OpenSolaris

  subsites   code review   repo   packages   bugs   defect   polls   planet
You are not signed in. Sign in or register.

Release Notes for the OpenSolaris 2008.05 Release

For additional information on the following issues, look up the bug number in Bugzilla.

Languages
Translation for this page

This page lists the following Release Notes for the OpenSolaris 2008.05 Release :

Important Package Update Instructions for All Users.

6696145 - Install Causes Panics on Intel 965 With GM965 Chipset.

64 - Installer, but not Desktop, Allows Empty Passwords.

1045 - beadm create Fails When Run as User.

1109 - Locale Information Inaccurate in Installer Summary Panel.

1156 - Booting a BE That Was Left Mounted Before the Reboot Fails.

1333 - Creating BE and Renaming BE Fails to Handle Subordinate Legacy-Mounted Datasets.

1334 - Compiz Hangs System that Uses Intel Graphics Card i915 or i965GM.

1472 - Creating a new BE From a Mounted BE Results in Incorrect Mount Points for Subordinate Datasets.

1559 - GRUB Entry for Windows Incorrect From USB Stick Install.

1565 - Nautilus Process Hangs After Gnome Login.

1771 - Install will fail if rpool ZFS pool already exists.

1685 - Renaming a Busy Boot Environment can Fail.

6676221 - setxkbmap Can't Locate Rules file for Xorg.

6687328 - vim Command Cannot Display Multibyte Characters.

6690194 - fdisk(1M) Doesn't Clear VTOC When Partition Starts After, or Exceeds, 4 Gbytes.

6690824 - fdisk(1M) Corrupts Solaris VTOC, if Linux Swap Primary Partition is Defined.

6694783 - Input Method Duplicate Causes IDE Crash.

2900 - Firefox dumps core in NV95.

Important Package Update Instructions for All Users

A number of issues in the Image Packaging System (IPS) have been identified and resolved since the initial release of OpenSolaris 2008.05. As a result, the following steps should be performed immediately following an OpenSolaris installation and boot in order to ensure correct operation of the software packaging system.

1. Once the system has booted following an installation, the user should login to the system and bring up a terminal window by performing a right-click on the desktop and selecting "Open Terminal". For users who have a single-button pointing device, the menu in question can be raised by holding the "Control" key down while clicking on the desktop.

2. In the terminal window, the following sequence of commands should be performed:

    $ pfexec pkg refresh
    $ pfexec pkg install SUNWipkg@0.5.11-0.86
    $ pfexec pkg install entire@0.5.11-0.86
    $ pfexec pkg install SUNWipkg-gui

After these steps have been performed, packaging operations using the pkg(1) command can be performed. For more information on this command and the IPS technology, refer to the pkg(1) and pkg(5) manual pages and Getting Started With the Image Packaging System.

If you have already installed packages from the repository without upgrading the IPS software, it is possible that your software installation may contain packages that are incompatible with one another. In some cases, the system may not boot completely or will boot with errors. If you would like to restore your system software, see Restoring Your System Software to the Initial Installation for more information.

Development builds of OpenSolaris are also available for use by using the pkg image-update command. For critical information on this feature, see Updating Your System to OpenSolaris Development Builds.

# 6696145 Install Causes Panics on Intel 965 With GM965 Chipset

Description

Installing the OpenSolaris release on certain laptops that are equipped with the Mobile Intel GM965 Express chipset (Intel GMA X3100 Graphics) is known to cause panics.

This issue has been reproduced on the following models that are equipped with the GM965 chipset.

  • Dell XPS M1330

  • Lenovo ThinkPad X61/T61

  • Toshiba Satellite Pro A200

How can I tell if my laptop is running the GM965 chipset?

You can use the Device Driver Utility on the LiveCD. Or, open a terminal window and run the following command:

pfexec /usr/X11/bin/scanpci

This command identifies which Graphics Controller you have on your laptop.

Workaround

Use the VESA driver to install the OpenSolaris 2008.05 release onto your system.

Use the following procedure for the vesa driver:

  1. From the GRUB menu, select the Text Console option.

  2. Login with user name, jack, and the password, jack.

  3. Execute the following commands

    jack@opensolaris:~$ pfexec 
    Xorg -configure
    jack@opensolaris:~$ sed '/Driver.*"intel"/s/"intel"/"vesa"/' xorg.conf.new > xorg.conf
    
    jack@opensolaris:~$ pfexec mv xorg.conf.new /etc/X11/xorg.conf
    jack@opensolaris:~$ svcadm enable gdm
  4. Install the release, per instructions at Installing OpenSolaris 2008.05 .

    After the system reboots, the desktop is displayed using the vesa driver.

  5. If you want to restart the Xserver using the intel driver, type the following:

    % pfexec rm /etc/X11/xorg.conf

    Restart the Xserver by logging out of your desktop session or hitting Ctrl-Alt-Backspace.

# 64 Installer, but not Desktop, Allows Empty Passwords

Description

If, during installation, no root password is entered and no user account is created, you will not be able to log in to the installed system.

Workaround

Select one of the following workarounds:

  • Enter a root password as follows:

    1. During boot, type “e” in the grub menu twice, in order to edit the boot entry.

    2. Add -s to the end of the boot entry, and boot with this line.

    3. Log in as root and use the passwd command to enter a password.

    4. Restart the system. Boot normally, and log in with the new password.

  • Set root so that it does not require a password as follows:

    1. During boot, type “e” in the grub menu twice, in order to edit the boot entry.

    2. Add -s to the end of the boot entry, and boot with this line.

    3. Log in to the system as root.

    4. Edit the /etc/X11/gdm/custom.conf file so that the value of PasswordRequired is false.

# 1045 beadm create Fails When Run as User

Description

When using the beadm command, as a non-root user, to create a new BE, the command fails with the following traceback:

user@opensolaris:~$ beadm create newBE
        Traceback (most recent call last):
          File "/usr/sbin/beadm", line 1296, in ?
            ret = main()
          File "/usr/sbin/beadm", line 557, in main
            return(parseCLI(sys.argv[1:]))
          File "/usr/sbin/beadm", line 532, in parseCLI
            rc = create(optsArgs)
          File "/usr/sbin/beadm", line 146, in create
            if initBELog("create", be) != 0:
          File "/usr/sbin/beadm", line 1244, in initBELog
            os.makedirs(os.path.dirname(be.log), 0644)
          File "/usr/lib/python2.4/os.py", line 156, in makedirs
            makedirs(head, mode)
          File "/usr/lib/python2.4/os.py", line 159, in makedirs
            mkdir(name, mode)
        OSError: [Errno 13] Permission denied: '/var/log/beadm' 
Workaround

Become root or assume root privileges before using the beadm command to create a new boot environment.

# 1109 Locale Information Inaccurate in Installer Summary Panel

Description

The final Summary panel in the installer displays inaccurate language support information.

The installer summary panel states that only the following language support will be installed:

Locale

  • Default Language: English (United States)

  • Language Support: English

The correct information is that the OpenSolaris installer installs all language support that is included in the Live CD.

# 1156 Booting a BE That Was Left Mounted Before the Reboot Fails

Description

If a boot environment (BE) is mounted by using the beadm mount command, and the system is rebooted before that BE gets unmounted, selecting that BE from the GRUB menu causes a boot failure.

In the particular case where a BE has been activated by using the beadm activate command, this BE is the default selection in the GRUB menu. Therefore, leaving this BE mounted before a reboot causes a boot failure.

A typical boot failure message is:

Use is subject to license terms.
        Hostname: opensolaris
        Apr 28 16:05:08 svc.startd[7]: svc:/system/sysevent:default: Method
        "/lib/svc/method/svc-syseventd start" failed with exit status 95.
        Apr 28 16:05:08 svc.startd[7]: system/sysevent:default failed fatally: 
            transition to maintenance (see 'svcs -xv' for details)
        '/usr/sbin/pmconfig: cannot open/create "/etc/.cpr_config", Bad file number
        Requesting System Maintenance Mode
        (See /lib/svc/share/README for more information.)
        Console login service(s) cannot run

        Root password for system maintenance (control-d to bypass): 
Workaround

Reboot the system and select the previously active boot environment. When the system comes up, reset the mount point of the activated boot environment by running the following command as root.

# beadm unmount boot_environment

Then reboot your system.

# 1333 Creating BE and Renaming BE Fails to Handle Subordinate Legacy-Mounted Datasets

Description

When a boot environment (BE) has a legacy mount point for one of its subordinate datasets, creating a new BE based on this BE by using the beadm create command fails. Also, renaming a BE that has a legacy mount point for one of its subordinate datasets fails to update the /etc/vfstab properly.

Workaround

Use one of the following workarounds:

  • In the case where a BE is created, do the following:

    After creating the new BE, use the beadm command to mount the BE. Manually update /etc/vfstab to correct the entry for the subordinate dataset in that BE.

    # beadm create -e origBE newBE
    
            # beadm mount newBE /mnt
            # vi /mnt/etc/vfstab
  • In cases where a BE is renamed, do the following:

    After renaming the BE, use beadm to mount the BE. Manually update the /etc/vfstab to correct the entry for the subordinate dataset in that BE.

    # beadm rename origBE newBE
    
            # beadm mount newBE /mnt
            # vi /mnt/etc/vfstab

# 1334 Compiz Hangs System that Uses Intel Graphics Card i915 or i965GM

Description

Enabling Compiz on a machine that has an Intel graphic card i915 or i965GM will probably hang the system due to driver issues.

