if you apply the patch 118833-36, you will notice that
/etc/init.d/sysetup*118833-36 gets created
Replace the default with it and it will fix your issue

Visibly facilities were missed in the old version

*** sysetup2    Mon Apr  2 22:25:30 2007
--- sysetup     Mon Apr  2 22:25:05 2007
***************
*** 1,9 ****
 #!/sbin/sh
 #
! # Copyright 2006 Sun Microsystems, Inc.  All rights reserved.
 # Use is subject to license terms.
 #
! # ident       "@(#)sysetup    1.27    06/06/05 SMI"

 if [ -z "$_INIT_PREV_LEVEL" ]; then
       set -- `/usr/bin/who -r`
--- 1,9 ----
 #!/sbin/sh
 #
! # Copyright 2004 Sun Microsystems, Inc.  All rights reserved.
 # Use is subject to license terms.
 #
! # ident       "@(#)sysetup    1.25    04/07/15 SMI"

 if [ -z "$_INIT_PREV_LEVEL" ]; then
       set -- `/usr/bin/who -r`
***************
*** 20,25 ****
--- 20,62 ----
 [ -d /etc/tm ] && /usr/bin/cat /etc/tm/* 2>/dev/null

 #
+ # Set default scheduling class
+ #
+ if [ -f /etc/dispadmin.conf ] && [ -x /usr/sbin/dispadmin ] && \
+      [ -x /usr/bin/priocntl ] && \
+      [ "${_INIT_ZONENAME:=`/sbin/zonename`}" = "global" ]; then
+       ERROR="$0: cannot set default scheduling class to "
+       DISPADMIN_D=`/usr/sbin/dispadmin -d`
+
+       if [ $? -eq 0 ]; then
+               #
+               # Inform the kernel about the default scheduling class.
+               #
+               dispadmin -u
+
+               DEFAULT_SCHEDULER=`echo $DISPADMIN_D | \
+                       /usr/bin/awk '{ print $1 }'`
+
+               /usr/bin/priocntl -s -c $DEFAULT_SCHEDULER \
+                       -i all >/dev/null 2>&1
+
+               if [ $? -ne 0 ]; then
+                       echo $ERROR $DEFAULT_SCHEDULER
+               else
+                       #
+                       # Also need to move init process explicitly
+                       # because it was ignored by "-i all".
+                       #
+                       /usr/bin/priocntl -s -c $DEFAULT_SCHEDULER \
+                               -i pid 1 >/dev/null 2>&1
+                       if [ $? -ne 0 ]; then
+                               echo $ERROR $DEFAULT_SCHEDULER
+                       fi
+               fi
+       fi
+ fi
+
+ #
 # Run rctladm to configure system resource controls based on the settings
 # previously saved by rctladm.  See rctladm(1m) for instructions on how to
 # modify resource control settings.
***************
*** 26,29 ****
--- 63,75 ----
 #
 if [ -f /etc/rctladm.conf ] && [ -x /usr/sbin/rctladm ]; then
       /usr/sbin/rctladm -u
+ fi
+
+ #
+ # Run pooladm to configure system pools.  See pooladm(1m) for more
information.
+ #
+ if [ -f /etc/pooladm.conf ] && [ -x /usr/sbin/pooladm ] && \
+     [ "$_INIT_ZONENAME" = "global" ]; then
+       /usr/sbin/pooladm -e
+       /usr/sbin/pooladm -c
 fi

Regards




On 4/2/07, steeles <[EMAIL PROTECTED]> wrote:

I followed sun's doc to setup a zone.

I enable and save conf with pooladm -e, and pooladm -s and I configure
zone with new pool and pset.

After I reboot server, zone is running, zonecfg -z zone info. It shows my
pool name(not default)

but if I run pooladm to check what pool I have. It tell me

"facility is not active"

But the fact is my zone is up running with my new pool name. If I do

#pooladm -e
#pooladm

it didn't give the pool that I setup before. It only shows default pool.


So my questions are.
1. Did I setup correctly?
2.pool is not necessary to run at startup?
3. if shutdown zone and bring it up again, it tells me that pool can't be
located and will use default pool instead. What did I do wrong?

Thanks.


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




--
Gael Martinez
_______________________________________________
zones-discuss mailing list
zones-discuss@opensolaris.org

Reply via email to