SATA - System Freezes

Henry Ritzlmayr fedora-list at rc0.at
Mon Jun 23 07:05:59 UTC 2008


Am Freitag, den 20.06.2008, 09:37 -0600 schrieb Robin Laing:
> Henry Ritzlmayr wrote:
> > Am Donnerstag, den 19.06.2008, 09:52 -0600 schrieb Robin Laing:
> >> Henry Ritzlmayr wrote:
> >>> Am Dienstag, den 17.06.2008, 13:25 -0400 schrieb Jorge Fábregas:
> >>>> Hello Everyone,
> >>>>
> >>>> I'm running Fedora 8 and my system freezes (for about 20 to 40 seconds) a 
> >>>> couple of times a day. When it does I see this on /var/log/messages:
> 
> >> -- 
> >> Robin Laing
> > 
> > 
> > Question to the devs - could you think of any way that the kernel output
> > could be a bit more informing, or don´t you get enough information from
> > the hardware for such an issue. I also checked smart for unusual power
> > cycle counts but to no avail. 
> > 
> > Henry
> > 
> 
> Henry, it would be nice but if the system, including the BIOS doesn't 
> know that there is a problem with the power supply, then how is the 
> hardware supposed to report it.  Maybe there can be a sensor added to 
> the harddrive to detect this type of error.

A sensor on the hardrive would be a good start I guess. But even better
would be a sensor on the power supply. I know that this is not something
which can be resolved on the fedora list, since it involves hardware
support which, as Alan already stated, is obviously not there. 

> I have looked at how lmsensors works and you can make some changes to 
> the configuration files to increase the accuracy of the reports.  I have 
> not played with it much though.  It is just as easy to pull the cover 
> and measure the voltages with a volt meter.

I will try lmsensors and see what it can do here. If this works out,
than my initial questions stands: Could such an info be included in the
kernel output. Maybe not (by design) but a question can´t hurt. We do
development within SAP here at our company, and for special cases, when
something fails, we don´t only throw a error message to the user. We try
to retrieve as much information from the underlying OS/DB as much as
possible to help for later debugging or at least being more informative
in the message we display to the user. 


> I have two of the same brand of power supply's (only ones available in 
> our area) with exactly the same fault.  A known fault that can be fixed 
> with the addition of three variable resistors.

Don´t get me wrong here, but this is not something I would like to see
in one of my systems here. I could imagine the first response from one
of our managers if anything fails (even if it is only a web page which
is not displayed properly) that this must be due to the newly installed
resistors ;-)

Henry 







More information about the fedora-list mailing list