and issues to note)

Hi all,
Think the subject says it all..
I have a Solaris 7 server running on Sparc E250 with Oracle8.1.5 on it. It shows that it s a 64bit system when booting up and it uses normal Solaris filesystem setup (no RAID on it) .
If I were to upgrade it to Solaris 8, what are the issues I need to take note.
Thanks.
Ld

This could be a big headache....but if there is a need to upgrade your OS from 7 to 8
then you should be ready for a good backup plan....meaning, if something goes wrong at any
point in the process, you could easily revert back to your old system environment.
If I were the one to do this, I will do the following.
1) I will make use of mirroring. It will be good if you could do it for the entire filesystem
of your OS and your DB.
2) Once I have a mirror, I will simply detach the other mirror and do my upgrade.
Anything goes wrong, I will simply make use of the other mirror to have my old system back.
When you upgrade your OS from 7 to 8, I had seen lots of problem where it could not complete the
process. And the reason is the size of the Filesystem for Solaris 8 specially the root requires
a big size as to any previous version. i usually just install from scratch. But the problem is if
your filesystem for the OS and Applications like your DB are crossing each other then you may have
to rebuilt your Applications too. Unless you had partition them such that they are totally separated.
When you had succesfully installed your Solaris 8, then your next headache is to make sure
all your applicatiions including your DB will be running fine with the new OS. Usually they
would not as they may need to be patched. So that means you have to check the Oracle websites
as for what are the patches needed to make your Oracle runs properlly in the new OS.
Most important is your backup plan, which is your mirrored-disk of your running environment.
Also you have to make a Plan of everything. Doing this will really take a lot of time.
Good luck. I have done this before.....I'm pretty sure you could do it.
just remember , your back-up plan should always be available anytime.

A few more Solaris questions

I'm sure this isn't the best place to post this, but I currently have a file server running Solaris 11 with a raid-z2 pool and I'm building a new vmware server out of some parts I've managed to get some good deals on might I add.
I'm building on a supermicro x8dt3-f board that has an LSI controller on board with a pair of xeon 5570's and 48gb ram. The processors and ram actually came from a sun blade that was tossed out for recycling, sadly the box had never been opened, but i got these for pennies on the dollar, so i'm happy
I'd like to move the current Solaris file server into a VM on the new ESXi 5.1 host and passthrough the LSI controller.
I've been reading, and reading and reading and i find the more i read, the more questions I have and the less clear some of the answers are getting.
First, from what I gather, Solaris 11 has only been added to the supported guest list in ESXi with the recent esxi 5.1 version as I understand? So all should be fine here? Can someone confirm?
second, i've read about issues with LSI controllers under solaris 11. Is this something that has been addressed in 11.1?
Third, trying to find the best method to conver the physical system to virtual under ESXi
The first thing I plan on doing is backing up my data from the pool, though it will be scattered across a few systems. Then I plan on exporting the pool to move the disks physically to the new controller in the VM. The question I have here, is will the share flags and permissions be retained when I import the pool? or do I have to redo all that?
But then, what's the best method to move it to a VM? The one document I see come up the most is moving a physical solaris system into a zone on another system. Can it be transferred to the global zone? Could I do a new install in a VM, and move the existing install to the global zone, import my pool and call it a daY?
Or do I dd the OS disk, convert the image and drop it into vmware, get the hardware working, and then import the pool?
Or is there a better way? Has anyone got any online docs in mind that may help specifically with this migration? everything i'm finding is scattered, maybe i'm not looking for the right things but I could use some pointers if anyone has suggestions.
I suppose I should note, i have it integrated with an Active directory, this is why i'm worried about permissions being retained when i import the pool
I just want this to go as quickly and smoothly as possible, with as little headache as possible. it's my home setup, so realistically it takes the time it takes as long as things go smooth

I know this may not be the answer you are looking for, but I think you are making it more difficult than it needs to be.
One other option is to leave your Solaris Storage server on the bare metal of this new beast of a machine you are piecing together. Then, use VirtualBox 4.2.6 which is support quite well in Solaris to run whatever virtual machines you where intending ESXi to be used for.
This way you have the fastest possible storage setup without the issues of hardware passthrough. And the fast storage now benefits the VMs running on it. Not to mention the other neat options now of running lzjb compression for the VMs. either using zvols for the VMs, or just virtualbox vdi files sitting on a compressed zfs filesystem.
As far as the LSI 1068E controller goes, their website only shows drivers for Solaris 10. So unless Solaris 11 has the drivers built in, you may not be able to use that controller. Believe me I feel your pain in this one.. I have the d#$$!est time finding good SAS HBAs for Solaris 11.1 and the few I did find had questionable drivers. Areca 1320 cards seem to work well, as do Adaptec 64xx and 68xx raid cards. LSI has a new line that supposedly works with Solaris 11.
My advice is to try a baremetal install of Solaris 11.1 on the new machine and see if you can recognize drives on the LSI controller, if not then use the 6 onboard SATA ports if that is enough for the drives. Otherwise purchase an Areca 1320 which is only like 230 bucks for the 8 port version. http://www.newegg.com/Product/Product.aspx?Item=N82E16816151116R

