2.6.19-git20 cciss: cmd f7b00000 timedout

2.6.19-git20 cciss: cmd f7b00000 timedout

[Steve Roemen]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

All,
    I tried out the 2.6.19-git20 kernel on one of my machines (HP
DL380 G3) that has the on board 5i controller (disabled),
2 smart array 642 controllers.

I get the error (cciss: cmd f7b00000 timedout) with Buffer I/O error
on device cciss/c (all cards, and disks) logical block 0, 1, 2, etc



Steve Roemen

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFgZk5cA4cgQlZoQ8RAgUYAKCCfn+SGLPy+/Ng/7Lyh/m/AigiJwCcDEO3
vNUZctiG+TxZ2oPcOVVigRY=
=6WKY
-----END PGP SIGNATURE-----

Re: 2.6.19-git20 cciss: cmd f7b00000 timedout

[Jens Axboe]

On Thu, Dec 14 2006, Steve Roemen wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> All,
>     I tried out the 2.6.19-git20 kernel on one of my machines (HP
> DL380 G3) that has the on board 5i controller (disabled),
> 2 smart array 642 controllers.
> 
> I get the error (cciss: cmd f7b00000 timedout) with Buffer I/O error
> on device cciss/c (all cards, and disks) logical block 0, 1, 2, etc

I saw this on another box, but it works on the ones that I have. Does
2.6.19 work? Any chance you can try and narrow down when it broke?

-- 
Jens Axboe

Re: 2.6.19-git20 cciss: cmd f7b00000 timedout

[Steve Roemen]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

2.6.19 does the same thing, except it's a cmd f7f00000 timedout error.

2.6.18 works just fine though.

Steve

Jens Axboe wrote:
> On Thu, Dec 14 2006, Steve Roemen wrote:
>> -----BEGIN PGP SIGNED MESSAGE-----
>> Hash: SHA1
>>
>> All,
>>     I tried out the 2.6.19-git20 kernel on one of my machines (HP
>> DL380 G3) that has the on board 5i controller (disabled),
>> 2 smart array 642 controllers.
>>
>> I get the error (cciss: cmd f7b00000 timedout) with Buffer I/O error
>> on device cciss/c (all cards, and disks) logical block 0, 1, 2, etc
>
> I saw this on another box, but it works on the ones that I have. Does
> 2.6.19 work? Any chance you can try and narrow down when it broke?
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFgag7cA4cgQlZoQ8RAiXhAJ9eRe0Nh3JqFKO2H6iYKmxsVIKGDACgrCrx
QWG2sAQH4bL8+fa5mPnVhnM=
=DxGc
-----END PGP SIGNATURE-----

Re: 2.6.19-git20 cciss: cmd f7b00000 timedout

[Jens Axboe]

On Thu, Dec 14 2006, Steve Roemen wrote:
>
> 2.6.19 does the same thing, except it's a cmd f7f00000 timedout error.<br>

Please don't send html only emails, that's just impossible to work with.

> 2.6.18 works just fine though.<br>

So the question is when it broke between 2.6.18 and 2.6.19. Can you try
and pin it down? Either use git bisect, or just start with the 2.6.19-rc
kernels and see which is the last one that works.

-- 
Jens Axboe

RE: 2.6.19-git20 cciss: cmd f7b00000 timedout

[Miller, Mike (OS Dev)]

 

> -----Original Message-----
> From: Jens Axboe [mailto:jens.axboe@oracle.com] 
> Sent: Thursday, December 14, 2006 12:51 PM
> To: Steve Roemen
> Cc: LKML; ISS StorageDev; Miller, Mike (OS Dev)
> Subject: Re: 2.6.19-git20 cciss: cmd f7b00000 timedout
> 
> On Thu, Dec 14 2006, Steve Roemen wrote:
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA1
> > 
> > All,
> >     I tried out the 2.6.19-git20 kernel on one of my machines (HP 
> > DL380 G3) that has the on board 5i controller (disabled),
> > 2 smart array 642 controllers.
> > 
> > I get the error (cciss: cmd f7b00000 timedout) with Buffer 
> I/O error 
> > on device cciss/c (all cards, and disks) logical block 0, 1, 2, etc
> 
> I saw this on another box, but it works on the ones that I have. Does
> 2.6.19 work? Any chance you can try and narrow down when it broke?

