Re: Best way to "Clone" a Linux Hard drive?

fl0w3r_w13ld1ng_$0ld13r_at_salmahayeksknockers.edu
Date: 02/26/04 

Date: Thu, 26 Feb 2004 09:44:58 GMT

P.T. Breuer <ptb@oboe.it.uc3m.es> wrote:
> It's not logfile integrity, but file system integrity. If the file

True, however, if we're talking about filesystem /var/log, we're talking
logfile integrity. Again, you jump to a conclusion with insufficient
information. If I have an HD failure, I'm not concerned if I can recovery
logfiles, only a working kernel, working applications, and user data.

> It's clear from everything you say that you are woefully ignorant of
> almost everything that constitutes "normal" knowledge of the operating
> system. It's taken this long to get you to the state of almost
> understanding what the two most obvious courses are, and what their
> tradeoffs are. It wouldn't take anyone normally competent more than a
> millisecond to do so.

Again, I think you make gross assumptions of my knowledge based on the
choices I made, which you pipe through your own arrogance.. I have
not covered some avenues, not because I am unaware of them, but because they
were already discarded as insufficient to meet my requirements, or required
further massaging in order to do so.

> Do you really think you can fake it?

I must be faking adequately, as I've been SAing HPUX, AIX, and Solaris for 5
years now, during which time I've had the pleasure of working with some of
the biggest iron for 1500 miles, not to mention many talented and
experienced SAs that I've felt honored to have learned from.

>> >> As mentioned, This process will be unattended. If it were unattended, and
>> >> I used a boot floppy
>> > You couldn't. You wouldn't be there to put it in the slot.
>> Goodness, thanks for pointing that out! That's why I discarded the option
>> in the first place...
> Nice of you to tell us, no?

Was pointed out previously in the thread. Assumed you saw it.

>> > I have no idea what [Your requirements] are, if that is what you mean!
>> >How could I?
>>
>> They have been explicitly covered in the thread.
> I see only what you write. Explicitly or otherwise.

I'll lay them out for you: I want to clone my hard drive to an identical
model, within the following constraints:

1) Unattended and cronable.
2) Minimum duration.
3) Minimal impact to system availability.
4) Maximum integrity of data.
5) Minimal effort to restore after HW failure (Assumed attended).

> What you are talking about is run-of-the-mill sysadmin-ery.

The basic tools are simple, but using them to reconcile the above contstraints
took some thought.

>> Truthfully, I find you expleasant to deal with, but are a knowledgeable
>> sounding board, and have offered a good idea or two (Especially passing
>> instructions to Lilo from the command line).
>They are not "ideas", but the most agonisingly patently obvious things.

I think you overestimate the universality of Linux-specific tools such as
Lilo.

(Is this the part where you helpfully inform me that lilo will compile on
 many platforms, and then take credit for teaching me about cross-platform
 portability?)

> It isn't simple, if you insist on using the hard disk at the same time.
> That's why one would normally not. If you insist on doing so then you
> have to put it in readonly mode, or your copy will be incoherent.

Absolutely incorrect. I have completed a clone, as I desired, meeting all of
the above requirements.

Here's how:

-Replaced RL4 scripts with all of the shutdown scripts for RL1.
Added script "S21Clonedisk" to rc4.d
-S21Clonedisk, obviously, starts up desired minimal services (In my case,
Apache, Bind, SSHd), Does the DD, and when complete, changes to RL3.

It's all kicked off with an "init 4". I've tested it, and made a clone in 2
hours, with 0 loss of necessary (to me) system functionality. After, the
system returned flawlessly to 100% functionality. The clone booted up with
little more than millisecond-long JFS log replays, and no missing or
corrupted data that I could find.

> Unfortunately for you, killing things that have been started in a
> runlevel requires you to back down to runlevel S first.

As I figured I'd have to do from the start. I was hoping someone would have
a more elegant or superior suggestion, but no dice.

>> This is a reasonable suggestion, however, I've found that doing a system
>> "hot" takes too much time, with all of the different processes competing for
>> I/O.
> This not so! Processes do not "compete for i/o". There will likely be
> only one process doing i/o at a time! That's what "system load" measures.
> It's the number of processes waiting on i/o.

It is evident that some form of contention is taking place, whether I/O or
CPU. An 80Gb DD at single user took 2 hours, whereas one at RL3 took 12+.

> A sophisticated approach to this would probably use xfs snapshots.

Possibly. As far as I'm concerned, I have confirmed what I knew from the
start: The ideal method would be silvering and breaking mirrored filesystems,
once a week (automated via cron). Bad luck with Linux Raid in the past
sent me looking for alternative ideas.

> But I seriously doubt that you care that much.

Nasty, nasty!

> What I suspect is
> that you are running an http server which makes moment to moment
> changes in its configuration and content, and you want to back it up.

More or less true. However, the actual functionality of the server is
only relevant to creating the requirements. Once the requirements are
provided, the functionality that determined them isn't relevant, it's just
TMI. (Unless requirements are revisited).

> Only of course you're too busy to say so.

Too busy dismissing your arrogant, insufficient suggestions! Mea culpa,
however, for not being more explicit from the start with my requirements.

> Peter

Like many technical people, you are shamelessly arrogant and pedantic,
but this dialogue has underscored the necessity of being vigilant against
manifestations of such tendancies in my own behavior. 

-- 
.............................................................................
"Jewish villages were built in the place of Arab villages You do not even
 know the names of these Arab villages, and I do not blame you because the
 geography books no longer exist  Not only do the books not exist, the Arab 
 villages are not there eitherThere is not one single place built in this 
 country that did not have a former Arab Population"
 	      	       -Moshe Dayan, Israeli military hero and statesman 
		        (Ha'aretz, 04/04/1969)
.............................................................................
dswan@m3m3t1ccand1ru.com                        http://www.memeticcandiru.com