On Fri, 2005-08-26 at 19:08 +0200, Brian Foster wrote:
> | Date: Fri, 26 Aug 2005 11:50:03 +0100
> | From: P at draigBrady.com> |
> | Patrick Stack wrote:
> | > I'm trying to backup my /home directory prior to upgrading FC4 using
> | > afio with cdrecord with no success.
> | > I'm using the following command which I've adapted from the example
> | > given in [ afio(1) ].
> | > find /home/patrick/* | afio -o -b -K -b 2048 -s 614400x '!cdrecord dev=0,0,0 -'
> | >
> | > This seems to work and checks the data (-K) [ ... ]
>> why do you think it “works“?
> why do you think it really is checking the data?
> what, in fact, do you think it is doing when
> checking the data?
I must admit I have little idea what it is doing and haven't thought
about it either. I'm a relative beginner having worked with Linux for
only 11 months and that mainly on php and mysql.
I'm picking things up on a need-to-know basis as time is pressing. So
all I'm interested in is getting things to work as quickly as possible.
> like Pádraig, I've never used afio(1) either,
> but I suspect at least part of the problem is
> the `-K'. it is very likely — and the manual
> page (for v2.5) implies this — that `-K' does
> read-after-write verification. it is unclear
> if it verifies on a sector-by-sector or image-
> by-image basis, but in either case, it cannot
> possibly work.
Yes, you're probably right. I stuck in the -K switch because I wanted
some way to verify the backup on the fly, which was probably a bad idea.
>> to start with the obvious, `afio' has _no_idea_
> what device to read back! and then, _if_ it is
> sector-by-sector, it clearly cannot work: you
> cannot read a CD that is still in the process
> of being recorded.
>> the example of using cdrecord(1) in afio(1) does
> not use `-K'.
> |[ ... ]
> | I've never used afio but I think the problem is
> | that cdrecord expects iso images not arbitrary data.
>> eh? AFAIK, in `-data' mode (which is the default),
> `cdrecord' simply writes a data-format CD, and does
> not give a hoot what about the format of the image.
> e.g., I have CDs that are tar(1) images, albeit I
> did not burn them and do not know if `cdrecord'
> or something else was used: but they do prove
> data-format and ISO-9660 are not isomorphic.
>> having said that, I concur Patrick should write
> ISO-9660 (Rockridge) images. there may be only
> one file in the ISO-9660 — an `afio' archive —
> but the containing software format should be
> ISO-9660, unless there is a Very Good reason
> to the contrary.
Having lost patience with afio I decided to use tar instead.
>> cheers!
> -blf-
> --
> Experienced (20+ yrs) kernel/software Eng: | Brian Foster Montpellier,
> • Unix, embedded, &tc; • Linux; • doc; | blf at utvinternet.ie FRANCE
> • IDL, automated testing, process, &tc. | Stop E$$o (ExxonMobile)!
> Résumé (CV) http://www.blf.utvinternet.ie | http://www.stopesso.com
Maintained by the ILUG website team. The aim of Linux.ie is to
support and help commercial and private users of Linux in Ireland. You can
display ILUG news in your own webpages, read backend
information to find out how. Networking services kindly provided by HEAnet, server kindly donated by
Dell. Linux is a trademark of Linus Torvalds,
used with permission. No penguins were harmed in the production or maintenance
of this highly praised website. Looking for the
Indian Linux Users' Group? Try here. If you've read all this and aren't a lawyer: you should be!