On Tue, Jul 19, 2005 at 10:36:40PM -0400, James K. Lowden wrote:
> bsd.petr@seznam.cz wrote:
> > James K. Lowden napsal(a):
> > >bsd.petr@seznam.cz wrote:
> > >  
> > >>i need to move my NetBSD 2.0.2 from old 4 GB HDD to new 1.2 GB HDD. 
> > >
> > >There are many free alternatives.  One is http://www.feyrer.de/g4u/.  
> > >
> > Thanks for above link. I have a idea. Is it possible way tu resize UFS 
> > partition without data-loss to a smaller (equal or lower then 1.2 GB) 
> > size (which program or command can do it)? 
> 
> Not that I know of, but more important, it's a bad idea.  
 
> One difference between newbies and gurus: gurus distrust magic. 
> Filesystems aren't *meant* to grow and shrink, and programs that grow and
> shrink them almost certainly -- not that I understand them in any detail
> -- take some shortcuts with your data.  If the "UFS shrinker" crashes in
> mid-shrink, what state do you suppose it will leave the filesystem in? 
> Where will your data be?  If it's OK to destroy the filesystem, why are we
> moving the data in the first place?  

i've been following thi thread with much interestand while i've not the
same requirement at present i am about to embark on a new way of
solving an old issue for myself and us few who are involveded
writing/publishing project ... 
 
> Treat your data with care.  Copy, verify, then remove.  Not only will you
> keep your data intact, you'll live longer, too.  :-)

with regards to this segment .. i'm new'ish to using tape backup but not to
bsd (freebsd since 2.0.5-release, aprox 10 year and netbsd a stint with
v0.9 and then v1.4.1 and i've now got a 2.0.2 cdrom that will become a
permanent fixture on a pentium pro 180 or a dec alphastation 255 with
233 mhz cpu.

one of these machines will have a tandberg tdc-4222 (5 gb compressed
acording to tandberg or 2g3 byte raw capacity) scsi streamer fitted and
become the network backup 'server' using the amanda package.

i'm new to this whole concept of backup/restore. i've used tar (gtar,
on ibm os/2 'server' in novell network for some years) can my
experience with that be converted to netbsd nouce so to speak ?

or, is the an alternative way, that would provvide better overal system
reliability and data integrety ?? i have tow needs to be accomodated ..

1, weekly partition (srry if that should be slice, i still have not
gotten used to teh diferent terminology), as its called in os/2
parlance.

2, daily several directories that need to be put on ice for a just
incase something goes wrong and (2a if you will) as required developer files
backed up for archival or keep while media cleaned up and build
restarted type stuff ... general house keeping.

can i build cron sequences that check 'flags' that then go about teh
various tasks and do whats required as the need arises.

just one more side track, i plan to use postgresql as teh
network/domain/web backbone rdbms antbody got some comments/war stories
about its 'hardiness' as regards living in a netbsd environment and how
it copes with having its files poked, proded and interupted (more or
less constantly) by some some out of school jockey trying to get it to
stream to tape 

much thanks appreciations and gratitued for work done on netbsd (hwole
bsd tree infact) its been a really nice fall back position and has
saved our data on many ocassiones. i (and teh other 4 old timers) are
looking to migrating from our old toolkit (ms dos - ibm os/2)
wordperfect, word for windows, framework and stuff we have built for
ourselves over the years that has made life easier for us over the
years.

hopefully we will be able to progress with the publishing side of our
project as smoothly, easily and in teh sme timely manner that the
(netbsd/freebsd) lan/wan has kept us connected over teh past 10-12
years  

regards

jonathan

-- 
================================================================
powered by ..
QNX, OS9 and freeBSD  --  http://caamora com au/operating system
==== === appropriate solution in an inappropriate world === ====