gmirror existing installation?

Oliver Fromme olli at lurza.secnetix.de
Fri Jan 26 19:04:47 UTC 2007


R. B. Riddick wrote:
 > Oliver Fromme wrote:
 > > Not really much more.  It's true that you have to copy over
 > > your files, but the computer does all that work itself.
 > > You only have to type one command, then spend your time
 > > doing something else.  You don't have to watch it during
 > > the copying.  (You don't have to watch the mirror being
 > > synchronized either, which has to be done anyhow and takes
 > > much longer, probably.)
 >
 > The service interruption is higher, because copying the system should
 > be done with a read-only root-fs...

That depends on your services.  I have done that procedure
several times without shuting down services.

You can also copy over your stuff once while services are
running, then stop the services (and remount red-only if
you think it's necessary) and perform a final sync that
will transfer only the files that have changed [*].
That will be pretty quick, and the downtime will be short.

 > Both methods need a reboot in order to change the root fs...

That's true.  At some point you'll have to switch your
root FS to gm0 ...  That's not avoidable.

 > It is always better to design the system before it is installed... :-)

Very true.  :-)

Best regards
   Oliver

[*]  E.g. you can use cpdup for that purposes, from
ports/sysutils/cpdup.  It's fast and easy to use.

-- 
Oliver Fromme, secnetix GmbH & Co. KG, Marktplatz 29, 85567 Grafing b. M.
Handelsregister: Registergericht Muenchen, HRA 74606, USt-Id: DE204219783
Any opinions expressed in this message are personal to the author and may
not necessarily reflect the opinions of secnetix GmbH & Co KG in any way.
FreeBSD-Dienstleistungen, -Produkte und mehr:  http://www.secnetix.de/bsd

"And believe me, as a C++ programmer, I don't hesitate to question
the decisions of language designers.  After a decent amount of C++
exposure, Python's flaws seem ridiculously small." -- Ville Vainio


More information about the freebsd-geom mailing list