A P P E N D I X  F

Troubleshooting

This chapter provides troubleshooting suggestions for the following symptoms:

Symptom: A server is not responding or a server may be down.

Make sure the services are running at the server.

1. (UNIX OS) Run the following command:


# ps -e | grep ss

The names ssmon and ssserver should both be present in the output. If they are not, proceed to Step 2. If they are, proceed to Step 4.

(Windows 2000) Choose Start right arrow Programs right arrow Administrative Tools right arrow Computer Management. Click Services & Applications right arrow Services and verify that SUNWscsd Monitor, SUNWscsd Server, and SUNWscsd Startup services are started. If they are not, proceed to Step 2. If they are, proceed to Step 4.

(Windows 2003) Choose Start right arrow Administrative Tools right arrow Computer Management right arrow Services & Applications right arrow Services and verify that SUNWscsd Monitor, SUNWscsd Server, and SUNWscsd Startup services are started. If they are not, proceed to Step 2. If they are, proceed to Step 4.

2. (UNIX OS) Stop and start the daemons as explained in the Sun StorEdge 3000 Family Software Installation Guide.

(Microsoft Windows) Stop and start the services by right-clicking the service you want to start or stop.

3. If the daemons/services do not stop/start properly, do a reconfiguration reboot.

4. Make sure that the TCP/IP protocol software is loaded, and that the network card and the network connection are working.

To test the TCP/IP connectivity at the command-line, type:


# ping {IP address of the server or the server name}

If you do not get a reply, there is a problem in the TCP/IP protocol services. Contact your MIS department for assistance.

5. Make sure that the correct server name and password are specified.

If the name or password is not correct, make the correction by editing the entries. See the “Administrative (User) Security Levels and Guidelines” section for your OS in the Sun StorEdge 3000 Family Software Installation Guide.

The password has to be the one established through the procedure for setting up users.

Symptom: No logical drives are listed when trying to add a logical volume from existing logical drives.

If you do not see any logical drives listed under Select a Logical Drive, the logical drives have not been unmapped and therefore are unavailable to select. You must unmap the logical drives first.

Symptom: An IP address of a server in a DHCP environment has changed.

In a DHCP environment, there is a remote possibility of a server’s IP address changing if the server has been offline for longer than three days.

In these cases, the console must be notified of the new IP address of the server because the console communicates with the server through its IP address.

To remedy this situation, the server’s new IP address needs to be entered into the Edit Server window of the console software.

1. Choose File right arrow Server List Setup.

2. Move the server name from the Managed Servers list to the Available Servers list.

The Edit Server window is displayed. This window shows the server name and the IP address as it was configured.

3. Select the name of the server in the Available Servers list, and click Edit.

The Edit Server window is displayed. This window shows the server name and the IP address as it was configured.

4. If the network address has changed, click Get IP Addr by Name.

The program searches for and displays the correct IP address if you provided the name of the server as it is recorded by the service used by your network. Otherwise, you must type the IP address.

If the name used for the server is not the same as the server’s network name, or if the naming service is not yet updated, you have to type the IP address manually.

5. Once the server IP address is entered, click OK to specify the changed address and go back to the Server List Setup window.

6. Move the server name back to the Managed Servers list.

7. Click OK to leave the Server List Setup window.

Symptom: Traps are not received from a server.

To troubleshoot why a trap from a server is not received on an SNMP management workstation, such as HP OpenView, use the following steps:

1. Make sure that the SNMP service has been started on the server.

Verify that sstrapd is running by running the following command


# ps -e | grep ss

The output includes the name sstrapd; if it does not, start or stop the agent as explained in the Sun StorEdge 3000 Family Software Installation Guide.

2. Check that the proper target address of the enterprise management console and the community string is set up on the server properly.

3. Make sure that the MIB is compiled under the SNMP management console.

On a Solaris, Linux, and HP-UX OS, the RST_OID.MIB file is located in /opt/SUNWsscs/ssagent. On an IBM AIX OS, this file is located in /usr/SUNWsscs/ssagent. On Microsoft Windows, this file is located in
\Program Files\Sun\ssagent. Refer to your SNMP management console documentation (such as HP OpenView) for information about how to compile the MIB.

Symptom: HP OpenView won’t install or the trap daemon won’t start.

If you installed HP OpenView on the same server that you set up to send SNMP traps, there are insufficient system resources for both the trap daemon and HP OpenView to run. Reconfigure system resources as explained in the following steps.

