weird messages in blktrace pertaining to manor and minor numbers

weird messages in blktrace pertaining to manor and minor numbers

[Srinivasa R Chamarthy]

Hello,

I am running blktrace against scsi disks on linux. I get the
following output which i am not able to infer some of the messages
in it.

2057,8440  1     8043     0.580951687  2047  A   W 14666 + 1 <-
(253,0) 14666
  8,0    1     8044     0.580951812  2047  Q   W 14666 + 1
[mkfs.ext3]
  8,0    1     8045     0.580952000  2047  M   W 14666 + 1
[mkfs.ext3]
2057,8440  1     8046     0.580953656  2047  A   W 14667 + 1 <-
(253,0) 14667
  8,0    1     8047     0.580953781  2047  Q   W 14667 + 1
[mkfs.ext3]
  8,0    1     8048     0.580953968  2047  M   W 14667 + 1
[mkfs.ext3]
2057,8440  1     8049     0.580955406  2047  A   W 14668 + 1 <-
(253,0) 14668
  8,0    1     8050     0.580955718  2047  Q   W 14668 + 1
[mkfs.ext3]

For some of the lines, it shows the list of major and minor numbers
(2057,8440) which are not there applicable to my system. I am not
able to understand the output pertaining to these numbers and i do
not find any documentation about them. Can anybody let me know what
this output means?

I have run blktrace as following:

blktrace -d /dev/sda /dev/sdb -o - | blkparse -i -

and ran mkfs.ext3 to the scsi disks.

-- 
Srinivasa R Chamarthy (schamart@linux.vnet.ibm.com)

Re: weird messages in blktrace pertaining to manor and minor

[Alan D. Brunelle]

On Thu, 2009-07-30 at 18:24 +0530, Srinivasa R Chamarthy wrote:
> Hello,
> 
> I am running blktrace against scsi disks on linux. I get the
> following output which i am not able to infer some of the messages
> in it.
> 
> 2057,8440  1     8043     0.580951687  2047  A   W 14666 + 1 <-
> (253,0) 14666
>   8,0    1     8044     0.580951812  2047  Q   W 14666 + 1
> [mkfs.ext3]
>   8,0    1     8045     0.580952000  2047  M   W 14666 + 1
> [mkfs.ext3]
> 2057,8440  1     8046     0.580953656  2047  A   W 14667 + 1 <-
> (253,0) 14667
>   8,0    1     8047     0.580953781  2047  Q   W 14667 + 1
> [mkfs.ext3]
>   8,0    1     8048     0.580953968  2047  M   W 14667 + 1
> [mkfs.ext3]
> 2057,8440  1     8049     0.580955406  2047  A   W 14668 + 1 <-
> (253,0) 14668
>   8,0    1     8050     0.580955718  2047  Q   W 14668 + 1
> [mkfs.ext3]
> 
> For some of the lines, it shows the list of major and minor numbers
> (2057,8440) which are not there applicable to my system. I am not
> able to understand the output pertaining to these numbers and i do
> not find any documentation about them. Can anybody let me know what
> this output means?
> 
> I have run blktrace as following:
> 
> blktrace -d /dev/sda /dev/sdb -o - | blkparse -i -
> 
> and ran mkfs.ext3 to the scsi disks.
> 

What Linux version are you using? Issues in MD/DM mappings were fixed in
kernels post-RHEL5/SLES10 for example. 

Alan

Re: weird messages in blktrace pertaining to manor and minor

[Srinivasa R Chamarthy]

Hi Alan,

Thanks for the reply.

I am using RHEL5.4 Snap-3 with kernel version 2.6.18-158.el5 for
testing on s390x for test.

The blktrace version is:
blktrace version 1.0.0

On Thu, Jul 30, 2009 at 09:03:16AM -0400, Alan D. Brunelle wrote:
> On Thu, 2009-07-30 at 18:24 +0530, Srinivasa R Chamarthy wrote:
> > Hello,
> > 
> > I am running blktrace against scsi disks on linux. I get the
> > following output which i am not able to infer some of the messages
> > in it.
> > 
> > 2057,8440  1     8043     0.580951687  2047  A   W 14666 + 1 <-
> > (253,0) 14666
> >   8,0    1     8044     0.580951812  2047  Q   W 14666 + 1
> > [mkfs.ext3]
> >   8,0    1     8045     0.580952000  2047  M   W 14666 + 1
> > [mkfs.ext3]
> > 2057,8440  1     8046     0.580953656  2047  A   W 14667 + 1 <-
> > (253,0) 14667
> >   8,0    1     8047     0.580953781  2047  Q   W 14667 + 1
> > [mkfs.ext3]
> >   8,0    1     8048     0.580953968  2047  M   W 14667 + 1
> > [mkfs.ext3]
> > 2057,8440  1     8049     0.580955406  2047  A   W 14668 + 1 <-
> > (253,0) 14668
> >   8,0    1     8050     0.580955718  2047  Q   W 14668 + 1
> > [mkfs.ext3]
> > 
> > For some of the lines, it shows the list of major and minor numbers
> > (2057,8440) which are not there applicable to my system. I am not
> > able to understand the output pertaining to these numbers and i do
> > not find any documentation about them. Can anybody let me know what
> > this output means?
> > 
> > I have run blktrace as following:
> > 
> > blktrace -d /dev/sda /dev/sdb -o - | blkparse -i -
> > 
> > and ran mkfs.ext3 to the scsi disks.
> > 
> 
> What Linux version are you using? Issues in MD/DM mappings were fixed in
> kernels post-RHEL5/SLES10 for example. 
> 
> Alan
> 

-- 
Srinivasa R Chamarthy (schamart@linux.vnet.ibm.com)
---------------------------------------------------
Linux on System z (Component & System Verification),
IBM India Systems and Technology Labs, Bangalore.
+91-80-41776451 / ITN: *869246992

Re: weird messages in blktrace pertaining to manor and minor numbers

[Eric Sandeen]

Alan D. Brunelle wrote:
> On Thu, 2009-07-30 at 18:24 +0530, Srinivasa R Chamarthy wrote:
>> Hello,
>>
>> I am running blktrace against scsi disks on linux. I get the
>> following output which i am not able to infer some of the messages
>> in it.

