[mythtv-users] ANN: PlugMyth, a new MythTV distribution for Plug computers

[Tortise]

----- Original Message ----- 
From: "David Whyte" <david.whyte at gmail.com>
To: "Discussion about MythTV" <mythtv-users at mythtv.org>
Sent: Thursday, August 26, 2010 10:04 PM
Subject: Re: [mythtv-users] ANN: PlugMyth,a new MythTV distribution for Plug computers


On Thu, Aug 26, 2010 at 7:57 PM, Simon Hobson <linux at thehobsons.co.uk> wrote:
> Christopher Kerr wrote:
>
>> I think for many (most?) users here, sleeping the master backend is an
>> unacceptable solution, because you lose mythweb.
>
> Indeed. Perhaps "best of both worlds" would be to run master backend on a
> plug (always on, very low power), and have it wake up slave BEs as needed.

>Looks good, except I run ZoneMinder on my BE and I am pretty sure a
plug PC won't have the grunt for that :(

While the Sheeva plug was the impetus for this software another supportive poster reports using an old 600M Laptop as well.  In 
terms of the myth community for me this project is all about using small power draw PC's and conserving power, for me all great 
objectives.  Laptops are great for power efficient computing, I've got a few 400M to 1200M ones lying around waiting for a new 
purpose in life.  Now if only....

Some thoughts occur to me when considering the arguments presented.

A system like this is presented as good for a small scale mythtv installation.  Is it expandable to work for more extensible multi 
tuner systems?  (I now have one BE with 2 dual Tuner PCI cards and 3 HDD's, 120G for OS, and 2 x 2TB's now all in one box under a 
430W PS)

While having one USB socket has a power and input limitation, using a powered USB hub presumably adds a whole new dimension of drive 
and tuner connectability.  I haven't tested formally, however my expectation is that when the PC powers down, the USB hub power 
drawer also idles to some lowish figure.  Conversely when on I expect (but do not know) that it would be all on, i.e. there is no 
possibility of powering HDD's connected to the USB hub on demand.

One thing I've been contemplating for some time is having small Ethernet devices that interface into mains relays.  They would be 
controlled over Ethernet in various ways, such as power on, power off, power on for 30 mins, power on @ 24:00 for 30 mins etc. 
While such devices would have broad applicability they could be useful here, e.g. turn NAS on and off or turn USB hub on and off! 
(They could be powered with their own PS, borrowing power from local devices or POE, however how many of us have POE switches at 
home?)  Another example use is turn on LAN printer, via a LAN webpage press the printer power button, for a print job and auto turn 
off when finished, say after 15 mins.

Similarly a serial output from a low power PC might be used to control the power of the USB hub, (thus the HDD and Tuners) using a 
relay.  (One hack I've realised for doing this is to take an old modem with a relay, steal (rewire) the relays switched outputs, and 
send appropriate commands to the modem to control the relay!)

I acknowledge the comments about more power savings might be sought elsewhere in the house, however the NIX technology to control 
these things has to start somewhere, on a smaller scale - and it also becomes extensible.

The arguments for separate HDD's seem to revolve around data streams being written to differing parts of the disc.  It seems to me 
the new electronic discs do not have this problem, and if so they should be designed into the equation for central BE use, while 
HDD's are powered on for storage and turned off again when not being used.  Spelling this out a recording stream is saved to the SSD 
during download, but when completed is then dumped onto a HDD turned on for the purpose, and then turned off again.  So HDD's would 
run on demand instead of all the time.  Decreasing significant SSD cost acknowledged, however in the context of potential power 
savings they might be seen as a cheaper option.

Anyway some thoughts that might trigger further useful discussion.

While not enjoying much positive feedback I'd like to express thanks to Tyler T for taking the trouble to assemble and present the 
software.