Inom
Hitta mer dokumentation
Supportresurser som ingår
| Ladda ner denna bok i PDF
Troubleshooting
E
- This appendix describes problems you may encounter when installing Solaris software. This appendix also suggests possible solutions for some of the problems.
- The following table shows common error messages and the page number where you can find causes and possible solutions to the related errors.
-
Specific Installation Errors
-
Error: Unknown client "host_name"
|
-
| Reason Error Occurred | How to Fix the Problem |
| The host_name argument in the add_install_client command must be a host in the name service. | Add the host host_name to the NIS or NIS+ name service. (Either update the NIS maps or populate the NIS+ tables.) Try executing the add_install_client command again. |
-
ip: joining multicasts failed on tr0 - will use link layer broadcasts for
multicast
|
-
| Reason Error Occurred | How to Fix the Problem |
| You will see this error message when you boot a system with a token ring card. Ethernet multicast and token ring multicast do not work the same way. The driver returns this error message to indicate that an invalid multicast address was given. | Ignore this error message. If multicast doesn't work, IP uses layer broadcasts instead. |
-
le0: No carrier - transceiver cable problem
|
-
| Reason Error Occurred | How to Fix the Problem |
This system is not connected to the
network. | If this is a non-networked system, ignore this message. If this is a networked
system, make sure the Ethernet cabling is attached securely. |
-
The file just loaded does not appear to be executable
|
-
| Reason Error Occurred | How to Fix the Problem |
| This message cannot find the proper media for booting. | Verify that you are using the correct boot command for your system. If installing from a CD-ROM drive, make sure the Solaris CD is in the CD-ROM drive and that the CD-ROM drive is mounted. . |
-
Still trying to find a RPL server...
|
-
| Reason Error Occurred | How to Fix the Problem |
| The system is trying to boot over the network, but the server is not set up to boot this system. | On the install server, execute add_install_client for the system to be installed. The add_install_client command sets up an /rplboot directory, which contains the necessary network boot program. |
-
Requesting Internet address for Ethernet_Address
|
-
| Reason Error Occurred | How to Fix the Problem |
| The client is trying to boot over the network, but it cannot find a system that knows about it. | Verify the system's host name is in the NIS or NIS+ name service. If the system's host name is in the NIS or NIS+ name service, and the system continues to print this error message, try rebooting. |
-
RPC: Timed out
No bootparams (whoami) server responding; still trying...
|
-
| Reason Error Occurred | How to Fix the Problem |
| The client is trying to boot over the network, but it cannot find a system with an entry in the /etc/bootparams file on the install server. | Use add_install_client on the install server. This will add the proper entry in the /etc/bootparams file, enabling the client to boot over the network. |
General Installation Problems
-
| Problem | How to Fix the Problem |
| The system boots over the network, but from a system other than the specified install server. | On the name server, update the /etc/bootparams entry for the system being installed. The entry should conform to the following syntax:
install_system root=boot_server:path install=install_server:path
|
-
| Problem | How to Fix the Problem |
| After setting up an install server and configuring the system to install over the network, the system still does not boot. | Be sure the tftpd daemon is running on the install server. Type the following command and press Return:
ps -ef | grep rpld
If this command does not return a line indicating the rpld daemon is running, execute the following command:
/usr/sbin/rpld
After making this change, try booting the system again.
|
-
| Problem | How to Fix the Problem |
| OpenWindows is not available to diskless and dataless clients. | The /usr/openwin may be on a separate, mounted file system. This directory is not automatically shared by admintool. The /usr/openwin file system must be manually added to the server's /etc/dfs/dfstab file and the file system must be shared. The entry looks like this:
share -F nfs /usr/openwin
Also, an explicit mount of the file system must be made in the client's /etc/vfstab file using the following:
server_name:/usr/openwin - /usr/openwin nfs - yes -
|
-
| Problem | How To Fix the Problem |
| Several processes that normally run when booting the system are not run when the system boots from the CD. (This is done to enable the system to boot and the Solaris installation program to run with only 16 Mbytes of memory.) | To boot from the CD to do disaster recovery (that is, restoring the root file system), you need to add functionality normally provided when booting. To provide this functionality, a script is provided, /sbin/setup_cd, which can be run to perform the additional setup that might be needed to do disaster recovery. setup_cd does the following tasks: - Finishes setting up the device file system for tapes and ports - Enables routing - Enables multicast - Invokes the statd(1M) daemon - Invokes the lockd(1M) daemon - Starts the automounter
Note that if setup_cd is run on a 16 Mbyte system, it may not be possible to run the Solaris installation program or other programs that need more memory. If a disk with a swap partition is available, the system administrator could avoid this problem by using the swap(1M) command to add swap space to the system before invoking setup_cd.
When booting from the Solaris CD during disaster recovery, it is possible to boot the system in single user mode from the CD-ROM. For example, at the menu option to install using Interactive or custom JumpStart, type:
b -s
|
-
| Problem | How to Fix the Problem |
| An installation using the upgrade option fails for reasons beyond your control, such as a power failure or a network connection failure. The system may be in an unbootable state. | 1. Reboot the system from the Solaris CD or from the network. 2. Choose the upgrade option for installation. The Solaris installation program will determine if the system has been partially upgraded and will continue the upgrade. |
-
| Problem | How to Fix the Problem |
| IDE disk drives do not automatically map out bad blocks like other drives supported by Solaris software. Before installing Solaris on an IDE disk, you may want to perform a surface analysis on the disk. | To perform surface analysis on an IDE disk, follow this procedure:
1) Start an interactive installation, as described in Chapter 6, "Booting and Installing Solaris: Interactive." The Solaris installation program will start either a graphical user interface (GUI) or a character user interface (CUI), depending on whether you have a graphics or non-graphics monitor. 2) When either the GUI or CUI program starts, enter information and select the Continue option on the first few screens. 3) When you see the Installing Solaris - Initial screen, select the Exit option and exit the installation. 4) If you are using the GUI installation program, open a command tool window for the remaining steps in this procedure. If you are using the CUI installation program, use the system shell for the remaining steps in this procedure. 5) Start the format program. Type format. 7) Specify the IDE disk drive on which you want to perform a surface analysis. Note: IDE drives do not include a target number. The IDE drive naming convention is cxdy, where cx is the controller number and dy is the device number. 6) At the format> prompt, type fdisk. Use the fdisk program to create a Solaris partition on the disk. (If a Solaris fdisk partition already exists, leave it alone.) 7) At the format> prompt, type analyze. 8) At the analyze> prompt, type config. This will show you the current settings for a surface analysis. If you want to change any settings, type setup. 9) At the analyze> prompt, type read, write, or compare for the type of surface analysis to be performed. If format finds bad blocks, it will remap them. 10) At the analyze> prompt, type quit. 11) You may want to specify blocks to remap. If so, at the format> prompt, type repair. 12) Type quit to quit the format program. 13) Type installtool to resume the GUI installation or suninstall to resume the CUI installation.
|
-
| Problem | How to Fix the Problem |
| The Solaris root partition must reside within the first 1024 cylinders of the disk. If it is not, the installation fails after booting. | If the first fdisk partition is primary DOS (PRI DOS), use the fdisk program to delete space from it and try booting Solaris again. If the first fdisk partition is extended DOS (EXT DOS) or another operating system, use the fdisk program to delete it and try booting Solaris again. |
|
|