...

> What Linux version are you using? Issues in MD/DM mappings were fixed in
> kernels post-RHEL5/SLES10 for example. 
> 
> Alan

Alan, if you happen to know, would you mind pointing me at the commits?
 I had a quick look through the blktrace changelogs and nothing popped
out at me...

Thanks,
-Eric

Re: weird messages in blktrace pertaining to manor and minor

[Alan D. Brunelle]

On Thu, 2009-07-30 at 08:57 -0500, Eric Sandeen wrote:
> Alan D. Brunelle wrote:
> > On Thu, 2009-07-30 at 18:24 +0530, Srinivasa R Chamarthy wrote:
> >> Hello,
> >>
> >> I am running blktrace against scsi disks on linux. I get the
> >> following output which i am not able to infer some of the messages
> >> in it.
> 
> ...
> 
> > What Linux version are you using? Issues in MD/DM mappings were fixed in
> > kernels post-RHEL5/SLES10 for example. 
> > 
> > Alan
> 
> Alan, if you happen to know, would you mind pointing me at the commits?
>  I had a quick look through the blktrace changelogs and nothing popped
> out at me...
> 
> Thanks,
> -Eric

Hi Eric-  

Here are some that I can recall off the top of my head - I think this
catches them all:

commit 79c5d3ce614d8fe706545c7bca2158b63db6bb5e
Author: Li Zefan <lizf@cn.fujitsu.com>
Date:   Mon May 11 15:06:46 2009 +0800

    blktrace: from-sector redundant in trace_block_remap, cleanup
    
    The last argument of block_remap prober is the original sector
    before remap, so it should be 'from', not 'to'.
    
    [ Impact: clean up ]
    
    Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
    Cc: "Alan D. Brunelle" <Alan.Brunelle@hp.com>
    Cc: Jens Axboe <jens.axboe@oracle.com>
    Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
    Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
    LKML-Reference: <4A07CE86.5090301@cn.fujitsu.com>
    Signed-off-by: Ingo Molnar <mingo@elte.hu>

commit 22a7c31a9659deaddafbbcec6562d44141e84474
Author: Alan D. Brunelle <Alan.Brunelle@hp.com>
Date:   Mon May 4 16:35:08 2009 -0400

    blktrace: from-sector redundant in trace_block_remap
    
    Remove redundant from-sector parameter: it's /always/ the bio's
sector
    passed in.
    
    [ Impact: cleanup ]
    
    Signed-off-by: Alan D. Brunelle <alan.brunelle@hp.com>
    Reviewed-by: Li Zefan <lizf@cn.fujitsu.com>
    Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
    Cc: Jens Axboe <jens.axboe@oracle.com>
    Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
    LKML-Reference: <49FF517C.7000503@hp.com>commit
a42aaa3bbce85ac487ad4fad5db99e8e91b7aac1
Author: Alan D. Brunelle <Alan.Brunelle@hp.com>
Date:   Mon May 4 16:27:26 2009 -0400

    blktrace: correct remap names
    
    This attempts to clarify names utilized during block I/O remap
    operations (partition, volume manager). It correctly matches up the
    /from/ information for both device & sector. This takes in the
concept
    from Kosaki Motohiro and extends it to include better naming for the
    "device_from" field.
    
    [ Impact: cleanup ]
    
    Signed-off-by: Alan D. Brunelle <alan.brunelle@hp.com>
    Reviewed-by: Li Zefan <lizf@cn.fujitsu.com>
    Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
    Cc: Jens Axboe <jens.axboe@oracle.com>
    Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
    LKML-Reference: <49FF4FAE.3000301@hp.com>
    Signed-off-by: Ingo Molnar <mingo@elte.hu>

    Signed-off-by: Ingo Molnar <mingo@elte.hu>

commit a42aaa3bbce85ac487ad4fad5db99e8e91b7aac1
Author: Alan D. Brunelle <Alan.Brunelle@hp.com>
Date:   Mon May 4 16:27:26 2009 -0400

    blktrace: correct remap names
    
    This attempts to clarify names utilized during block I/O remap
    operations (partition, volume manager). It correctly matches up the
    /from/ information for both device & sector. This takes in the
concept
    from Kosaki Motohiro and extends it to include better naming for the
    "device_from" field.
    
    [ Impact: cleanup ]
    
    Signed-off-by: Alan D. Brunelle <alan.brunelle@hp.com>
    Reviewed-by: Li Zefan <lizf@cn.fujitsu.com>
    Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
    Cc: Jens Axboe <jens.axboe@oracle.com>
    Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
    LKML-Reference: <49FF4FAE.3000301@hp.com>
    Signed-off-by: Ingo Molnar <mingo@elte.hu>