[mythtv] Re: New idea for storing recordings to disks

Kevin Elliott kevin at phunc.com
Thu Jun 30 19:22:57 UTC 2005


I tend to agree, I use RAID, but I also have a TeraServer device on my
network that I'd like to utilize. I have over 2TB of disk space for those
two alone, then add on the additional 300gb drives in random machines. I'd
like to make use of all the space, without disturbing the contents on the
drives already (i.e. I don't want to thrash the OS's just so I can do raid
over Ethernet, or something like that)...

I think having multiple mount point hunt would be good.

-----Original Message-----
From: mythtv-dev-bounces at mythtv.org [mailto:mythtv-dev-bounces at mythtv.org]
On Behalf Of Joseph A. Caputo
Sent: Thursday, June 30, 2005 8:08 AM
To: mythtv-dev at mythtv.org
Subject: Re: [mythtv] Re: New idea for storing recordings to disks

On Wednesday 29 June 2005 18:13, Hamish Moffatt wrote:
> On Wed, Jun 29, 2005 at 09:50:29AM -0400, Joseph A. Caputo wrote:
> > Here's a pie-in-the-sky thought (and I realize this is probably 
> > *way*  
> > overkill, but it's an interesting idea nonetheless).  What if Myth 
> > had  
> > a sort of virtual filesystem?  That is, not a filesystem per se, but 
> > more of a content management layer, where a recording could be 
> > 'sliced'  
> 
> I think that's a bad idea personally. 

That's fine, everyone is entitled to an opinion.

> UNIX design philosopy is to  
> split 
> up systems into small components that connect together. The kernel is
> good at aggregating storage (RAID, LVM etc) so don't replicate that in
> the application.

Yes, but LVM and RAID operate in a 'best-case' world where hardware 
rarely fails.  When hardware does fail (or even when you want to make 
changes to the volume group, etc), it usually requires superior 
sysadmin-type skills to manage the situation.  It's virtually 
impossible to deal with a failed disk in an LVM group or RAID array at 
the application level.  Think of the WAF.  If Myth managed its own 
storage, then it could mark 'damaged' (some slices lost) or 
'unavailable' (all slices lost) recordings.  The user could set (via 
gui) a policy on what to do with these (auto-delete, or keep and let 
user decide).  

> I don't really understand why you wouldn't just put enough disk space 
> in 
> the backend. Disk is cheap.

Yeah, and 640K ought to be enough for anybody :-)
Plus, the cheaper it gets the more we'll want a fault-tolerant method of 
dealing with multiple disks :-)

-JAC
_______________________________________________
mythtv-dev mailing list
mythtv-dev at mythtv.org
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev



More information about the mythtv-dev mailing list