[DRBD-user] drbd-0.8pre6 (svn rev 2590) - kernel fails

Lars Ellenberg Lars.Ellenberg at linbit.com
Wed Nov 8 17:37:54 CET 2006

Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.


/ 2006-11-08 09:37:30 +0300
\ Vitaly Kuznetsov:
> Lars Ellenberg wrote:
> 
> > 
> > there are not too many ways __bio_clone
> > could do something stupid like that.
> > 
> > from the disassembly of the "Code:", this is the
> > 
> > 	memcpy(bio->bi_io_vec, bio_src->bi_io_vec,
> > 		bio_src->bi_max_vecs * sizeof(struct bio_vec));
> > 
> > so bio->bi_io_vec (esi) is 0xea66e000,
> > and for some reason your kernel
> > thinks it cannot access that address.
> > 
> > no more I can figure from what we have so far, sorry.
> > 
> > 
> 
> Maybe I'll insert some debuginfo? Or there are some other ways to debug
> it? But what and where?
> This (bug?) occurs regulary on different servers.

I have no idea, sorry.
this appears to me to be a valid address.
it is not even drbd that "invents" this address,
it is the bio_split or bio_alloc or whatever.

and then there is a memcpy from some source (passed in by upper layers)
to some destination address (provided by kernel allocation functions)
that fails with an invalid paging request...
really, I don't know what to do about that.

maybe ask some vm guru, why exactly a kernel paging request for an
address that apears to be perfectly valid can fail...

-- 
: Lars Ellenberg                                  Tel +43-1-8178292-0  :
: LINBIT Information Technologies GmbH            Fax +43-1-8178292-82 :
: Schoenbrunner Str. 244, A-1120 Vienna/Europe   http://www.linbit.com :
__
please use the "List-Reply" function of your email client.



More information about the drbd-user mailing list