Jds3 in solaris 10 x86 broken

Hi, i am a novice to unix and have been trying to learn solaris for some time now. I finally have a system that will allow me to install the os with minimum headaches. The problem i have been experiencing is that when i install kde OR SUN GNOME, something related to it causes jds to brake. I get a black background when i log in to jds and can only use some applications. if i try to change the background through the jds tool it crashes. can someone advise me what may be affecting gnome and how to fix it please. (JDS WORKS FINE IF KDE AND ITS DEPENDANCIES NOT INSTALLED)
my system specs:
AMD 3000+ PROC
ASUS A7N8XE-DELUXE MB
1.5 GIG RAM
80 GIG HD
256 MB ATI VID CARD
PINNACLE CAPTURE CARD
2 PROMISE PCI CONTROLLER CARDS(4 80 GIG HD ATTACHED)

I apologize, someone usually responds on this forum
within a day or two, and I'm currently working on
newer opensolaris releases and one of the older linux
based JDS releases, so your combination of JDS3
Solaris with KDE wasn't easily to replicate without
wiping a system that I'm using for something else.
In the future I would suggest that you post such
questions to one of the opensolaris forums:
http://www.opensolaris.org/jive/forum.jspa?forumID=39
start=0
Here is a thread on installing KDE from blastwave:
http://www.opensolaris.org/jive/thread.jspa?messageID=
26374�B
If you think this combination of JDS3 and KDE would
be useful to others, I'd appreciate if you could post
your solution. Even "newbies" can contribute.nitz thank you for responding, it has been frustrating trying to get this beast up and running. First i had resolved the problem by cleanining up my /etc/profile file as there were duplicate path statements and statements in incorrect order, caused by me following stuff i got over the internet and not realizing that solaris puts some of those statements automatically into env path. This was working fine until i had to reinstall thus mu handle REINSTALLED :-)
because i forgot to create a hard link between /usr and /opt/csw which my soft link got trashed. now after reinstalling the OS and programs from blastwave the black background is back in JDS. any help or suggestions will be appreciated. on another note i have been able to fix the sig 11 error associated with kde as follows;
*make sure /opt/csw/kde-gcc/bin and /opt/csw/bin is at the end of the path list, because the compilier was interfering with the css compilier.
also do not install qt from blastwave as kde also has its own qt files built in. when i installed qt for use with another program i get the sig 11 error and kde will not login or work properly, but when i just pkgrm /opt/CSWqt , everything worked normal again.
Thanks again for responding.

[Solved] Arch Linux in a Solaris branded Zone

Hello,
After having read this article at the Genunix WiKi, I 'd very much like to install Arch Linux (http://204.152.191.100/wiki/index.php/I … anded_zone) in a Solaris branded Zone. However, the Arch Linux file to download is about two years old now, so I 'd like to roll one with a bit more recent kernel.
I just wonder what might be the procedure to do so.
a) Is it just unpacking the ISO inside the zone and let it enroll whilst booting?, or
b) is it advisable to install the old package and upgrade from there?
I am quite curious to learn how this works.
TIA, Algey
Last edited by algernonz (2011-11-28 18:42:31)

Hi,
it does indeed sound like an interesting little project. You should, however, revert to your procedure (a) and use one of the recent ISO images the Arch team has updated this year.
Using that old tar will give you numerous headaches when trying to upgrade. Doing a simple upgrade from that old file will most probably break the system, thats why they took the effort to provide the new ones.
Good luck.

