X-MimeOLE: Produced By Microsoft Exchange V6.5
Received: by onstor-exch02.onstor.net 
	id <01C87D65.06972ED2@onstor-exch02.onstor.net>; Mon, 3 Mar 2008 12:30:23 -0700
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
Content-class: urn:content-classes:message
Subject: RE: concrete info on CF status problem
Date: Mon, 3 Mar 2008 12:30:23 -0700
Message-ID: <BB375AF679D4A34E9CA8DFA650E2B04E089F23D3@onstor-exch02.onstor.net>
In-Reply-To: <20080303111338.2e074e05@ripper.onstor.net>
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Thread-Topic: concrete info on CF status problem
Thread-Index: Ach9Yq9AGGXrYqX5Rouzs9nuyJ8jYQAAC9Sw
References: <20080225140859.55145d81@ripper.onstor.net><BB375AF679D4A34E9CA8DFA650E2B04E089F185F@onstor-exch02.onstor.net><20080228162754.3c0479e9@ripper.onstor.net><BB375AF679D4A34E9CA8DFA650E2B04E089F1902@onstor-exch02.onstor.net><20080228174605.206a0a38@ripper.onstor.net><BB375AF679D4A34E9CA8DFA650E2B04E089F1971@onstor-exch02.onstor.net><20080228185120.016da30b@ripper.onstor.net><BB375AF679D4A34E9CA8DFA650E2B04E089F19C5@onstor-exch02.onstor.net><20080228202749.036b8be7@ripper.onstor.net><BB375AF679D4A34E9CA8DFA650E2B04E089F238B@onstor-exch02.onstor.net> <20080303111338.2e074e05@ripper.onstor.net>
From: "Brian Stark" <brian.stark@onstor.com>
To: "Andy Sharp" <andy.sharp@onstor.com>

You're da man!

Since this wasn't a hardware issue, I'll let you know what I decide the
spoils of the bet are.  I don't think a million donuts will do, but I'm
thinking I'll choose something that's somewhere between a lunch and a
new shiny Mac notebook...



