Re: MA111 problem at boot on Fedora FC3

Hi Peter,

--- Peter Ison <[email protected]> wrote:
> I uncommented the debug lines in hotplug.functions, usb.agent and
> wlan.agent 
> and rebooted. I attach the log for the relevant hotplug, usb.agent, 
> wlan.agent, Prism2 events. I sorted the file because I noticed that
> the 
> kernel events were out of sequence in /var/log/messages.

Do you mean that the order in the log didn't correspond to the date
order? I dare say there are various possible reasons for this, with
multiple processes writing to the log, but one factor that I've noticed
in my logs is that the system clock gets set during boot. This happens
in /etc/rc.sysinit, and in my case also during ntpd startup. I actually
get a jump forward by an hour (give or take a minute) and a subsequent
jump back, presumably due to a UTC versus localtime or DST issue that I
haven't got around to sorting out. Even without the hour jump, system
clock adjustments during boot may mean that the time field of log
entries is not a reliable guide to their actual sequence.

So, AFAICS, sorting the log may produce a misleading result. But I'm
not an expert on these issues and will be happy to defer to someone who
is!

> All activity seems to be initiated through hotplug (or coldplug as
> part of 
> the boot process). However, in reply to your question, there are no 
> significant events after the linkstatus DISCONNECTED/CONNECTED
> sequence.

Interesting - so in a sense, this fits with Karl's findings and mine,
that it's usb probing activity after the linkstatus=CONNECTED that
causes the lockup. What's odd is that the sequence of events on your
FC3 system is so different from mine. Ah well. Maybe the MA111 just
takes longer to start up?

> Hope this helps.

Yes, thanks for taking the trouble to post your findings.

Cheers,

Dave


	
	
		
___________________________________________________________ 
Yahoo! Messenger - NEW crystal clear PC to PC calling worldwide with voicemail http://uk.messenger.yahoo.com