How can I tell if my laptop is running the GM965 chipset?

Use the Device Driver Utility on the LiveCD. Or, open a terminal window and run the following command:

pfexec /usr/X11/bin/scanpci

This command identifies which Graphics Controller you have on your laptop.

Workaround

Do not enable the visual effects in the GNOME appearance preference dialog.

# 1472 Creating a new BE From a Mounted BE Results in Incorrect Mount Points for Subordinate Datasets

Description

When you create a new boot environment (BE) from an BE that is currently mounted, some of the subordinate datasets of the new BE have incorrect mount points.

Workaround

Use the zfs list command to look at the mount points for the datasets of the new BE.

For example:

# zfs list -r rpool/ROOT/be2
        NAME                 USED  AVAIL  REFER  MOUNTPOINT
        rpool/ROOT/be2      73.5K  31.0G  2.23G  legacy
        rpool/ROOT/be2/opt      0  31.0G  3.60M  /mnt/opt

Use the zfs set command to reset the subordinate file system at the correct mount point.

# zfs set mountpoint=/opt rpool/ROOT/be2/opt

# 1559 GRUB Entry for Windows Incorrect From USB Stick Install

Description

When the OpenSolaris 2008.95 release is installed from a USB stick, the GRUB entry for Windows does not reference the correct disk device.

Workaround

Edit the /rpool/boot/grub/menu.lst file. Modify the rootnoverify entry in that file.

Typically a USB installed system will have the following entry:

rootnoverify (hd1,0)

This entry should be changed to the following:

rootnoverify (hd0,0)

#---------- ADDED BY BOOTADM - DO NOT EDIT ----------
title Windows
rootnoverify (hd0,0)
chainloader +1
#------------------- END BOOTADM -------------------- 

If a diagnostic partition exists, a similar change is required.

# 1565 Nautilus Process Hangs After Gnome Login

Description

When you log in to the Gnome desktop, Nautilus might hang. In this case, the desktop background is a solid, blue-gray color and right-clicking on the background does not open a menu. The nautilus-debug-log.txt is added in your home directory.

If Nautilus hangs, it can consume 50% of the CPU time.

Workaround

Manually kill the broken Nautilus process as follows:

  1. Launch a Gnome terminal by selecting Applications->System Tool->Terminal on the menu bar.

  2. Kill the oldest nautilus process in the gnome-terminal. For example:

    bash$ ps -ef | grep naut
        test 13425 13374   0 19:59:30 ?           0:00 nautilus --no-default-window
    --sm-client-id default2
        test 13227     1  50 17:39:39 ?          39:36 nautilus --no-default-window
    --sm-client-id default2
    
    bash$ kill 13227
  3. Log out.

# 1771 Install will fail if rpool ZFS pool already exists

Description

If a ZFS pool that is named “rpool” is present on the system , the installer fails when the Install button is clicked on the installation summary screen. A failure screen is displayed immediately.

The last message in the installation log is:

Root pool rpool exists, we can't proceed with the installation.

A root pool that is named “rpool” might be present on the system under any of the following conditions:

  • The user uses zpool create command to create pool that is named “rpool.”

  • The user uses zpool import command to import a pool that is named “rpool.”

  • The installer fails after it creates the pool named “rpool.”

You can use the zpool list rpool command to confirm existence of a ZFS pool that is named “rpool.”

Workaround

Select one of the following workarounds:

  • If you want to preserve data in that pool, reboot the system, and don't import the pool. The pool will remain invisible to the installer. If it was created on the target disk, the pool can't be preserved.

  • If you don't want to preserve data in the pool, destroy the pool before you invoke the installer by using the following command.

    $ pfexec zpool destroy -f rpool

# 1685 Renaming a Busy Boot Environment can Fail

Description

Renaming a boot environment (BE) might fail if the BE has a dependent clone that is currently mounted or otherwise busy.

A typical example of this situation is when the live BE has been successfully upgraded. If you then try to rename the auto-generated and upgraded BE, the renaming fails. The renaming fails because the upgraded BE has been activated and the live BE is now its clone.

Workaround

Use the following procedure:

  1. Activate the live boot environment.

  2. Rename the upgraded boot environment.

  3. Activate the upgraded boot environment.

For example, if the live boot environment name is “opensolaris”, and the auto-generated BE name of the upgraded boot environment is “opensolaris-1,” use the following commands:

# beadm activate opensolaris
# beadm rename opensolaris-1 new_name
# beadm activate new_name

# 6676221 setxkbmap Can't Locate Rules file for Xorg

Description

By default, setxkbmap has not defined the correct path to the rules file for Xorg, /usr/X11/lib/X11/xkb/rules/xorg. When you try to set a Keyboard Layout by using setxkbmap, the following error message appears:

/bash-3.2$ /usr/X11/bin/setxkbmap br
Couldn't find rules file (xorg) 
Workaround

To correctly set the path to the rules file, type:

-bash-3.2$ setxkbmap -rules /usr/X11/lib/X11/xkb/rules/xorg/ 

Once this path is set, the system remembers the path until you restart your Xserver.

