Panic/Action Error Messages - Extreme Networks BlackDiamond 6804 Troubleshooting Manual

Panic/Action Error Messages

In addition to fabric checksum error messages, there are also error messages that indicate a panic
condition or a condition to which the system health checking subsystem is responding by initiating the
configured action. Actions taken can include: error logging, auto scan automatic single-bit memory
mapping by the packet memory subsystem, card shutdown, or switch shutdown.
NOTE
The primary use of panic messages is to assist in isolating problems to a given switch module. This
information is also useful when communicating with Extreme Networks Technical Assistance Center
(TAC) personnel about a system problem.
For more information on how the system health checker can be configured to react to these panic
conditions, refer to "On Demand Packet Memory Scan" on page 59.
These panic/action messages take the format:
mm/dd/yyyy hh:mm:ss <level:from> sys-health-check [ACTION] (check) (error)
on location
and indicate that the system health check has taken the configured action upon detecting a systematic
error, where:
check
error
The PBUS checksum type might indicate a problem with the packet memory, but it might also point to
other problem types, indicating failures with other hardware components detected during additional
checking performed under the umbrella of the Extreme diagnostics suite. For example, any other
component in the path between the ingress and egress points could malfunction, resulting in a
corrupted checksum. You should be aware that PBUS checksum messages and conditions can be caused
Advanced System Diagnostics and Troubleshooting Guide
The type of problem detected, based on the health check diagnostic component that
triggered the action, from among the following:
• CPU diag packet—Action was taken because excessive CPU health check packets were
missed. Examine the CPU health check diagnostics. Errors on more than one slot
probably indicate MSM management bus transceiver issues. Errors on single slots
probably indicate that the specified slot is in error.
• Hardware failure—Indicates a card diagnostic failure during:
- System boot
- Card initialization
- Manual diagnostics
- Transceiver test
• PBUS checksum—Fabric checksum error detected for:
- Data traffic, [INT] or [EXT]
- CPU-bound traffic [CPU}
- EDP packets [EDP]
• Backplane link—Indicates that health check packets were lost on one or more backplane
links connecting an MSM module to an I/O module. Either module might be in error;
check the transceiver diagnostics.
The error condition that summarizes the failure. One of many values that points to a specific
hardware component or software test (for further problem isolation).
Panic/Action Error Messages
35