Jens/Steve:
We also encountered a time out issue on the 642. This one is connected
to an MSA500. Do either of you have MSA500? What controller fw are you
running? Check /proc/driver/cciss/ccissN.

mikem


> 
> --
> Jens Axboe
> 
> 

Re: 2.6.19-git20 cciss: cmd f7b00000 timedout

[Steve Roemen]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

MIke, Yes it's connected to a MSA 500 G2

the two 642's firmware are 2.34 for card 1, and 2.58 for card 2


Steve
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFgbB1cA4cgQlZoQ8RAoWGAKCPgdqX2QYl69V0i7cJACmP3LJ40wCfX932
ZEZEQGPhxYhJej0ovQ+6Spw=
=uIik
-----END PGP SIGNATURE-----

Re: 2.6.19-git20 cciss: cmd f7b00000 timedout

[Jens Axboe]

On Thu, Dec 14 2006, Miller, Mike (OS Dev) wrote:
>  
> 
> > -----Original Message-----
> > From: Jens Axboe [mailto:jens.axboe@oracle.com] 
> > Sent: Thursday, December 14, 2006 12:51 PM
> > To: Steve Roemen
> > Cc: LKML; ISS StorageDev; Miller, Mike (OS Dev)
> > Subject: Re: 2.6.19-git20 cciss: cmd f7b00000 timedout
> > 
> > On Thu, Dec 14 2006, Steve Roemen wrote:
> > > -----BEGIN PGP SIGNED MESSAGE-----
> > > Hash: SHA1
> > > 
> > > All,
> > >     I tried out the 2.6.19-git20 kernel on one of my machines (HP 
> > > DL380 G3) that has the on board 5i controller (disabled),
> > > 2 smart array 642 controllers.
> > > 
> > > I get the error (cciss: cmd f7b00000 timedout) with Buffer 
> > I/O error 
> > > on device cciss/c (all cards, and disks) logical block 0, 1, 2, etc
> > 
> > I saw this on another box, but it works on the ones that I have. Does
> > 2.6.19 work? Any chance you can try and narrow down when it broke?
> 
> Jens/Steve:
> We also encountered a time out issue on the 642. This one is connected
> to an MSA500. Do either of you have MSA500? What controller fw are you
> running? Check /proc/driver/cciss/ccissN.

I think the case I know about here is an MSA500 as well, I can check in
the morning.

-- 
Jens Axboe

RE: 2.6.19-git20 cciss: cmd f7b00000 timedout

[Miller, Mike (OS Dev)]

 

> -----Original Message-----
> From: Steve Roemen [mailto:stever@carlislefsp.com] 
> Sent: Thursday, December 14, 2006 2:14 PM
> To: Miller, Mike (OS Dev)
> Cc: Jens Axboe; LKML; ISS StorageDev; Frazier, Daniel Kent
> Subject: Re: 2.6.19-git20 cciss: cmd f7b00000 timedout
> 
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> MIke, Yes it's connected to a MSA 500 G2
> 
> the two 642's firmware are 2.34 for card 1, and 2.58 for card 2

What's the fw on the msa? You should be able to get that from the LCD
display.

mikem

> 
> 
> Steve
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.3 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
> 
> iD8DBQFFgbB1cA4cgQlZoQ8RAoWGAKCPgdqX2QYl69V0i7cJACmP3LJ40wCfX932
> ZEZEQGPhxYhJej0ovQ+6Spw=
> =uIik
> -----END PGP SIGNATURE-----
> 
> 

Re: 2.6.19-git20 cciss: cmd f7b00000 timedout

[dann frazier]

On Thu, Dec 14, 2006 at 01:44:34PM -0600, Miller, Mike (OS Dev) wrote:
>  
> 
> > -----Original Message-----
> > From: Jens Axboe [mailto:jens.axboe@oracle.com] 
> > Sent: Thursday, December 14, 2006 12:51 PM
> > To: Steve Roemen
> > Cc: LKML; ISS StorageDev; Miller, Mike (OS Dev)
> > Subject: Re: 2.6.19-git20 cciss: cmd f7b00000 timedout
> > 
> > On Thu, Dec 14 2006, Steve Roemen wrote:
> > > -----BEGIN PGP SIGNED MESSAGE-----
> > > Hash: SHA1
> > > 
> > > All,
> > >     I tried out the 2.6.19-git20 kernel on one of my machines (HP 
> > > DL380 G3) that has the on board 5i controller (disabled),
> > > 2 smart array 642 controllers.
> > > 
> > > I get the error (cciss: cmd f7b00000 timedout) with Buffer 
> > I/O error 
> > > on device cciss/c (all cards, and disks) logical block 0, 1, 2, etc
> > 
> > I saw this on another box, but it works on the ones that I have. Does
> > 2.6.19 work? Any chance you can try and narrow down when it broke?
> 
> Jens/Steve:
> We also encountered a time out issue on the 642. This one is connected
> to an MSA500. Do either of you have MSA500? What controller fw are you
> running? Check /proc/driver/cciss/ccissN.

