Internal Ethernet Adapter Duplex Settings Cause Hang on E105 LED
Environment
OS Level: AIX Version 4.3.2
Type/Model: 9076
Problem
When attempting to run a network boot, the procedure stops and the LED first
displays E105. It then displays 40B00100. The screen shows
the following information where it appears to be waiting for input:
chrp DEFAULT CATCH!, code FFF00700 at %SRR0:00C522CC %SRR1:00083000
ok
0>
In general, the LED E105 error is caused by an invalid bootlist.
Various steps, such as those listed in the "Solution summary"
section of this document, should be taken first to rule out the more likely causes
of this error. In this particular instance, however, the error was caused by an
internal Ethernet adapter duplex setting.
The en0 adapter was incorrectly set to full duplex on node 25.
Solution
The LED code E105 generally indicates a network problem. It may be caused by a
malfunctioning Ethernet card or by an Ethernet adapter being set to an incorrect
duplex setting. If after other, potential causes of this problem have been ruled
out, such as those listed in the "Solution summary" section of
this document, and you suspect the adapter setting is the cause, complete the
following steps.
- Verify the duplex setting on your nodes with the lsattr command. For
example:
node1>lsattr -Dl ent0
busio Bus I/O address False
busintr Bus interrupt level False
intr_priority 3 Interrupt priority False
tx_que_size 64 TRANSMIT queue size True
rx_que_size 32 RECEIVE queue size True
full_duplex no Full duplex True
use_alt_addr no Enable ALTERNATE ETHERNET address True
alt_addr 0x000000000000 ALTERNATE ETHERNET address True
node1>
- Verify that the full_duplex setting is correct on all nodes for
your particular network environment.
- If all adapters are correctly set, perform node diagnostics to determine
if there is a malfunctioning adapter card in your system. If so, contact IBM
support to have it replaced.
Listed is a history of the steps taken to determine the cause of the problem.
Instructions on how to accomplish these steps are beyond the purpose of this
document. If necessary, contact your AIX support center for assistance with
these steps.
- The NVRAM (non-volatile RAM) was checked to see if it needed to be drained.
- Checked to see if there was a compressed file system.
- An attempt was made to network boot the system for install or into Maintenance
mode but was not successful.
- The SPOT was rebuilt many times.
- The latest fixes for PSSP 3.1 (U467374) and AIX Version 4.3.2 were installed.
- It was possible to run a manual node conditioning, run the
ping test, and boot from disk without problems. An attempt to do a network boot
failed displaying the error message described in the "Problem" section.
- An attempt was made to boot node 29 into Maintenance mode using AIX Version
4.3.2 SPOT. It hung with the preceding errors and then with LED 40b and E075.
- The SPOT and the Network Install Manager (NIM) were put into Debug mode. After
several manual node conditionings, the system booted into Service mode, went passed
LED E105 and eventually was able to install. The result of the tasks in this step is
inconsistent across nodes. Depending on the node, this procedure may work or may
result in the same errors. For this reason, this problem is often misdiagnosed as
a hardware-related problem.
- To attempt to isolate the problem, the en0 adapter was unplugged from
node 49 and 53 and terminated using two different terminators. The same
problem occurred.
- Node 53 was reinstalled; however, the node could not be put back into
Maintenance mode. It still hung on LED E105 then 4B246110.
- The problem was due to an en0 adapter set to full duplex on node25.
[ Doc Ref: 94892114414900 Publish Date: Feb. 16, 2000 4FAX Ref: none ]