HP Support Information Digests

===============================================================================
o IT Resource Center World Wide Web Service
   ---------------------------------------------------

   If you subscribed through the IT Resource Center and would
   like to be REMOVED from this mailing list, access the
   IT Resource Center on the World Wide Web at:

     http://itrc.hp.com/

   Login using your IT Resource Center User ID and Password.
   Then select Support Information Digests (located under
   Maintenance and Support). You may then unsubscribe from the
   appropriate digest.
===============================================================================


Digest Name: monthly HP-UX technical tips digest
    Created: Mon Jan 15 3:10:05 PST 2001

Table of Contents:

Document ID Title
--------------- -----------
3100372189 Sys Interrupts: dmesg buffer contains "Unexpected interru
3100501431 LVM: how many alternate links does LVM (Logical Volume Manager
3100472165 OB II:
3100481788 Sys Adm: A4576A color monitor does not respond to Ctrl-B

The documents are listed below.
-------------------------------------------------------------------------------


Document ID: 3100372189
Date Loaded: 20010104
      Title: Sys Interrupts: dmesg buffer contains "Unexpected interrupt on EIRR bit 3&q

PROBLEM

The dmesg buffer contains many repeats of the following message:

     Unexpected interrupt on EIRR bit 3

The kernel is constantly logging this message as is evidenced by entering
the following dmesg(1M) command as root:

     # dmesg -

Also, the server crashed two times, each time with the aforementioned
messages in the buffer. No HPMC (High Priority Machine Check) chassis
codes were recorded for the event. STM (Support Tool Manager) diagnostic
logs found problems with all 4 cards in the GSC backplane. The Event
Monitoring Service was reporting the following errors:

  Sep 18 18:16:32 [nodename] EMS [1831]: ------ EMS Event Notification ------
  Value: "CRITICAL (5)" for Resource:
  "/storage/events/disk_arrays/High_Availability/8_0.1" (Threshold: >= "3")
  Execute the following command to obtain event details:
  /opt/resmon/bin/resdata -R 119996422 -r
  /storage/events/disk_arrays/High_Availability/8_0.1 -n 119996417 -a

How can these problems be corrected?

CONFIGURATION

Operating System - HP-UX
Version - 10.20
Hardware System - HP9000
Series - K460

RESOLUTION

Any of the following items could cause the conditions observed above.
Check each of these items and perform the recommended actions:

  1. The device may have been powered off and may be in the process
      of being powered on.

  2. One or both of the terminators on the SCSI bus may be missing.
      Install the terminators in their proper locations at the ends
      of the SCSI bus.

  3. The SCSI cable may have become detached from the device. Re-
      attach the cable.

  4. The SCSI cable may have failed. Replace it.

  5. The device may be in a state where it could not process this, or
      any request. Cycle power to the device.

  6. There could be more than one device with the same address on the
      SCSI bus. Make all the addresses on the SCSI bus unique.

  7. The total length of all cable segments on the SCSI bus exceeds 25
      meters. Replace one or more cable segments until the total length
      is less than this value.

  8. If all of the above fail to correct the problem, then the device
      has experienced a hardware failure. Repair or replace it as
      necessary.

  9. If messages corresponding to this condition appear in the log for
      more than one device on the SCSI bus, then the device adapter may
      be in a state from which it cannot extract itself. Perform a
      system shutdown, cycle power to the computer and wait for it to
      reboot.

 10. If, after reboot, messages corresponding to this condition continue
      to appear in the log for this SCSI bus, replace the device adapter.

In this case, a SCSI cable was found to have bent pins and was replaced.

<!-- ALT KEYWORDS -->
-----End of Document ID: 3100372189------------------------------------------


Document ID: 3100501431
Date Loaded: 20010103
      Title: LVM: how many alternate links does LVM (Logical Volume Manager) allow

PROBLEM

How many alternate links does LVM (Logical Volume Manager) allow?

CONFIGURATION

Operating System - HP-UX
Version - 11.0
Hardware System - HP9000
Series - V-Class
Subsystem - LVM (Logical Volume Manager)

RESOLUTION

LVM allows up to eight pvlinks to any one device. However, LVM only
allows four paths to boot devices per Volume Group.

The four paths include mirrors, primary, and alternate paths. For
example, root, the root mirror, and their pvlinks equal four paths.

<!-- ALT KEYWORDS -->
-----End of Document ID: 3100501431------------------------------------------


Document ID: 3100472165
Date Loaded: 20010103
      Title: OB II:

PROBLEM

The "/opt/omni/bin/omnimm -check_protection <LABEL>" command and the GUI (Graphical
User Interface) Media Management window report different media protection.

Why do the omnimm command and the GUI Media Management window report
different information?

CONFIGURATION

Operating System - HP-UX
Version - 11.0
Hardware System - HP9000
Series - L2000
Subsystem - OB II (OmniBack II)

RESOLUTION

omnimm and the GUI Media Management window report different information
if LABEL is duplicated in another Media Pool. omnimm and the Media
Management window display information for two different tapes.

OmniBack permits the designation of duplicate LABELs (initially entered
in the "Description" field when initializing Media). Furthermore,
OmniBack maintains unique, internal "Medium IDs" that look similar
to the following string:

     0f11b85c:3a1f05aa:0671:0001

Using the "Medium ID" in the omnimm command will display the correct
protection information.

For example:

/opt/omni/bin/omnimm -check_protection 0f11b85c:3a1f05aa:0671:0001
would display the correct protection information, such as:

Medium 0f11b85c:3a1f05aa:0671:0001 protected until Mon Dec 4 21:00:40 2000

<!-- ALT KEYWORDS -->
-----End of Document ID: 3100472165------------------------------------------


Document ID: 3100481788
Date Loaded: 20001219
      Title: Sys Adm: A4576A color monitor does not respond to Ctrl-B

PROBLEM

The K580 server hung, and would not respond to Ctrl-B. The users had
to power cycle the system.

Why doesn't this system respond to the Ctrl-B key sequence?

CONFIGURATION

Operating System - HP-UX
Version - 11.0
Hardware System - HP9000
Series - K580
Subsystem - System Administration

RESOLUTION

The system was configured with a A4576A color monitor.

The A4575A/A4576A color (graphics) monitors use a different port than
the standard 700/96 console monitors. The A4575A/A4576A monitors do
not respond to Ctrl-B. The monitor must be connected to the standard
console port.

To resolve the problem, connect a 700/96 monitor to the console port.
The K580 will respond to Ctrl-B once the 700/96 monitor is connected
to the system.

<!-- ALT KEYWORDS -->
-----End of Document ID: 3100481788------------------------------------------

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]