On 06/09/09 14:45, Paul B. Henson wrote:
On Mon, 8 Jun 2009, David Zhang wrote:
With fmdump -eV, we can find what ereport cause this issue, and do
further analyze.
Here is another example of diagnostic output I got from someone running
NexentaStor (based on b104 ON bits) who also has an X25-E reported as
failing self-test, if it provides any other clues to what's going on. His
x4500 doesn't have any failure lights lit though, not sure why.
self-test-failure = (embedded nvlist)
nvlist version: 0
result-code = 0xb
timestamp = 0xb400
segment = 0x0
address = 0xb400b400b4
(end self-test-failure)
As with the last one, this data doesn't make any sense. Besides the
curious repetitive 0xb4, the result-code of '0xb' is completely invalid.
I still have to take a look at the core you gave me, but I do notice
the code isn't quite structured as I would have expected. In
particular, if there is a failed self test followed by a successful one,
then we don't log the error. So I've identified at least two problems
with the current code, but neither one explains exactly what is going on.
- Eric
--
Eric Schrock, Fishworks http://blogs.sun.com/eschrock
_______________________________________________
fm-discuss mailing list
fm-discuss@opensolaris.org
