[zones-discuss] zone create on sol10u6

2008-12-01 Thread Maciej Browarski
Hello,
I try to install zone on zfs file system using solaris 10 update 6.
I have separate storage pool which is names zone. (mount in /zone).
So I type:
#zonecfg -z new
zonecfg:new create
zonecfg:new set zonepath=/zone/new
zonecfg:new exit
#zoneadm -z new install
After that I don't see zone/new filesystems. On Nevada systems I always 
have separate zfs.
In man documentation from zoneadm I read that option -x means that don't 
use  zfs filesystem.
Why is this don't working with solaris10u6 ?

Regards


-- 
Maciej Browarski


___
zones-discuss mailing list
zones-discuss@opensolaris.org


Re: [zones-discuss] zone create on sol10u6

2008-12-01 Thread Enda O'Connor
On 12/01/08 13:36, Maciej Browarski wrote:
 Hello,
 I try to install zone on zfs file system using solaris 10 update 6.
 I have separate storage pool which is names zone. (mount in /zone).
 So I type:
 #zonecfg -z new
 zonecfg:new create
 zonecfg:new set zonepath=/zone/new
 zonecfg:new exit
 #zoneadm -z new install
 After that I don't see zone/new filesystems. On Nevada systems I always 
 have separate zfs.
 In man documentation from zoneadm I read that option -x means that don't 
 use  zfs filesystem.
 Why is this don't working with solaris10u6 ?
 
 Regards
 
 
Hi
Could you clarify what you mean by I don't see zone/new filesystems
is it that it's not in zfs list, i.e. how did you create /zone/new to 
start with.

Enda

-- 
Enda O'Connor x19781  Software Product Engineering
Patch System Test : Ireland : x19781/353-1-8199718
___
zones-discuss mailing list
zones-discuss@opensolaris.org


Re: [zones-discuss] zone create on sol10u6

2008-12-01 Thread bob netherton
  No Solaris 10u6 does support zones on zfs, but not as cleanly as nevada,
  there are some differences, as new features are integrated into nevada,
  then later they might get backported to soalris updates.

Reminds me of the car commercial - the suggested price is blah,
nicely appointed for blah + $$.   As Enda says, zfs and zones
are available, but will be nicely appointed in a future release.

the automatic creation of the datasets does not work.  use of zfs
snapshot and clones when cloning a zone do not work.for LU the 
zoneroots must
be datasets (file systems).

   On s10u6 you
  have to create the zonepath yourself, either mkdir or zfs create,
  setting appropriate permissions too,

If the zoneroot is no present, a mkdir will be done for you.   and
the permissions will be set correctly.   this has been that way since
u1 or u2 (u1 IIRC).  A zfs dataset will not be automatically created
like it does in nevada.  and if you do create the zoneroot, make sure
and set the permissions to 700 like Enda says - this is only done
automatically when we mkdir on your behalf.  if you do it then you must 
do it right.


zone1 is on ufs.   zone2 will be a similar zone on zfs.
/dune/zones/s10u6 is already a dataset (file system).   this will
(should) be live upgradable.

# zonecfg -z zone2 create -t zone1
# zonecfg -z zone2 set zonepath=/dune/zones/s10u6/zone2
# zfs create dune/zones/s10u6/zone2
# chmod 700 /dune/zones/s10u6/zone2
# zoneadm -z zone2 install
Preparing to install zone zone2.
Creating list of files to copy from the global zone.

 time passes, stuff happens.



bob
___
zones-discuss mailing list
zones-discuss@opensolaris.org


Re: [zones-discuss] zone create on sol10u6

2008-12-01 Thread Enda O'Connor
On 12/01/08 14:09, bob netherton wrote:
   No Solaris 10u6 does support zones on zfs, but not as cleanly as nevada,
   there are some differences, as new features are integrated into nevada,
   then later they might get backported to soalris updates.
 
 Reminds me of the car commercial - the suggested price is blah,
 nicely appointed for blah + $$.   As Enda says, zfs and zones
 are available, but will be nicely appointed in a future release.
 
 the automatic creation of the datasets does not work.  use of zfs
 snapshot and clones when cloning a zone do not work.for LU the 
 zoneroots must
 be datasets (file systems).

So just to clarify, the next Lu patch to release ( 121430-30/121431-31, 
saprc/x86 ) that releases will relax that restriction in terms of 
upgrade of zones on zfs, basically allowing zones on zfs to live 
anywhere, ie top level dataset, lower level dataset, or a directory ina 
dataset
ie
zfs create zones/zonehome
zonecfg : set path=/zones/zonehome/myzone-01

Now the only caveat is that you must also apply this patch to the ABE, 
prior to booting it, ie
apply LU patch 121430-30 to live system
run lucreate
run luupgrade ( if upgrading from pre u6 )
apply 121430-30 to the new BE using luupgrade -t ( or patchadd -R )
The last step MUST be done prior to luactivate, so even if you are 
creating a BE for patching ie skipping the luupgrade step, you still 
need to apply the patch after lucreate and before luactivate.
Not doign so can result in not being able to boot back to original BE 
once new BE is activated.