fyi, we've been seeing this in Debian too (which is why Mike added me
to the CC list), and I've narrowed it down to the 2TB patch that went
into 2.6.19:
  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=402787
-- 
dann frazier | HP Open Source and Linux Organization

RE: 2.6.19-git20 cciss: cmd f7b00000 timedout

[Miller, Mike (OS Dev)]

 

> -----Original Message-----
> From: Frazier, Daniel Kent 
> Sent: Thursday, December 14, 2006 3:12 PM
> To: Miller, Mike (OS Dev)
> Cc: Jens Axboe; Steve Roemen; LKML; ISS StorageDev
> Subject: Re: 2.6.19-git20 cciss: cmd f7b00000 timedout
> 
> On Thu, Dec 14, 2006 at 01:44:34PM -0600, Miller, Mike (OS Dev) wrote:
> >  
> > 
> > > -----Original Message-----
> > > From: Jens Axboe [mailto:jens.axboe@oracle.com]
> > > Sent: Thursday, December 14, 2006 12:51 PM
> > > To: Steve Roemen
> > > Cc: LKML; ISS StorageDev; Miller, Mike (OS Dev)
> > > Subject: Re: 2.6.19-git20 cciss: cmd f7b00000 timedout
> > > 
> > > On Thu, Dec 14 2006, Steve Roemen wrote:
> > > > -----BEGIN PGP SIGNED MESSAGE-----
> > > > Hash: SHA1
> > > > 
> > > > All,
> > > >     I tried out the 2.6.19-git20 kernel on one of my 
> machines (HP 
> > > > DL380 G3) that has the on board 5i controller (disabled),
> > > > 2 smart array 642 controllers.
> > > > 
> > > > I get the error (cciss: cmd f7b00000 timedout) with Buffer
> > > I/O error
> > > > on device cciss/c (all cards, and disks) logical block 0, 1, 2, 
> > > > etc
> > > 
> > > I saw this on another box, but it works on the ones that I have. 
> > > Does
> > > 2.6.19 work? Any chance you can try and narrow down when it broke?
> > 
> > Jens/Steve:
> > We also encountered a time out issue on the 642. This one 
> is connected 
> > to an MSA500. Do either of you have MSA500? What controller 
> fw are you 
> > running? Check /proc/driver/cciss/ccissN.
> 
> fyi, we've been seeing this in Debian too (which is why Mike 
> added me to the CC list), and I've narrowed it down to the 
> 2TB patch that went into 2.6.19:
>   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=402787

Hmmmm. Dann, did you see this on 32-bit Debian? I have this running in
the lab on x86_64 and ia64. Someone else was _supposed_ to test ia32 for
me. Dammit.

Jens/Steve:
Are your os'es 32-bit?

mikem


> --


> dann frazier | HP Open Source and Linux Organization
> 

Re: 2.6.19-git20 cciss: cmd f7b00000 timedout

[Steve Roemen]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Yes,  32bit Debian

Miller, Mike (OS Dev) wrote:
> 
>
>> -----Original Message-----
>> From: Frazier, Daniel Kent
>> Sent: Thursday, December 14, 2006 3:12 PM
>> To: Miller, Mike (OS Dev)
>> Cc: Jens Axboe; Steve Roemen; LKML; ISS StorageDev
>> Subject: Re: 2.6.19-git20 cciss: cmd f7b00000 timedout
>>
>> On Thu, Dec 14, 2006 at 01:44:34PM -0600, Miller, Mike (OS Dev) wrote:
>>> 
>>>
>>>> -----Original Message-----
>>>> From: Jens Axboe [mailto:jens.axboe@oracle.com]
>>>> Sent: Thursday, December 14, 2006 12:51 PM
>>>> To: Steve Roemen
>>>> Cc: LKML; ISS StorageDev; Miller, Mike (OS Dev)
>>>> Subject: Re: 2.6.19-git20 cciss: cmd f7b00000 timedout
>>>>
>>>> On Thu, Dec 14 2006, Steve Roemen wrote:
>>>>> -----BEGIN PGP SIGNED MESSAGE-----
>>>>> Hash: SHA1
>>>>>
>>>>> All,
>>>>>     I tried out the 2.6.19-git20 kernel on one of my
>> machines (HP
>>>>> DL380 G3) that has the on board 5i controller (disabled),
>>>>> 2 smart array 642 controllers.
>>>>>
>>>>> I get the error (cciss: cmd f7b00000 timedout) with Buffer
>>>> I/O error
>>>>> on device cciss/c (all cards, and disks) logical block 0, 1, 2,
>>>>> etc
>>>> I saw this on another box, but it works on the ones that I have.
>>>> Does
>>>> 2.6.19 work? Any chance you can try and narrow down when it broke?
>>> Jens/Steve:
>>> We also encountered a time out issue on the 642. This one
>> is connected
>>> to an MSA500. Do either of you have MSA500? What controller
>> fw are you
>>> running? Check /proc/driver/cciss/ccissN.
>> fyi, we've been seeing this in Debian too (which is why Mike
>> added me to the CC list), and I've narrowed it down to the
>> 2TB patch that went into 2.6.19:
>>   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=402787
>
> Hmmmm. Dann, did you see this on 32-bit Debian? I have this running in
> the lab on x86_64 and ia64. Someone else was _supposed_ to test ia32 for
> me. Dammit.
>
> Jens/Steve:
> Are your os'es 32-bit?
>
> mikem
>
>
>> --
>
>
>> dann frazier | HP Open Source and Linux Organization
>>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFgc6UcA4cgQlZoQ8RAnRHAKDRshyyEIJTECKRTe/ghkNNkC8a0QCfQYnO
To0nPyou9F3OWbyxzTnZ5JU=
=0z54
-----END PGP SIGNATURE-----

