Re: [zones-discuss] Getting local zones patches back in sync with the global

2007-03-21 Thread EndaO'Connor

Rich Pedano wrote:


Hi,

My customer has a system with 3 zones.

Zone 1 was created prior to global zone patching.
Zone 2 was created prior to global zone patching.
Zone 3 was created after global zone patching.

Zone 3 inherited the patches in the global as expected.

Zone 1 and 2's  zone path was not available at the time of global zone patching 
because the zone paths are located
on external disks that were unmounted. The external disks that support the zone paths were unmounted  because the 
patches were installed in single user mode per the patch cluster README.


As a result Zone 1 and 2 did not inherit the patches installed in the global. 


Zone 2 was re-installed and the patches propagated from the global as expected.

Are there any other workarounds to get a local zone's patches in sync with the 
global  other than re-installation ?

Thank you,

-rich


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

also not clear as to why patchadd dependency checks failed, patchadd 
should have checked dependencies in all zonesand noticed that zone 1  2 
were not available.


Would it be possible to re test this using a patch not installed and run 
patchadd -a,
the -a does a dryrun so will not actually install anything, I would like 
to see the output from such a scenario where the zonepath was not available


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


Re: [zones-discuss] zoneadm install

2007-06-04 Thread EndaO'Connor

F.V.(Phil)Porcella wrote:


***Im getting some strange errors from zoneadmin install
bash-3.00# zoneadm -z CIS2 install
Preparing to install zone CIS2.
Creating list of files to copy from the global zone.
Copying 213 files to the zone.
Initializing zone product registry.
ERROR: Read-only file system: cannot create zone product registry text database file 
/zones/CIS2/root/var/sadm/install/contents
ERROR: cannot initialize zone product registry at /zones/CIS2/root
ERROR: cannot copy and initialize packages from global zone to local zone CIS2 path 
/zones/CIS2/root
ERROR: cannot create zone boot environment CIS2
zoneadm: zone 'CIS2': '/usr/lib/lu/lucreatezone' failed with exit code 74.
*** my configuration file looks like this:
bash-3.00# zonecfg -z CIS2
zonecfg:CIS2 export
create -b
set zonepath=/zones/CIS2
set autoboot=true
add inherit-pkg-dir
set dir=/lib
end
add inherit-pkg-dir
set dir=/platform
end
add inherit-pkg-dir
set dir=/sbin
end
add inherit-pkg-dir
set dir=/usr
end
add inherit-pkg-dir
set dir=/opt
end
add inherit-pkg-dir
set dir=/var
end
add net
set address=156.12.1.129
set physical=eri0
end
zonecfg:CIS2 exit
bash-3.00#
*** Im patched up to the solaris cluster that came out last week, so
*** Any ideas?


This message posted from opensolaris.org
___
zones-discuss mailing list
zones-discuss@opensolaris.org
 


Hi
Create -b creates a whole root zone, esssentially /lib /sbin/ platform 
and /usr are no longer inherited, but then in the zone config you 
specify these as being inherited anyway,
esentially drop the -b, now the real problem is the last inherit /var,  
var must be writqable, and is what is causing the failure.

so
1 no need to use -b to create,
2 no need to specify inherit for /lib /sbin /platform and usr once -b is 
not used

3 lose the inherit of /var

Why are you inheriting /var, might be able to give better advise once we 
understand issue


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


Re: [zones-discuss] Question on new Nov patch cluster for a zoned system.

2007-12-19 Thread EndaO'Connor
Stewart, Tom L. wrote:
 All,

 I have a T2000 with a couple of zones and when I tried to install the
 patch cluster from Nov 07 in the Global zone, it failed.
   

any idea why it failed, are there logs, patchadd output?
it should not have failed
 According to the CLUSTER_README

 WARNING!! IT IS HIGHLY RECOMMENDED that the installation of this patch 
 cluster be performed in single-user mode (Run Level S).

 So I shutdown the zones and rebooted to single user mode. The
 install_cluster process took less than a minute and gave errors on
 trying to patch the first zone.

 Do I need to run the patch in normal mode or is this some sort of bug?
   
if you boot to Single User Mode ( as opposed to dropping to Single User 
from Multi User ) then you need to run the filesystem/local:default SMF, 
or mountall -l
As otherwise the zone's zonepath will not be mounted.

What errors did you see?
Enda
 Thank you,
 Tom
 ___
 zones-discuss mailing list
 zones-discuss@opensolaris.org
   

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


Re: [zones-discuss] Question on new Nov patch cluster for a zoned system.

2007-12-19 Thread EndaO'Connor
Hi
If the cluster has 127111 or 127112 ( sparc | x86 KU ), and if 120011-14 
is also being installed, then a reboot is required after applying 
120011-1 4, otherwise corruption will occur,
if the cluster does contain 127111|127113 then get the latest cluster as 
these have been removed, pending a fix to the patch tools.

Enda
Stewart, Tom L. wrote:
 Ahh,

 Now it makes sense

 The error was ( I had to change the server name to protect the
 innocent).

 zoneadm: /zones/x: No such file or directory
 could not verify zonepath /zones/x because of the above errors.
 zoneadm: zone xx failed to verify

 I didn't even think of that as I did do a shutdown and booted back to
 single user.

 Many thanks,

 Tom

 -Original Message-
 From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] 
 Sent: Wednesday, December 19, 2007 10:50 AM
 To: Stewart, Tom L.
 Cc: zones-discuss@opensolaris.org
 Subject: Re: [zones-discuss] Question on new Nov patch cluster for a
 zoned system.

 Stewart, Tom L. wrote:
   
 All,

 I have a T2000 with a couple of zones and when I tried to install the
 patch cluster from Nov 07 in the Global zone, it failed.
   
 

 any idea why it failed, are there logs, patchadd output?
 it should not have failed
   
 According to the CLUSTER_README

 WARNING!! IT IS HIGHLY RECOMMENDED that the installation of this patch
 

   
 cluster be performed in single-user mode (Run Level S).

 So I shutdown the zones and rebooted to single user mode. The
 install_cluster process took less than a minute and gave errors on
 trying to patch the first zone.

 Do I need to run the patch in normal mode or is this some sort of bug?
   
 
 if you boot to Single User Mode ( as opposed to dropping to Single User 
 from Multi User ) then you need to run the filesystem/local:default SMF,

 or mountall -l
 As otherwise the zone's zonepath will not be mounted.

 What errors did you see?
 Enda
   
 Thank you,
 Tom
 ___
 zones-discuss mailing list
 zones-discuss@opensolaris.org
   
 


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

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