Re: slink_cd v 1.07 out
Steve McIntyre <stevem@chiark.greenend.org.uk> writes:
> On 24 Feb 1999, John Lapeyre wrote:
>
> > Hi, I am trying slink_cd 1.07 and have several problems.
>
> OK, fire away.
>
> > image1 does not seem to work. It still trys to build all images.
> > changing DISKLIST by hand in the script takes care of most of this.
>
> Really, what errors are you seeing? If it's a complain about missing volid
> files, don't worry. It's a bug, but a non-serious one. Ah, or are you just
> noticing it's building trees for the others too? What command-line are you
> using?
Sorry , I don't have it anymore. I used image1 and not images, I got no
error messages ( execpt maybe no volid, no extras for disk 1 ). Yes, I was
noticing that it was building trees for other disks. Maybe this is
necessary ? I didn't make it far enought to see which images were built.
> You should run the genlist step to make _any_ binary CDs, unless you want
> to supply the complete list of packages yourself. Perhaps I should
> document this better, it was not always the case.
This probably wasn't too clear. Eventually, the docs should
be improved, but at this point there are probably more important things
for you to do, like making sure the images are perfect within a week (pressure ! :-) )
I did genlist and things seemed to be working. I can't finish the
test, because my mirror is not complete. I moved it from one HD to another
and the links into hamm were not made (probably forgot the -a in cp -ar ).
Ill go back tonight, maybe, and
copy the my original mirror of hamm and slink, and give it a try again over
the next couple of days. I don't plan on modifying your package lists
any, but at least you will have one more data point.
I have trouble mirroring. I finally got 'fmirror' to work
pretty well. 'mirror' sometimes transfers only a few K per s (don't
know why). 'lurkftp' used to work a little for me. Now it just kind of
sits there and doesn't do anything. fmirror just starts downloading
files as fast as my connection will allow, which is what I want. I
found that includes and excludes don't quite work as documented, but
after an hour or so of work, I got something that did the trick. I
had to mirror all of hamm with fmirror, because if I ask it to follow
links, then it makes a seperate mirror for frozen and slink, and god
knows what else it would duplicate. I guess with some scripting, one
could make something that works more efficiently.
thanks again for the work.
--
John Lapeyre <lapeyre@physics.arizona.edu>
Tucson,AZ http://www.physics.arizona.edu/~lapeyre
Reply to: