[RFC] /var versus /srv

Krzysztof Halasa khc at pm.waw.pl
Sat Sep 29 00:22:40 UTC 2007


Lamont Peterson <lamont at gurulabs.com> writes:

> Let's just go to an example.
>
> Bob runs IT for a large organization.  The systems that Bob deals with
> are firmly in the category of "Enterprise-Grade."
>
> Bob has 10,000 server to backup.

So anything non-SOHO means 10,000 servers and perhaps hundreds
of thousands people. Ok, let it be.

> Each of Bob's servers has an average of, say, 30GB of stuff (everything
> on the disk added up together).
>
> Of that average of 30GB/server, only 4GB is not data the server manipulate.
>  Or in other words, 4GB is OS, libraries, application code, etc. (of
> course, this number would be significantly higher for Bob's Windows
> server, but let's keep this scenario simple for now, shall we?).

IOW, 13.3% of your data is the OS.
How about a bit larger systems? Like the one I have here with
ca. 440 GB of data? 1% = the OS. It's still a small machine.

I don't know about MS Windows.

> So, 10,000 times 4GB of stuff that's either the same on all those servers
> or otherwise doesn't need to be backed up because Bob can reinstall it
> by other means.  Let's see that 40TB of data he'd be backing up if he
> followed your advice.

40 TB per 10,000 servers, with the whole backup size of 360 TB?
Seems fair, don't you think?
Assuming, of course, that you store all copies of identical files.

BTW: are those 10,000 servers identical (WRT to the set of RPMs)?
Why don't you use a master server and replicate?

>  Oh, and it would take *longer* to restore those systems.

Assuming you would reinstall your OS using kickstart in no time :-)
it would take just 13% longer. Or just <1% longer in another case.
Now I'm not exactly sure about this "no time kickstart". And, after
the install finishes, you do a "yum update", don't you?
What if the installation, for example, keeps crashing for some
reason till morning?

> He wasn't advocating that the quality of the backup wasn't important.
>  What I got from it was that he's merely saying that *state*
> information may no longer be valid by the time one needs to restore
> from backup.

Sure, but the fresh backup is still way better than a fresh install.

> I'm sorry, I must have missed the part of this thread where people
> were wanting to change things in order to break the system.

Do you think moving things out of /var won't break anything?
-- 
Krzysztof Halasa




More information about the fedora-devel-list mailing list