On Fri, 26 Apr 2002, Sean Davis wrote:

> Okay, I updated lfs_cleanerd, got rev 1.11 of print.c in it now. So far so
> good - none of those messages in syslog yet. I'll let you know if anything
> else weird happens.

> One question re: the difference between 1.10 and 1.11: Is the main effect of
> the change increased verbosity of syslogged stuff, or was the problem with
> it not being able to allocate memory due to it using u_long instead of
> u_int32_t?

The "main" change from the point of view of your problem was

> - datap = (u_long *)malloc((lfsp->lfs_ssize*lfsp->lfs_frag) * sizeof(u_long));
> + datap = (u_int32_t *)malloc((lfsp->lfs_ssize / lfsp->lfs_fsize) * sizeof(u_int32_t));
                                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
it was trying to get too much memory, for real.  I'm not sure what kind of
machine you're running it on, but most likely for you u_long and u_int32_t
are equivalent; the u_long/u_int32_t change was because I was testing it
on Alpha, where they're not.
						Konrad Schroder
						perseant@hhhh.org