A.M.P.S. - problems seting up on solaris 10

help,
I am trying to setup A.M.P.S. ... I have managed to configure this setup on a windows 2003 server (with headaches) but got it working. (eventually) ... i have joined this list in the hope that someone out there will be able to help.
I am a Windows engineer with little unix exspereince (i've just bought the complete solaris 10 reference) - which has helped, but after managing to get apache 2 up & working i then installed php & related packages but was unable to get the "info" script to publish the php info ...
i will happily copy and parts of the httpd.conf file or the php.ini file that you wish to see ...
I am willing to reinstall the o/s then start from scratch adding the packages nesacery to get this setup working ... I am running solaris 10 (sparc) on a sunblade 1000 .... using php 5.0.4 ... and would ideally like a rundown similar to this on solaris 8 ...
http://ampubsvc.com/~meljr/a2m4p5s8.html
I haven't even looked at the mysql setup but beleive it may be possible to configure this at the sametime as php.
please help ... before i go insane ...

Sorry, Stephen, but I don't think you're going to get much help in this particular group of forums. They're strictly for issues on lumps of hardware.
Your question seems to be 100% software.
You just happen to be running Sun's OS on a Sun workstation.
You may get more information out of these <i>other</i> Sun forums:
http://forum.sun.com/index.jspa
Use the search function over there, keywords <u>apache + php</u>.

Porting Solaris 7 audio driver to Solaris 8.

I am trying to load solaris 7 audio driver on solaris 8. It is not working.
Also I have recompiled drivers codes on solaris 8 still not working. Something must have changed in the audio interface from solaris 8 to solaris 7. Can anyone help me to find these changes. Thanks

This could be a big headache....but if there is a need to upgrade your OS from 7 to 8
then you should be ready for a good backup plan....meaning, if something goes wrong at any
point in the process, you could easily revert back to your old system environment.
If I were the one to do this, I will do the following.
1) I will make use of mirroring. It will be good if you could do it for the entire filesystem
of your OS and your DB.
2) Once I have a mirror, I will simply detach the other mirror and do my upgrade.
Anything goes wrong, I will simply make use of the other mirror to have my old system back.
When you upgrade your OS from 7 to 8, I had seen lots of problem where it could not complete the
process. And the reason is the size of the Filesystem for Solaris 8 specially the root requires
a big size as to any previous version. i usually just install from scratch. But the problem is if
your filesystem for the OS and Applications like your DB are crossing each other then you may have
to rebuilt your Applications too. Unless you had partition them such that they are totally separated.
When you had succesfully installed your Solaris 8, then your next headache is to make sure
all your applicatiions including your DB will be running fine with the new OS. Usually they
would not as they may need to be patched. So that means you have to check the Oracle websites
as for what are the patches needed to make your Oracle runs properlly in the new OS.
Most important is your backup plan, which is your mirrored-disk of your running environment.
Also you have to make a Plan of everything. Doing this will really take a lot of time.
Good luck. I have done this before.....I'm pretty sure you could do it.
just remember , your back-up plan should always be available anytime.

Solaris root password & installer

I have installed oracle solaris 11 exp, but :
I have difficulties finding root password, have tried solaris & my user password but it's not going,
during the installation I have used solaris passwrd for the partition manager and it worked, but no later,
I also would like to ERASE ALL PASSWORDS of the system, ... give password, give password, give password, it gives you a headache,
and by the way, do I have to buy an application installer or I have with the package?
thank in advance,
max.

root is a role by default in Solaris 11, so i don't think it actually have a password, or am i wrong here? You can set a password by running pfexec passwd rood from the user who you created as an administrative user..
.7/M.

Can Solaris run on AMD Sempron

I am considering evaluating Solaris 10.
I have a new motherboard with an AMD Sempron CPU.
I currently run WinXP on a 30 GB hard drive, partitioned as two 15GB drives.
I need to re-install the operating system as I am replacing my current motherboard.
This will allow me to partition one 15 GB drive into two smaller drives, probabaly one 10 GB and one 5GB
Is it possible to configure this as a dual boot machine, one booting into Windows, and one booting into Solaris.
My concern with Solaris is how hard will it be to learn and configure Solaris, and how many programs that I currently run on Windows can run on Solaris.
I use sevral EDA programs on Windows, Schematic Capture, FPGA design programs, PCB layout.
Thanks in advance for your help.

Well I suppose it will run on AMD Sempron, I haven't tried, but since AMD Sempron is an i386 compatible CPU, It will have no problem...
About using windows programs on Solaris, you should use Wine to use the windows executables. Watch www.winehq.com.
About the ease of configuration of Solaris, it all depends on haw much time you want to dedicate to it. It's not as simple as configuring Windows XP but it doesn't give you so much headaches: things just work if you do the correct.

