[Coco] Large file deletion fails on NitrOS9/6309
Jeff Teunissen
deek at d2dc.net
Mon Jun 1 00:29:15 EDT 2020
It's not something you have ever noticed, but it's not something most
people _would_ notice, since the only time it could possibly hurt
anything is if you try to delete one of the "offending" files. And
even then, most files are small, and the most common result of a "bad"
file deletion is a couple of sectors left allocated. I only noticed
because I found one that broke hard, and in another configuration
crashed the OS.
And yeah, it's probably something RBF should be able to deal with, but
it's not something it's ever had to deal with, because RBF is
incapable of making files that would cause it to overrun its buffer
space!
As far as I know, this "bug" is in every possible version of
OS-9/6809, so it's less a bug than a design limitation, and one our
tools should probably have to work around.
On Sun, May 31, 2020 at 9:40 PM Gene Heskett <gheskett at shentel.net> wrote:
>
> On Sunday 31 May 2020 20:25:50 Jeff Teunissen wrote:
>
> > Yes, that sounds exactly right, and probably a good idea.
> >
> > On Sun, May 31, 2020 at 6:31 PM Tormod Volden <lists.tormod at gmail.com>
> wrote:
> > > Jeff,
> > > Thanks for the explanation. So for a disk image / file system with 1
> > > sector/cluster, it would break if a file has a segment using e.g.
> > > both LSN 2047 and LSN 2048?
> > > Maybe for a start we could include a check in Toolshed os9dheck for
> > > this?
> > >
> > > Tormod
> > >
> I have to gently disagree guys, as its papering over a bug in rbf.mn. I'm
> no longer able to help, but in the long run this bug should be found and
> squashed. The dcheck check should advise the user of the problem since
> this is an admittedly rare case. With 2 hard drives, one dual formatted
> and running a cluster size of 4, and one fully os9 formatted running a
> cluster size of 16, both one GB drives with now very long grey hair,
> this is not something I have ever encountered.
>
> > > On Sun, May 31, 2020 at 5:47 AM Jeff Teunissen wrote:
> > > > Tormod,
> > > > 6x09 RBF will never make any file that crosses a 2048-cluster
> > > > boundary (it actually can't), but Toolshed has no facility
> > > > preventing it from making files that RBF will explode on when
> > > > trying to delete them. On a
> > >
> > > --
> > > Coco mailing list
> > > Coco at maltedmedia.com
> > > https://pairlist5.pair.net/mailman/listinfo/coco
>
>
> Cheers, Gene Heskett
> --
> "There are four boxes to be used in defense of liberty:
> soap, ballot, jury, and ammo. Please use in that order."
> -Ed Howdershelt (Author)
> If we desire respect for the law, we must first make the law respectable.
> - Louis D. Brandeis
> Genes Web page <http://geneslinuxbox.net:6309/gene>
>
> --
> Coco mailing list
> Coco at maltedmedia.com
> https://pairlist5.pair.net/mailman/listinfo/coco
More information about the Coco
mailing list