mdadm message (auto-read-only) on "/dev/md1" = swap file system

Dear Linux -raid community masters,


I have a quick question regarding something i noticed on my first time
mdadm linux raid configuration server:


Current System:

Debian Lenny 5.0
Two 250GB hard drives identical brand, size, model, speed, etc... they
are the same.
3GB memory ram
1 AMD athlon 2500 CPU

Here I am pasting an example of another machine I configure with the
same result notice the "(auto-read-only)" message on the swap file
system:


md2 : active raid1 hde3[0] hdg3[1]
153557184 blocks [2/2] [UU]

md1 : active (auto-read-only) raid1 hde2[0] hdg2[1]
2634560 blocks [2/2] [UU]



md0 : active raid1 hde1[0] hdg1[1]
96320 blocks [2/2] [UU]

unused devices:


My question for this is:

1. Is this the way suppose to work on the device md1 = swap filesystem raid?

2. If this is normal where can I find documentation about the
different "messages" displayed on the command "cat /proc/mdstat"?

3. If this is a bug, is there a work around and what will happen if
this continue to appear?

4. I ran the command mdadm -w /dev/md1 and the message disappear, but
after rebooting the system came back up.

I appreciate any information or document, links, ideas you can provide to me.

I look forward to your response.

Sincerely,

--
Jairzhino Bolivar (Jair)
GNU/Linux & Unix Administrator
Mobile: (312)404-6530
Fax: (214)602-4405
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: mdadm message (auto-read-only) on "/dev/md1" = swap file system

On 11/06/2009, at 3:10 AM, Ing. Jair wrote:

> Dear Linux -raid community masters,
>
>
> I have a quick question regarding something i noticed on my first time
> mdadm linux raid configuration server:
>
>
> Current System:
>
> Debian Lenny 5.0
> Two 250GB hard drives identical brand, size, model, speed, etc... they
> are the same.
> 3GB memory ram
> 1 AMD athlon 2500 CPU
>
> Here I am pasting an example of another machine I configure with the
> same result notice the "(auto-read-only)" message on the swap file
> system:
>
>
> md2 : active raid1 hde3[0] hdg3[1]
> 153557184 blocks [2/2] [UU]
>
> md1 : active (auto-read-only) raid1 hde2[0] hdg2[1]
> 2634560 blocks [2/2] [UU]
>
>
>
> md0 : active raid1 hde1[0] hdg1[1]
> 96320 blocks [2/2] [UU]
>
> unused devices:
>
>
> My question for this is:
>
> 1. Is this the way suppose to work on the device md1 = swap
> filesystem raid?

I believe the array will stay read-only until the first write is
performed - at which time the array will automatically switch to read/
write.

> 2. If this is normal where can I find documentation about the
> different "messages" displayed on the command "cat /proc/mdstat"?
>
> 3. If this is a bug, is there a work around and what will happen if
> this continue to appear?
>
> 4. I ran the command mdadm -w /dev/md1 and the message disappear, but
> after rebooting the system came back up.
>
> I appreciate any information or document, links, ideas you can
> provide to me.


The following bug report shows this behaviour, and the reply shows
this behaviour as 'not a bug'.
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=464560

I can't find any more solid references to this however... This being
said, logic says that if this is correct, the array would only go read/
write when something writes to your swap space...

--
Steven Haigh

Email: netwiz@crc.id.au
Web: http://www.crc.id.au
Phone: (03) 9001 6090 - 0412 935 897
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: mdadm message (auto-read-only) on "/dev/md1" = swap file system

On 10/06/2009 18:30, Steven Haigh wrote:
> On 11/06/2009, at 3:10 AM, Ing. Jair wrote:
[...]
>> md1 : active (auto-read-only) raid1 hde2[0] hdg2[1]
>> 2634560 blocks [2/2] [UU]
[...]
>> My question for this is:
>>
>> 1. Is this the way suppose to work on the device md1 = swap filesystem
>> raid?
>
> I believe the array will stay read-only until the first write is
> performed - at which time the array will automatically switch to
> read/write.

I think so too; the boss said something similar about a month ago:
http://marc.info/?l=linux-raid&m=124178553609572&w=2

To the OP: what problem are you experiencing?

Cheers,

John.

--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: mdadm message (auto-read-only) on "/dev/md1" = swap file system

Thank you very much Steven for your reply. I am not having any issues
since the two drives are working fine. But my concern was if this
will cause problems if in the future I have a hard drive failure.

=3D)

Thank you again,

On Wed, Jun 10, 2009 at 12:30 PM, Steven Haigh wrote:
> On 11/06/2009, at 3:10 AM, Ing. Jair wrote:
>
>> Dear Linux -raid community masters,
>>
>>
>> I have a quick question regarding something i noticed on my first ti=
me
>> mdadm linux raid configuration server:
>>
>>
>> Current System:
>>
>> Debian Lenny 5.0
>> Two 250GB hard drives identical brand, size, model, speed, etc... th=
ey
>> are the same.
>> 3GB memory ram
>> 1 AMD athlon 2500 CPU
>>
>> Here I am pasting an example of another machine I configure with the
>> same result notice the "(auto-read-only)" message on the swap file
>> system:
>>
>>
>> md2 : active raid1 hde3[0] hdg3[1]
>> =A0 =A0 153557184 blocks [2/2] [UU]
>>
>> md1 : active (auto-read-only) raid1 hde2[0] hdg2[1]
>> =A0 =A0 2634560 blocks [2/2] [UU]
>>
>>
>>
>> md0 : active raid1 hde1[0] hdg1[1]
>> =A0 =A0 96320 blocks [2/2] [UU]
>>
>> unused devices:
>>
>>
>> My question for this is:
>>
>> 1. Is this the way suppose to work on the device md1 =3D swap filesy=
stem
>> raid?
>
> I believe the array will stay read-only until the first write is perf=
ormed -
> at which time the array will automatically switch to read/write.
>
>> 2. If this is normal where can I find documentation about the
>> different "messages" displayed on the command "cat /proc/mdstat"?
>>
>> 3. If this is a bug, is there a work around and what will happen if
>> this continue to appear?
>>
>> 4. I ran the command mdadm -w /dev/md1 and the message disappear, bu=
t
>> after rebooting the system came back up.
>>
>> I appreciate any information or document, links, ideas you can provi=
de to
>> me.
>
>
> The following bug report shows this behaviour, and the reply shows th=
is
> behaviour as 'not a bug'.
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=3D464560
>
> I can't find any more solid references to this however... This being =
said,
> logic says that if this is correct, the array would only go read/writ=
e when
> something writes to your swap space...
>
> --
> Steven Haigh
>
> Email: netwiz@crc.id.au
> Web: http://www.crc.id.au
> Phone: (03) 9001 6090 - 0412 935 897
> --
> To unsubscribe from this list: send the line "unsubscribe linux-raid"=
in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at =A0http://vger.kernel.org/majordomo-info.html
>



--=20
Jairzhino Bolivar (Jair)
GNU/Linux & Unix Administrator
Mobile: (312)404-6530
=46ax: (214)602-4405
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" i=
n
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

RE: mdadm message (auto-read-only) on "/dev/md1" = swap file system

The only issue I have seen with this was when I was recovering a disk a=
nd the swap partition was rebuilding when my system reset for another r=
eason. On reboot, the swap partition was marked auto-read-only and the=
recovering member was marked as hot spare, but was not recovering. On=
ce I used mdadm --readwrite /dev/md2, the hot spare was immediately rec=
overed.

Simon.

-----Original Message-----
=46rom: linux-raid-owner@vger.kernel.org [mailto:linux-raid-owner@vger.=
kernel.org] On Behalf Of Ing. Jair
Sent: 10 June 2009 20:24
To: Steven Haigh
Cc: linux-raid@vger.kernel.org
Subject: Re: mdadm message (auto-read-only) on "/dev/md1" =3D swap file=
system

Thank you very much Steven for your reply. I am not having any issues
since the two drives are working fine. But my concern was if this
will cause problems if in the future I have a hard drive failure.

=3D)

Thank you again,

On Wed, Jun 10, 2009 at 12:30 PM, Steven Haigh wrote:
> On 11/06/2009, at 3:10 AM, Ing. Jair wrote:
>
>> Dear Linux -raid community masters,
>>
>>
>> I have a quick question regarding something i noticed on my first ti=
me
>> mdadm linux raid configuration server:
>>
>>
>> Current System:
>>
>> Debian Lenny 5.0
>> Two 250GB hard drives identical brand, size, model, speed, etc... th=
ey
>> are the same.
>> 3GB memory ram
>> 1 AMD athlon 2500 CPU
>>
>> Here I am pasting an example of another machine I configure with the
>> same result notice the "(auto-read-only)" message on the swap file
>> system:
>>
>>
>> md2 : active raid1 hde3[0] hdg3[1]
>> =A0 =A0 153557184 blocks [2/2] [UU]
>>
>> md1 : active (auto-read-only) raid1 hde2[0] hdg2[1]
>> =A0 =A0 2634560 blocks [2/2] [UU]
>>
>>
>>
>> md0 : active raid1 hde1[0] hdg1[1]
>> =A0 =A0 96320 blocks [2/2] [UU]
>>
>> unused devices:
>>
>>
>> My question for this is:
>>
>> 1. Is this the way suppose to work on the device md1 =3D swap filesy=
stem
>> raid?
>
> I believe the array will stay read-only until the first write is perf=
ormed -
> at which time the array will automatically switch to read/write.
>
>> 2. If this is normal where can I find documentation about the
>> different "messages" displayed on the command "cat /proc/mdstat"?
>>
>> 3. If this is a bug, is there a work around and what will happen if
>> this continue to appear?
>>
>> 4. I ran the command mdadm -w /dev/md1 and the message disappear, bu=
t
>> after rebooting the system came back up.
>>
>> I appreciate any information or document, links, ideas you can provi=
de to
>> me.
>
>
> The following bug report shows this behaviour, and the reply shows th=
is
> behaviour as 'not a bug'.
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=3D464560
>
> I can't find any more solid references to this however... This being =
said,
> logic says that if this is correct, the array would only go read/writ=
e when
> something writes to your swap space...
>
> --
> Steven Haigh
>
> Email: netwiz@crc.id.au
> Web: http://www.crc.id.au
> Phone: (03) 9001 6090 - 0412 935 897
> --
> To unsubscribe from this list: send the line "unsubscribe linux-raid"=
in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at =A0http://vger.kernel.org/majordomo-info.html
>