Re: 2.6.19-git20 cciss: cmd f7b00000 timedout

[dann frazier]

On Thu, Dec 14, 2006 at 04:16:39PM -0600, Miller, Mike (OS Dev) wrote:
> Hmmmm. Dann, did you see this on 32-bit Debian?

yep - all reports I've seen so far are on i386

Re: 2.6.19-git20 cciss: cmd f7b00000 timedout

[Jens Axboe]

On Thu, Dec 14 2006, Miller, Mike (OS Dev) wrote:
>  
> 
> > -----Original Message-----
> > From: Frazier, Daniel Kent 
> > Sent: Thursday, December 14, 2006 3:12 PM
> > To: Miller, Mike (OS Dev)
> > Cc: Jens Axboe; Steve Roemen; LKML; ISS StorageDev
> > Subject: Re: 2.6.19-git20 cciss: cmd f7b00000 timedout
> > 
> > On Thu, Dec 14, 2006 at 01:44:34PM -0600, Miller, Mike (OS Dev) wrote:
> > >  
> > > 
> > > > -----Original Message-----
> > > > From: Jens Axboe [mailto:jens.axboe@oracle.com]
> > > > Sent: Thursday, December 14, 2006 12:51 PM
> > > > To: Steve Roemen
> > > > Cc: LKML; ISS StorageDev; Miller, Mike (OS Dev)
> > > > Subject: Re: 2.6.19-git20 cciss: cmd f7b00000 timedout
> > > > 
> > > > On Thu, Dec 14 2006, Steve Roemen wrote:
> > > > > -----BEGIN PGP SIGNED MESSAGE-----
> > > > > Hash: SHA1
> > > > > 
> > > > > All,
> > > > >     I tried out the 2.6.19-git20 kernel on one of my 
> > machines (HP 
> > > > > DL380 G3) that has the on board 5i controller (disabled),
> > > > > 2 smart array 642 controllers.
> > > > > 
> > > > > I get the error (cciss: cmd f7b00000 timedout) with Buffer
> > > > I/O error
> > > > > on device cciss/c (all cards, and disks) logical block 0, 1, 2, 
> > > > > etc
> > > > 
> > > > I saw this on another box, but it works on the ones that I have. 
> > > > Does
> > > > 2.6.19 work? Any chance you can try and narrow down when it broke?
> > > 
> > > Jens/Steve:
> > > We also encountered a time out issue on the 642. This one 
> > is connected 
> > > to an MSA500. Do either of you have MSA500? What controller 
> > fw are you 
> > > running? Check /proc/driver/cciss/ccissN.
> > 
> > fyi, we've been seeing this in Debian too (which is why Mike 
> > added me to the CC list), and I've narrowed it down to the 
> > 2TB patch that went into 2.6.19:
> >   http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=402787
> 
> Hmmmm. Dann, did you see this on 32-bit Debian? I have this running in
> the lab on x86_64 and ia64. Someone else was _supposed_ to test ia32 for
> me. Dammit.
> 
> Jens/Steve:
> Are your os'es 32-bit?

Yep, the machine I saw this on is 32-bit. The one I tested on myself is
x86-64, and that works. It's not driving an MSA500 either, but it's a
data point for you.

-- 
Jens Axboe