Storage probing problem

[David Zeuthen]

Hi,

Sorry for the lag,

On Thu, 2006-09-28 at 15:35 +0300, S.Çağlar Onur wrote:
> 27 Eyl 2006 Çar 19:41 tarihinde, S.Çağlar Onur şunları yazmıştı: 
> > Thanks for info but i think this problem is not related that one, i added
> > several lines to debug and i think i found the problem (double free causes
> > seg. fault), im trying to find why its happining and a proper way to fix :)
> 
> OK further investigations shows me there is _may be_ a problem with partutil/ 
> code inside the hal. 

Right. It's problematic that I can't reproduce it either :-(

> I said maybe because in order to confirm this i replicate the same partition 
> table structure into a usb disk but _unmodified hal_ works well without a 
> problem on this and i cant reproduce this error again.
> 
> On this problematic system first probe-volume crashes on part_table_free call 
> if this call removed then probe-volume crashes on the same call and also 
> part_table_find can't fill pe_offset and pe_size properly. 
> 
> But somehow for testing purposes attached _stupidly wrong patch_ works for 
> this system and hal probes the disk as aimed on that machine (didnt test on 
> other machines). 
> 
> I'll investigate more tonight but im not sure to where to look? I'm also try 
> to find any problem on partition table on disk but fdisk and other partition 
> probe/change tools didnt report any errors so tables seems good. Any idea?

Any chance you can take the usb disk, dd the image up (e.g. 'dd
if=/dev/sda of=usbdisk.img'), zip it, and send it to me personally, e.g.
off-list? If it's huge alternatively put it online somewhere and send me
the URL.

Probably a good idea if you can zero out the actual data partitions in
order to reduce size and if you are worried that I'm going to look at
the data. Thanks.

     David