> -----Original Message-----
> From: Andy Sharp=20
> Sent: Monday, March 03, 2008 11:14 AM
> To: Brian Stark
> Subject: Re: concrete info on CF status problem
>=20
> Yeah, I found the issue on Friday.  There was no issue.  Or,=20
> to put it another way, it may have been working when I was up=20
> there in the lab.
> But apparently Max made some change to /etc/sysctl.conf which=20
> stopped the kernel from outputting messages with a level=20
> lower than WARNING (which would be INFO and NOTICE) so my=20
> printfs were being surpressed.
> <Bastard!>
>=20
> So that is part 1 of the problem.  Part 2 is that the package=20
> pcmciautils was not being installed properly ("Laaareee!?")=20
> and may have somehow convinced the kernel to deregister the=20
> 1520 device.  I would have bet a cop a million donuts that=20
> you couldn't do that without the devices above that being=20
> deregistered as well, not to mention the pcmcia stuff should=20
> probably have stopped working entirely, but I'm still doing=20
> testing to validate this theory.
>=20
> Anyway, when I fixed the pcmciautils configuration file and=20
> turned back on kernel messages, it worked as expected on my=20
> cougar.  The exact fix is still a WIP.
>=20
> Cheers,
>=20
> a
>=20
> On Mon, 3 Mar 2008 11:02:11 -0800 "Brian Stark"
> <brian.stark@onstor.com> wrote:
>=20
> > Any updates on this?  Of course, I'm hoping that you've found and=20
> > fixed the issue...
> >=20
> >=20
> >=20
> > > -----Original Message-----
> > > From: Andy Sharp
> > > Sent: Thursday, February 28, 2008 8:28 PM
> > > To: Brian Stark
> > > Cc: Warren Gale
> > > Subject: Re: concrete info on CF status problem
> > >=20
> > > On Thu, 28 Feb 2008 19:54:27 -0800 "Brian Stark"
> > > <brian.stark@onstor.com> wrote:
> > >=20
> > > > You've sapped all my energy, so I'm taking tomorrow off. =20
> > > I'll have to
> > > > miss the ice cream social tomorrow.  Darn it.
> > >=20
> > > Ah, too bad.  You're going to miss out, because I've=20
> already picked=20
> > > up a case of Jack Daniels to bring to the ICS tomorrow,=20
> along with a=20
> > > fresh batch of grenades. Should be very interesting.
> > >=20
> > > > Seriously, I might have to be at home in the afternoon
> > > tomorrow, so I
> > > > won't be coming down to Campbell.  I'll be in Pleasanton in the=20
> > > > morning and will happily eject the CF to your heart's
> > > content.  I gave
> > > > you the terminal info, so just let me know if you want me
> > > to do this. =20
> > > > If we don't figure it out tomorrow, then I'll bring the
> > > system down to
> > > > Campbell on Monday (again).
> > >=20
> > > I'm dying to get after this, it's the only non-ignorable=20
> bug on my=20
> > > list, and I've fixed all the ignorable ones anyway.  I think I'll=20
> > > head up to P-town in the morning.
> > >=20
> > > > I look forward to winning the bet.  Then again, no matter what=20
> > > > happens, I'm sure we'll never admit it was either a kernel
> > > or hardware
> > > > bug. We'll somehow conclude it was a PROM bug and blame it on=20
> > > > Rick.
> > >=20
> > > Well duh.  I mean, wait, Rick who?  Actually I think the answer is
> > > clear: there is no bug because there couldn't be.  QA=20
> "sees things"=20
> > > all the time, it's obviously just another one of those.
> > >=20
> > > >=20
> > > > =20
> > > >=20
> > > > > -----Original Message-----
> > > > > From: Andy Sharp
> > > > > Sent: Thursday, February 28, 2008 6:51 PM
> > > > > To: Brian Stark
> > > > > Cc: Warren Gale
> > > > > Subject: Re: concrete info on CF status problem
> > > > >=20
> > > > > On Thu, 28 Feb 2008 17:51:50 -0800 "Brian Stark"
> > > > > <brian.stark@onstor.com> wrote:
> > > > >=20
> > > > > > =20
> > > > > >=20
> > > > > > > -----Original Message-----
> > > > > > > From: Andy Sharp
> > > > > > > Sent: Thursday, February 28, 2008 5:46 PM
> > > > > > > To: Brian Stark
> > > > > > > Cc: Warren Gale
> > > > > > > Subject: Re: concrete info on CF status problem
> > > > > > >=20
> > > > > > > On Thu, 28 Feb 2008 16:51:10 -0800 "Brian Stark"
> > > > > > > <brian.stark@onstor.com> wrote:
> > > > > > >=20
> > > > > > > > No, I see nothing.  This is not a board problem -- the
> > > > > interrupt
> > > > > > > > is firing and something is then reading the status
> > > reg in the
> > > > > > > > 1520.
> > > > > > >=20
> > > > > > > That's OK, I recognize your denial and it's only human,=20
> > > > > > > after all.
> > > > > >=20
> > > > > > Good one.
> > > > >=20
> > > > > Thangyouverymuch.  I'm here all week.  Until the Ice cream=20
> > > > > social (wtf?) tomorrow, that is.
> > > > >=20
> > > > > > > > I've tried your old Rev 2 board, a Rev 3 board, and a
> > > > > Rev 4 board.
> > > > > > > > Plus, I can show you the diffs in the netlists=20
> between Rev
> > > > > > > 2 - Rev 4,
> > > > > > > > and there's nothing in the hardware that's changed with
> > > > > any of the
> > > > > > > > connections between CF, 1520, and 1125.
> > > > > > >=20
> > > > > > > What about the 1520 itself?  Different rev, different
> > > part no.,
> > > > > > > different supplier, different version? Just because
> > > > > nothing changed
> > > > > > > in the netlists doesn't rule out something not right with=20
> > > > > > > the way they were built.  Or if we switched to a slightly
> > > different
> > > > > > > part that's RoHS compatible or something.
> > > > > >=20
> > > > > > Nope, the 1520 is sole-sourced to TI and it hasn't=20
> changed in=20
> > > > > > eons. Nice try.  It ain't the 1520 or the board -- the=20
> > > > > > interrupt
> > > > > is firing
> > > > > > to the 1125 and then the status reg is read.  I think I
> > > > > need to draw
> > > > > > you a picture.
> > > > >=20
> > > > > Damn, you're right, its most likely a kernel bug.  It=20
> obviously=20
> > > > > means nothing that between PROM, software and=20
> hardware, it's the=20
> > > > > only thing that hasn't changed in months.
> > > > > Because it couldn't be a hardware issue, despite that=20
> this all=20
> > > > > started on the affected systems immediately after=20
> they had their=20
> > > > > boards replaced.
> > > > >=20
> > > > > On a serious side, can my old board be set up in the=20
> lab here in=20
> > > > > Camphell tomorrow?  Are you or someone coming here=20
> tomorrow?  I=20
> > > > > assume so, you guys couldn't be contemplating missing the ICS.
> > > > >=20
> > >=20
>=20
