Bela Lubkin wrote:
>>> Why without timestamp? Sometimes time information is
>>> significant in a crash.
>
>> The main reason is that SEL size is limited and I think the timestamp
>> area (4 bytes) of each entry is quite large.
>> I'm not sure whether this timestamp is so important.
>> Could you please explain a bit more about this?
>
> Debugging a panic, it is helpful to know what was going on right
> before failure. There might have been a SEL entry right before
> ("RAM failure"?). There's a big difference between these two logs:
>
> 17:45:10 RAM failure bank blah blah blah
> 23:30:31 oops...
>
> and
>
> 17:45:10 RAM failure bank blah blah blah
> 17:45:11 oops...
Thanks for the above explanation. I see, it makes sense.
I didn't realize the possibility, there is an entry causes panic.
>
> How about use the time stamp in the _first_ record, then leave it
> out. Put it back in if more than 10 seconds have elapsed since
> the last stamp. Then for prolific messages you don't consume
> much, but you still have reasonably accurate times for for
> sporadic messages.
Sounds good. Will consider and update.
Thanks,
Hiroshi
------------------------------------------------------------------------------
SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
The future of the web can't happen without you. Join us at MIX09 to help
pave the way to the Next Web now. Learn more and register at
http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/
_______________________________________________
Openipmi-developer mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/openipmi-developer