Enda
 
On s10u6 you
   have to create the zonepath yourself, either mkdir or zfs create,
   setting appropriate permissions too,
 
 If the zoneroot is no present, a mkdir will be done for you.   and
 the permissions will be set correctly.   this has been that way since
 u1 or u2 (u1 IIRC).  A zfs dataset will not be automatically created
 like it does in nevada.  and if you do create the zoneroot, make sure
 and set the permissions to 700 like Enda says - this is only done
 automatically when we mkdir on your behalf.  if you do it then you must 
 do it right.
 
 
 zone1 is on ufs.   zone2 will be a similar zone on zfs.
 /dune/zones/s10u6 is already a dataset (file system).   this will
 (should) be live upgradable.
 
 # zonecfg -z zone2 create -t zone1
 # zonecfg -z zone2 set zonepath=/dune/zones/s10u6/zone2
 # zfs create dune/zones/s10u6/zone2
 # chmod 700 /dune/zones/s10u6/zone2
 # zoneadm -z zone2 install
 Preparing to install zone zone2.
 Creating list of files to copy from the global zone.
 
  time passes, stuff happens.
 
 
 
 bob
 ___
 zones-discuss mailing list
 zones-discuss@opensolaris.org


-- 
Enda O'Connor x19781  Software Product Engineering
Patch System Test : Ireland : x19781/353-1-8199718
___
zones-discuss mailing list
zones-discuss@opensolaris.org


Re: [zones-discuss] zone create on sol10u6

2008-12-01 Thread Ralf Weber
Moin!

On 01.12.2008, at 17:36, Enda O'Connor wrote:
 So just to clarify, the next Lu patch to release  
 ( 121430-30/121431-31,
 saprc/x86 ) that releases will relax that restriction in terms of
 upgrade of zones on zfs, basically allowing zones on zfs to live
 anywhere, ie top level dataset, lower level dataset, or a directory  
 ina
 dataset
 ie
 zfs create zones/zonehome
 zonecfg : set path=/zones/zonehome/myzone-01
So what are the restrictions? I thought you where able to use zones  
and LU in update 6. Now creating my own zfs pool or file system in the  
root pool isn't a problem. But what is the required setup to use in  
10u6 to get it live upgradable without the patch?

So long
-Ralf
---
Ralf Weber


___
zones-discuss mailing list
zones-discuss@opensolaris.org


[zones-discuss] Any solaris native UI tool/WEB interface available to manage Solaris Containers/Zones?

2008-12-01 Thread Challa, Narsimha Reddy (STSD-HYD)
Hi,

 Could anybody please let me know whether there are any solaris native 
tools or web interface available to manage solaris containers?

Regards,
-Narsimha
___
zones-discuss mailing list
zones-discuss@opensolaris.org


Re: [zones-discuss] Any solaris native UI tool/WEB interface available to manage Solaris Containers/Zones?

2008-12-01 Thread Nicolas Dorfsman

Hi,

Le 1 déc. 08 à 18:24, Challa, Narsimha Reddy (STSD-HYD) a écrit :

 Could anybody please let me know whether there are any solaris  
 native tools

zonecfg and zoneadm ?

There's also zonemgr to help.

 or web interface available to manage solaris containers?

Sun xVM something.




___
zones-discuss mailing list
zones-discuss@opensolaris.org


Re: [zones-discuss] zone create on sol10u6

2008-12-01 Thread Enda O'Connor
Steve Lawrence wrote:
 Now the only caveat is that you must also apply this patch to the ABE, 
 prior to booting it, ie
 apply LU patch 121430-30 to live system
 run lucreate
 run luupgrade ( if upgrading from pre u6 )
 apply 121430-30 to the new BE using luupgrade -t ( or patchadd -R )
 The last step MUST be done prior to luactivate, so even if you are 
 creating a BE for patching ie skipping the luupgrade step, you still 
 need to apply the patch after lucreate and before luactivate.
 Not doign so can result in not being able to boot back to original BE 
 once new BE is activated.
 
 Hey Enda,
 
 For clarity,  once I'm up and running on a BE with u6 + 121430-30, if I
 create another ABE for the purpose of patching or upgradeing, do I need to
 re-apply the latest LU patch to it again?
 
 Example:
 
 (running u5, the decribed scenerio):
 patchadd 121430-30
 lucreate ABE
 luupgrade ABE u6 (or perhaps patch up to some u6 KU)
 luupgrade ABE -t 121430-30
 luactivate ABE
 reboot

yes above is necessary
 
 Now I want to patch to th next KU, or upgrade to U7)
 (running u6 or some equivalent ku)
 patchadd latest u6 lu patch
 lucreate BBE
 luupgrade BBE to next KU patch (or maybe update to u7)
 (do I need to add latest lu patch to BBE??)
if you upgrade to u7, you would be ok, if you creating an ABE to patch, 
id would be ok too, as you'd get the patched LU bits.