# 6687328 vim Command Cannot Display Multibyte Characters

Description

The vim command can not display multibyte characters. This bug affects other commands which symbolically link to /usr/bin/vim such as vi, view, and ex commands.

Workaround

Add the $HOME/.vimrc file that contains the following line:

set encoding=utf-8

Or, use the /usr/has/bin/vi command instead.

# 6690194 fdisk(1M) Doesn't Clear VTOC When Partition Starts After, or Exceeds, 4 Gbytes

Description

The symptom of this bug in OpenSolaris 2008.05 is that the installation will fail immediately after starting, and one of the last messages in the installation log will be "No space left on device: '/a/./.livecd'".

When a new Solaris2 partition is created, or an existing Solaris2 partition is resized or moved, the following problems can occur, if the Solaris2 partition starts after, or exceeds, 4 Gbytes.

The fdisk(1M) command clears the primary VTOC, as well as clearing the backup VTOCs. So, the default VTOC is chosen by the kernel. The appropriate operations which handle clearing the VTOC, lseek(2) and write(2), are 32-bit limited. This situation causes the following issues.

  • A Solaris2 partition that starts after the 4 Gbyte limit cannot be re-sized. If the starting sector is unchanged, deleting the partition and recreating it doesn't solve the problem.

  • When the fdisk(1M) command clears the VTOC for a Solaris2 partition that crosses the 4 Gbyte boundary, the utility actually clears a couple of sectors within the 4 Gbytes, due to the 32-bit overflow. If valid data is defined in those sectors, the data is corrupted.

    For example, if there is a NTFS partition occupying the first 4 Gbytes, during creation of Solaris2 partition, the fdisk(1M) command actually clears about 6 random sectors somewhere in the NTFS or boot block, depending on starting location of the Solaris2 partition to be created.

Workaround

A Solaris2 partition must always fit into first 4 Gbytes, otherwise data corruption might occur.

# 6690824 fdisk(1M) Corrupts Solaris VTOC, if Linux Swap Primary Partition is Defined

Description

The symptom of this bug in OpenSolaris 2008.05 is that the installation will succeed, but the system will fail to boot with a "Bad PBR sig" or just a "grub>" prompt.

If you install on a Solaris2 partition which precedes a Linux swap partition, the following problem occurs.

After the installation, while marking the Solaris2 partition as active, the fdisk(1M) utility corrupts Solaris2 VTOC, thus making the installed Solaris partition unusable.

Workaround

Use one of the following workarounds.

  • Avoid defining a Linux swap as a primary partition on the same disk where a Solaris2 primary partition exists.

  • Before installation, change the Linux swap partition type to another type, excluding 0xbf or 0x82. After the installation finishes, change the Linux swap partition type back to 0x82.

  • Create a Solaris2 partition outside the installer and mark it as active. Then install into this preexisting partition. After the installation finishes, calling fdisk(1M) to activate the partition will then be a null operation. Corruption will not occur.

# 6694783 Input Method Duplicate Causes IDE Crash

Description

On rare occasions, when you log in to the desktop, you may encounter the following issue.

The Input Method(IM) status icon is displayed at the bottom of main window. When you open a new child window, such as when you select File->Open, a new IM status icon displays in the main window. When you close this child window, the new IM status icon remains. If you click on an IM status icon at this point, IDE crashes.

Workaround

  1. Find out if the iiim daemon is running by typing this command.

    # ps -ef | grep iiim
  2. If you do not find a process that is related to iiimx or iiimd, restart the iiim daemon by typing this command.

    # iiimx -iiimd &
  3. Try to use the Internet/Intranet Input Method instead of the X Input Method. You can select this option by right-clicking on the Gnome Applications icon.

# 2900 Firefox crashes on NV 95

Description

Firefox dumps core in NV build 95 due to Flash Player bug. Flash plugin v9.0 r124 has a bug that causes Firefox to crash. This issue has been fixed in the flash plugin v 9.0 r125.

Note - This affects only users of Flash Player v9.0 r124 and below.

Workaround
  1. Start firefox.

  2. From the Menu bar, click on Tools -> Add-ons -> Plugins.

  3. Right click on Shockwave Flash and choose the Disable option to disable the plugin.

  4. Manually install the updated version from the following url: http://www.adobe.com/products/flashplayer/productinfo/instructions/#section-4