Confused about ZFS filesystems created with Solaris 11 Zone

Hello.
Installing a blank Zone in Solaris *10* with "zonepath=/export/zones/TESTvm01" just creates one zfs filesystem:
+"zfs list+
+...+
+rzpool/export/zones/TESTvm01 4.62G 31.3G 4.62G /export/zones/TESTvm01"+
Doing the same steps with Solaris *11* will ?create? more filesystems:
+"zfs list+
+...+
+rpool/export/zones/TESTvm05 335M 156G 32K /export/zones/TESTvm05+
+rpool/export/zones/TESTvm05/rpool 335M 156G 31K /rpool+
+rpool/export/zones/TESTvm05/rpool/ROOT 335M 156G 31K legacy+
+rpool/export/zones/TESTvm05/rpool/ROOT/solaris 335M 156G 310M /export/zones/TESTvm05/root+
+rpool/export/zones/TESTvm05/rpool/ROOT/solaris/var 24.4M 156G 23.5M /export/zones/TESTvm05/root/var+
+rpool/export/zones/TESTvm05/rpool/export 62K 156G 31K /export+
+rpool/export/zones/TESTvm05/rpool/export/home 31K 156G 31K /export/home"+
I dont understand why Solaris 11 is doing that. Just one FS (like in Solaris 10) would be better for my setup. I want to configure all created volumes by myself.
Is it possible to deactivate this automatic "feature"?

There are several reasons that it works like this, all guided by the simple idea "everything in a zone should work exactly like it does in the global zone, unless that is impractical." By having this layout we get:
* The same zfs administrative practices within a zone that are found in the global zone. This allows, for example, compression, encryption, etc. of parts of the zone.
* beadm(1M) and pkg(1) are able to create boot environments within the zone, thus making it easy to keep the global zone software in sync with non-global zone software as the system is updated (equivalent of patching in Solaris 10). Note that when Solaris 11 updates the kernel, core libraries, and perhaps other things, a new boot environment is automatically created (for the global zone and each zone) and the updates are done to the new boot environment(s). Thus, you get the benefits that Live Upgrade offered without the severe headaches that sometimes come with Live Upgrade.
* The ability to have a separate /var file system. This is required by policies at some large customers, such as the US Department of Defense via the DISA STIG.
* The ability to perform a p2v of a global zone into a zone (see solaris(5) for examples) without losing the dataset hierarchy or properties (e.g. compression, etc.) set on datasets in that hierarchy.
When this dataset hierarchy is combined with the fact that the ZFS namespace is virtualized in a zone (a feature called "dataset aliasing"), you see the same hierarchy in the zone that you would see in the global zone. Thus, you don't have confusing output from df saying that / is mounted on / and such.
Because there is integration between pkg, beadm, zones, and zfs, there is no way to disable this behavior. You can remove and optionally replace /export with something else if you wish.
If your goal is to prevent zone administrators from altering the dataset hierarchy, you may be able to accomplish this with immutable zones (see zones admin guide or file-mac-profile in zonecfg(1M)). This will have other effects as well, such as making all or most of the zone unwritable. If needed, you can add fs or dataset resources which will not be subject to file-mac-profile and as such will be writable.

MPxIO vs PowerPath for EMC SAN - Solaris 10

I'm just wondering what everyone's experience has been with either option. We are in the process of upgrading to a T5120 server using Solaris 10 and Oracle 9i (we can't upgrade yet due to limitations in our current blood bank software) and connect to an EMC SAN. Our vendor says that his preference is PowerPath because he's heard of systems hanging up when MPxIO option is used. Any ideas?

I'm not sure what kind of commentary you are looking for.
We use Powerpath on Solaris 10 (x86 now, but in the past on T5120s) for multipathing to LUNs on EMC VNX and CX4 SANs. We followed the EMC Powerpath setup guide for Solaris 10 exactly.
After disabling MPxIO, Powerpath was able to take control. Some older versions of powerpath had a bug where the pseudo names for LUNs would change causing some headaches.
We keep a mapping of LUN to pseudo name to solaris dsk labels
If you use ZFS I have found a 1:1 match between the zfs_max_vdev_pending value and the max queued-IOs value shown in Powerpath
When we've had to fail/trespass LUNs over to other SPs on the EMC SANs, Powerpath has handled this elegantly with the expected warnings in /var/adm/messages
I recall that we had to explicitly set the powerpath options for CLARiiON and VNX to managed and the policy to claropt
When adding LUNs, to see all paths there is a routine to go through with cfgadm, devfsadm and the powerpath utilities. We use qlogic HBAs and the qlc leadville driver.

Solaris 10 headaches

Similar Messages

Maybe you are looking for