Just upgrading to u6 is the problem, as you'd end up with the old u6 
lu bits.

Enda
 luactivate BBE
 reboot
 
 Comments?
 
 -Steve L.
 
 ___
 zones-discuss mailing list
 zones-discuss@opensolaris.org

___
zones-discuss mailing list
zones-discuss@opensolaris.org


Re: [zones-discuss] zone create on sol10u6

2008-12-01 Thread Edward Pilatowicz
On Mon, Dec 01, 2008 at 02:36:08PM +0100, Maciej Browarski wrote:
 Hello,
 I try to install zone on zfs file system using solaris 10 update 6.
 I have separate storage pool which is names zone. (mount in /zone).
 So I type:
 #zonecfg -z new
 zonecfg:new create
 zonecfg:new set zonepath=/zone/new
 zonecfg:new exit
 #zoneadm -z new install
 After that I don't see zone/new filesystems. On Nevada systems I always
 have separate zfs.
 In man documentation from zoneadm I read that option -x means that don't
 use  zfs filesystem.
 Why is this don't working with solaris10u6 ?


currently only nevada creates a new zfs filesystem for zones (assuming
that the parent directory is a zfs filesystem).  this functionality
is currently planned to be included in s10u7.  (it's covered via the
following bugids: 6383119 and 6716853.)

ed
___
zones-discuss mailing list
zones-discuss@opensolaris.org


Re: [zones-discuss] Any solaris native UI tool/WEB interface available to manage Solaris Containers/Zones?

2008-12-01 Thread Michael Barrett
Currently Sun Management Center 4.0's Container Manager module is what 
Sun has in terms of a web UI interface for zones.  I believe there are 
webadmin plugins out in the outsource land as well.  SunMC allows you to 
discover, build, destroy, allocate resources, trend utilization, move, 
migration over HTTP, etc.

Anyone can download and use SunMC 4.0 for free in production or 
elsewhere for as long as they want.  Should you require support or 
assistance from Sun we asked they you purchase a support license and 
right to use license at that time.

http://www.sun.com/software/products/sunmanagementcenter/get.html

Sun will re-instrument these features and add more in xVM Ops Center 2.x 
during the first half of 2009.

Thanks,
Mike



Challa, Narsimha Reddy (STSD-HYD) wrote:
 Hi,
 
  Could anybody please let me know whether there are any solaris native 
 tools or web interface available to manage solaris containers?
 
 Regards,
 -Narsimha
 ___
 zones-discuss mailing list
 zones-discuss@opensolaris.org
___
zones-discuss mailing list
zones-discuss@opensolaris.org


Re: [zones-discuss] DVD/CD burning in a Non Global Zone

2008-12-01 Thread Edward Pilatowicz
thanks for the info.
i've saved a copy of this discussion in 4964815.
ed

On Fri, Nov 28, 2008 at 01:29:27PM +0100, Joerg Schilling wrote:
 Edward Pilatowicz [EMAIL PROTECTED] wrote:
 
  i don't know how linus differentiates between good and bad scsi
  commands, but from a zones perspective, if a zone has access to a cdrom
  device on a SCSI bus, then cdrecord should be able to send any SCSI
  command that it wants to that device (including vendor specific
  commands).  but cdrecord should be prevented from sending any SCSI
  commands that access other devices on the same bus.  ideally, it should
  also be prevented from abusing commands that could reset or hang the
  entire bus, but this needs more investigation.
 
 This sounds reasonable. As you are able to delegate drives, you should 
 be able to delegate the permission to send any SCSI command to them as well.
 
 Any SCSI command _may_ hang the SCSI bus under some conditions. But similar 
 things may happen with the whole machine.
 
 You should know that A program that is able to send random SCSI commands to a 
 target usually is able to replace the firmware. If you like to prevent this,
 you just should not give this permission into a zone.
 
 BTW: There are target device resets and there are bus device resets. 
 
  i'm not familiar enough with SCSI commands to know how feasible it would
  be to implement such a filter.  if it's not technically possible, then
  the second option is to require that a zone be given access to all the
  devices on a scsi bus before we allow it to issue SCSI commands to that
  bus.
 
 Sending a bus device reset is not a SCSI command. There is a reset line in 
 the 
 bus.
 
 Sending a taarget reset is done via a SCSI message that is usually not 
 directly 
 sendable from user space.
 
 Jörg
 
 -- 
  EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin
[EMAIL PROTECTED](uni)  
[EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/
  URL:  http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily
___
zones-discuss mailing list
zones-discuss@opensolaris.org