gnu-arch-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnu-arch-users] Arch Configuration: Replication and fail safe


From: Brian May
Subject: Re: [Gnu-arch-users] Arch Configuration: Replication and fail safe
Date: Wed, 11 Feb 2004 09:12:50 +1100
User-agent: Gnus/5.1006 (Gnus v5.10.6) Emacs/21.3 (gnu/linux)

>>>>> "ALEJANDROLANDIM" == ALEJANDROLANDIM  <address@hidden> writes:

    ALEJANDROLANDIM> I'd like to have two or more machines always
    ALEJANDROLANDIM> synchronized using tla.  I want to have
    ALEJANDROLANDIM> replication and be virus/trojan safe.

I am not sure precisely of your requirements, do you want
- the arch archive to be synchronised?
- a copy of the project files to be synchronised between the two computers?
- both?

What type of data?

For a starting point, lookup the section in the tutorial on creating
and using mirrors.

    ALEJANDROLANDIM> If one machine disappears, then the other ones
    ALEJANDROLANDIM> have all the information.  If one machine gets
    ALEJANDROLANDIM> compromised and the archive modified without
    ALEJANDROLANDIM> knowledge of anyone the other machines have all
    ALEJANDROLANDIM> the changesets and a possible alarm could go on
    ALEJANDROLANDIM> notifying that some mirror is in bad state.

Also, lookup information on using GPG with the latest development
versions of TLA.

    ALEJANDROLANDIM> Is this configuration possible?  Can the machines
    ALEJANDROLANDIM> replicate automatically (whenever there is a link
    ALEJANDROLANDIM> between them) using SSL or SSH or GPG?  Do you
    ALEJANDROLANDIM> have any better configuration?  Anyone knows
    ALEJANDROLANDIM> about other solutions?

It should be possible. Is it feasible and the best solution for you?
You will have to work that out.

I once tried using arch to manage my website (327 Megs, most of it is
binary photos), but concluded that TLA was extremely slow to manage
such data (I suspect this will improve with proposed optimisations),
even with a revision tree. Not to mention the other issue of limited
disk space on the server ;-).

For now, I have abandoned this approach, although I have kept the *.id
files around to make it easier to redo at a latter date.
-- 
Brian May <address@hidden>




reply via email to

[Prev in Thread] Current Thread [Next in Thread]