Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Lab 4

0 views
Skip to first unread message

Kevin

unread,
May 8, 2005, 11:34:03 PM5/8/05
to
I was able to match the md5sum, contents, and rc's of the following tests:
teste07.txt
test07.txt
test08.txt

but on teste03.txt I was able to match only "contents", it seems that the
only difference
between the rc's is that the processes are writing in different order, thus
getting different
block numbers. The first process that writes is process 1300 using block 5
and not 200 using block 5 like their rc. Has anyone run into this problem or
knows what's wrong?


Murat Deligonul

unread,
May 8, 2005, 11:38:30 PM5/8/05
to

Yes, this happens because of the differences in everyone's scheduling
method.

There has been discussion about this already. I believe we've been told
that for lab 5 (and lab 4 regrades) that only the file contents will be
checked (using DiskDumper) instead of just an md5sum.

So as long as the files have the correct data it won't matter what blocks
they occupy.

Murat

Sidd

unread,
May 8, 2005, 11:39:47 PM5/8/05
to
Yup, a lot of us have. U can see my post (Lab4 prob) as well as the main
post "teste03.txt scheduler". The consensus seems to be that as long as the
diskdump matches, ur good. And as we only use the diskdump for the contents,
and not for the rc, I'm assuming we should be good.


"Kevin" <kev...@hotmail.com> wrote in message
news:d5mlnh$8pd$1...@geraldo.cc.utexas.edu...

Emmett Witchel

unread,
May 9, 2005, 3:05:35 AM5/9/05
to

Hi,

We just posted a modified version of ReadCheck.java which does not print the block number. Please check that your output matches the new blockless rc file.

Emmett

0 new messages