1. Add the following lines to the end of /etc/system:


set shmsys:shminfo_shmmax=0x2000000
set shmsys:shminfo_shmmin=1
set shmsys:shminfo_shmmni=256
set shmsys:shminfo_shmseg=256
 
set semsys:seminfo_semmap=256
set semsys:seminfo_semmni=512
set semsys:seminfo_semmns=512
set semsys:seminfo_semmsl=32



Note - If, due to the requirements of another application installed on your system, /etc/system already contains statements assigning values to any of these kernel parameters, you need to merge the parameter assignments shown in Step 1 into the file so that each parameter is only assigned once. If a previous parameter value is different than one specified in Step 1, specify the larger value.


2. Reboot the server.

Symptom: The console does not show changes when hardware is added or replaced.

If you add new equipment or if you replace a failed device, such as a disk drive, tape drive, power supply, or fan, the console does not always show updated information. At times, the tree view and the other views associated with it do not reflect the actual status of a device.

If a new device is added and you want this device to be recognized before the next periodic scan, click Rescan in the Server View window. The server immediately rescans for the inventory, and the console updates its server inventory and present the result of any changes in the tree view. To ensure a timely inventory, do a manual scan.

You might have to click Rescan more than once. When the server is in the process of a rescan and the console sends a request for the inventory, the server sends the console the server inventory at its last update, because it has not yet finished the current scan.

Symptom: Logical drive is not present on a Solaris host.

If the logical drive is not being displayed, be sure the logical drive is labeled and not excluded by Sun StorEdge Configuration Service (if installed).

Symptom: Environmental alarms are not being reported.

Storage enclosures with SCSI-based enclosure monitoring capabilities (such as SAF-TE cards) send environmental alarms to the console. The alarm state can be caused by a failure of a drive, fan, power supply, or battery, or by an abnormal temperature.

If an environmental alarm occurs, you must click Reset in the View Enclosure window to clear the alarm. You need to have configuration security privileges to click Reset in the software.

Symptom: Cannot silence the alarm.

Storage enclosures with SCSI-based enclosure monitoring capabilities (such as SAF-TE cards) send environmental alarms to the console. The alarm state can be caused by a failure of a drive, fan, power supply, or battery, or by an abnormal temperature. To silence an environmental alarm, you must push the Reset button on the right ear of the array.

An alarm can also be caused by a controller event, such as when a logical drive fails, during a rebuild, or when adding a drive. Refer to the “Event Messages” appendix in the Sun StorEdge 3000 Family RAID Firmware User’s Guide for more information about controller events. To silence the controller, see To Mute the Controller Beeper.



Note - Pushing the Reset button has no effect on controller event alarms and muting the beeper has no effect on failed component alarms.


Symptom: The console appears to run slowly.

Sun StorEdge Configuration Service can monitor and manage up to 32 arrays at one time. However, console response time can decrease as the number of arrays being managed increases.

When the amount of memory used approaches the total available virtual memory (physical memory plus the page file size), the excessive paging likely causes problems, resulting in poor performance of all applications on that workstation.

Increase the physical memory and page file size to increase the overall virtual memory. To change the page file size, select Control Panel right arrow System, and then choose the Performance tab in the System Properties window that is displayed.

Symptom: Sun StorEdge Diagnostic Reporter stops working.

(UNIX OS) There are three conditions under which Sun StorEdge Diagnostic Reporter stops working and does not report its condition. The workaround is to stop and restart it.

For a Solaris host and Linux host, stop and restart Sun StorEdge Diagnostic Reporter by typing:


# /etc/init.d/ssdgrptd stop 
# /etc/init.d ssdgrptd start

For an HP-UX host, stop and restart Sun StorEdge Diagnostic Reporter by typing:


# /sbin/init.d/ssdgrptd stop 
# /sbin/init.d ssdgrptd start

For an IBM AIX host, stop and restart Sun StorEdge Diagnostic Reporter by typing:


# /usr/sbin/ssdgrptd start
# /usr/sbin/ssdgrptd stop

Symptom: (UNIX OS) The online help is not displayed.

Make sure the absolute path name of the web browser you want to use to display online help has been specified.

1. For the Solaris, Linux, and HP-UX OS, change to /opt/SUNWsscs/sscsconsole. For the IBM AIX OS, change to /usr/SUNWsscs/sscsconsole.

2. Type:


./config_sscon

3. Enter the absolute path name of the web browser.