--=20
Jairzhino Bolivar (Jair)
GNU/Linux & Unix Administrator
Mobile: (312)404-6530
=46ax: (214)602-4405
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" i=
n
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" i=
n
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: mdadm message (auto-read-only) on "/dev/md1" = swap file system

I see.

I will keep my eye on the disks and also on the mailing list =3D)

Thank you very much,

On Thu, Jun 11, 2009 at 5:00 AM, Simon Jackson wr=
ote:
> The only issue I have seen with this was when I was recovering a disk=
and the swap partition was rebuilding when my system reset for another=
reason. =A0On reboot, the swap partition was marked auto-read-only and=
the recovering member was marked as hot spare, but was not recovering.=
=A0Once I used mdadm --readwrite /dev/md2, the hot spare was immediate=
ly recovered.
>
> Simon.
>
> -----Original Message-----
> From: linux-raid-owner@vger.kernel.org [mailto:linux-raid-owner@vger.=
kernel.org] On Behalf Of Ing. Jair
> Sent: 10 June 2009 20:24
> To: Steven Haigh
> Cc: linux-raid@vger.kernel.org
> Subject: Re: mdadm message (auto-read-only) on "/dev/md1" =3D swap fi=
le system
>
> Thank you very much Steven for your reply. I am not having any issues
> since the two drives are working fine. =A0But my concern was if this
> will cause problems if in the future I have a hard drive failure.
>
> =3D)
>
> Thank you again,
>
> On Wed, Jun 10, 2009 at 12:30 PM, Steven Haigh wrot=
e:
>> On 11/06/2009, at 3:10 AM, Ing. Jair wrote:
>>
>>> Dear Linux -raid community masters,
>>>
>>>
>>> I have a quick question regarding something i noticed on my first t=
ime
>>> mdadm linux raid configuration server:
>>>
>>>
>>> Current System:
>>>
>>> Debian Lenny 5.0
>>> Two 250GB hard drives identical brand, size, model, speed, etc... t=
hey
>>> are the same.
>>> 3GB memory ram
>>> 1 AMD athlon 2500 CPU
>>>
>>> Here I am pasting an example of another machine I configure with th=
e
>>> same result notice the "(auto-read-only)" message on the swap file
>>> system:
>>>
>>>
>>> md2 : active raid1 hde3[0] hdg3[1]
>>> =A0 =A0 153557184 blocks [2/2] [UU]
>>>
>>> md1 : active (auto-read-only) raid1 hde2[0] hdg2[1]
>>> =A0 =A0 2634560 blocks [2/2] [UU]
>>>
>>>
>>>
>>> md0 : active raid1 hde1[0] hdg1[1]
>>> =A0 =A0 96320 blocks [2/2] [UU]
>>>
>>> unused devices:
>>>
>>>
>>> My question for this is:
>>>
>>> 1. Is this the way suppose to work on the device md1 =3D swap files=
ystem
>>> raid?
>>
>> I believe the array will stay read-only until the first write is per=
formed -
>> at which time the array will automatically switch to read/write.
>>
>>> 2. If this is normal where can I find documentation about the
>>> different "messages" displayed on the command "cat /proc/mdstat"?
>>>
>>> 3. If this is a bug, is there a work around and what will happen if
>>> this continue to appear?
>>>
>>> 4. I ran the command mdadm -w /dev/md1 and the message disappear, b=
ut
>>> after rebooting the system came back up.
>>>
>>> I appreciate any information or document, links, ideas you can prov=
ide to
>>> me.
>>
>>
>> The following bug report shows this behaviour, and the reply shows t=
his
>> behaviour as 'not a bug'.
>> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=3D464560
>>
>> I can't find any more solid references to this however... This being=
said,
>> logic says that if this is correct, the array would only go read/wri=
te when
>> something writes to your swap space...
>>
>> --
>> Steven Haigh
>>
>> Email: netwiz@crc.id.au
>> Web: http://www.crc.id.au
>> Phone: (03) 9001 6090 - 0412 935 897
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-raid=
" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at =A0http://vger.kernel.org/majordomo-info.html
>>
>
>
>
> --
> Jairzhino Bolivar (Jair)
> GNU/Linux & Unix Administrator
> Mobile: (312)404-6530
> Fax: (214)602-4405
> --
> To unsubscribe from this list: send the line "unsubscribe linux-raid"=
in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at =A0http://vger.kernel.org/majordomo-info.html
>



--=20
Jairzhino Bolivar (Jair)
GNU/Linux & Unix Administrator
Mobile: (312)404-6530
=46ax: (214)602-4405
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" i=
n
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html