[pvrusb2] pvrusb2 Digest, Vol 26, Issue 6

[Matt Nyerges]

On 10/16/07, pvrusb2-request at isely.net <pvrusb2-request at isely.net> wrote:
>
>
> Date: Tue, 16 Oct 2007 10:31:12 -0500 (CDT)
> From: Mike Isely <isely at isely.net>
> Subject: Re: [pvrusb2] Getting error in increasing frequency
> To: Communications nexus for pvrusb2 driver <pvrusb2 at isely.net>
> Message-ID: <Pine.LNX.4.64.0710161018530.21458 at cnc.isely.net>
> Content-Type: text/plain; charset=us-ascii
>
>
> On Mon, 15 Oct 2007, Matt Nyerges wrote:
>
> > I noticed this error a few times last year, but not much.  Recently,
> > however, I see this error a lot more.  Whenever it happens my program
> that
> > cat's /dev/video0 gets killed when the video0 file gets removed then
> added
> > back.  I wrote something that watches for this and resumes recording,
> but it
> > would be great to find out what is causing this.  Has anyone seen
> anything
> > like this?
>
> Matt:
>
> This sounds a lot like a piece of hardware is drifting out of tolerance.
> The error here is a USB protocol failure.  Everything else you see is a
> cascade from that first error.  The error code -71 is coming from the
> USB core in Linux; it is a result of an error in communicating with the
> USB device.  Generally this can indicate bad cables, a flakey hub (if
> you are using one) or possibly USB devices experiencing some kind of
> power supply issue (glitches, noise, voltage out of spec, etc).
>
> A couple things you can try:
>
> I assume you only have a single device so unfortunately you can't swap
> pvrusb2 parts around.
>
> Try using a different USB port (or even better, controller).  Honestly I
> don't expect that is going to make any difference but it's easy to try
> and if it does make a difference then that's a clue.
>
> Can you run the device under Windows with Hauppauge's drivers?  If so,
> then clearly if the hardware is having trouble then you should also see
> problems there - at which point you should RMA the device.
>
> There's a module option you can set called "tolerance" that will try to
> paste over this sort of problem in the driver.  Basically it will cause
> the driver to ignore -71 errors if they remain at a rate below a set
> threshold.  This is NOT a fix but it is a way to endure the pain a
> little better.  It also is far from perfect because certain cases of
> that error are simply not recoverable.  But it might help.  See here for
> more info on the parameter:
>
> http://www.isely.net/pvrusb2/usage.html#Logging
>
> It is worth emphasizing that if setting the tolerance parameter helps
> you then you very likely have some kind of hardware problem going on
> that you should fix.  Eventually, even with tolerance enabled, you're
> still going to get a -71 error at an inopportune point which will still
> kill control of the device.  (It's also possible that tolerance won't
> help at all, if in fact this is a case where the device is actually
> crashing.)
>
>   -Mike
>
>
> --
>                         |         Mike Isely          |     PGP
> fingerprint
>      Spammers Die!!     |                             | 03 54 43 4D 75 E5
> CC 92
>                         |   isely @ pobox (dot) com   | 71 16 01 E2 B5 F5
> C1 E8
>                         |                             |
>
>
> ------------------------------
>
>

Mike,

Thanks for the prompt reply.  I changed USB port, didn't really notice
anything (like I said, it is fairly intermittent.)

I changed the outlet for power and changed the power cord to a universal and
again didn't notice anything.

I changed the cord that ran from my splitter to the unit and again didn't
really notice anything.

Then someone went downstairs, and there was the error!!!

What happened,  my wife turned on the light over the foyer.  Turning it on
and off will 90% of the time cause the unit to error.  So, I tested all the
switches in the house and found that the fan switch in the bathroom does the
same thing.

So, looks like I need to get a UPS on it and all should be fine.

Thanks again for your help on this.

-Matt