[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Duplicity-talk] a quick suggestion and a question
From: |
Ben Escoto |
Subject: |
Re: [Duplicity-talk] a quick suggestion and a question |
Date: |
Sat, 16 Nov 2002 10:53:13 -0800 |
>>>>> "RB" == ry4an-duplicity <address@hidden>
>>>>> wrote the following on Sat, 16 Nov 2002 00:54:06 -0600
RB> I love the whole idea of duplicity. One thing that threw me for
RB> a bit a loop which I think should be included in the document is
RB> that the scp: backend paths are relative to the user's home
RB> directory.
Ok, I added this as the first FAQ question at
http://www.nongnu.org/duplicity/FAQ.html
RB> Have you considered going with the scp standard and using:
RB> scp://address@hidden:/mnt/backup/toohey
RB> The two colons in there still yeilds a perfectly valid URN given
RB> the RFC2141 URN definitions.
Perhaps I'm read the RFC wrong, but in section two it says:
All URNs have the following syntax (phrases enclosed in quotes are
REQUIRED):
<URN> ::= "urn:" <NID> ":" <NSS>
which I think means that all URNs have to start with "urn:" and not,
for instance, "scp:". Also, this paragraph from RFC2396 section 1.2:
A URI can be further classified as a locator, a name, or both. The
term "Uniform Resource Locator" (URL) refers to the subset of URI
that identify resources via a representation of their primary access
mechanism (e.g., their network "location"), rather than identifying
the resource by name or by some other attribute(s) of that resource.
The term "Uniform Resource Name" (URN) refers to the subset of URI
that are required to remain globally unique and persistent even when
the resource ceases to exist or becomes unavailable.
implies that a URL is the appropriate category, because what duplicity
wants is not just a location but an access mechanism. Finally, some
examples given in that same rfc (section 1.3):
ftp://ftp.is.co.za/rfc/rfc1808.txt
-- ftp scheme for File Transfer Protocol services
gopher://spinaltap.micro.umn.edu/00/Weather/California/Los%20Angeles
-- gopher scheme for Gopher and Gopher+ Protocol services
http://www.math.uio.no/faq/compression-faq/part1.html
-- http scheme for Hypertext Transfer Protocol services
mailto:address@hidden
-- mailto scheme for electronic mail addresses
news:comp.infosystems.www.servers.unix
-- news scheme for USENET news groups and articles
telnet://melvyl.ucop.edu/
-- telnet scheme for interactive services via the TELNET Protocol
suggest the foo://..../ syntax, since scp: seems analogous to ftp,
http, and telnet.
Still though, I'm open to alternate ways of specifying things, and
I haven't read any rfc in detail. Perhaps you can find something
that would let me allow some more intuitive syntax..
RB> That said, I'm wondering what solutions people have used for
RB> storing their password in backup scripts. I want to set up a
RB> cron job to run my nightly backups, but I don't want to just put
RB> the password in the script. What's everyone else doing?
Probably I shouldn't be announcing this in a public list, but I just
put the password in the script. But currently I'm ftping to a
computer that doesn't have anything valuable on it. If you want to
use scp and can set up your account on the remote site, try giving
yourself a chrooted account that can only access the backup
directory.
You may find these instructions helpful too:
http://www.arctic.org/~dean/rdiff-backup/unattended.html
they describe how to set up rdiff-backup for unattended access, but
perhaps they can be adapted for duplicity. The idea is that you can
give yourself passwordless access to only certain commands. I don't
know, maybe there is some way of doing that with scp too.
--
Ben Escoto
pgpe5O7tQNKWV.pgp
Description: PGP signature