software raid

Phillip Susi psusi at cfl.rr.com
Thu Mar 23 03:46:39 UTC 2006


Your messages are missing the customary User-Agent: header identifying 
what horribly broken email client you are using, but whatever it is, it 
is very annoyingly broken.  I can't make heads or tails out of this 
message because it line wrapped the original quoted message after 
prefixing each line with a '>', resulting in a series of quoted lines 
intermixed with what appears to be a single word unquoted line.  It also 
appears to be attempting to use broken mime encoding because it is 
missing the main header and part separators.

 > It is really annoying trying to read quotes that look like
this
 > Don't you think?


Please stop using whatever horribly broken mail client you are using and 
complain loudly to its authors.


David Woyciesjes wrote:
> Carl Karsten wrote:
> 
>> I have been messing around with software raid1 - mirroring 2 drives - 
>> including the boot sectors. The goal is to be able to remove either 
>> drive and still have a functioning system.
>>
>> I realize that this is "not supported" given that "Add support for 
>> software RAID." is on the http://grub.enbug.org/TodoList but I have 
>> heard it can be done with some trickery: install grub on both drives 
> and
>> rely on the fact that /dev/hda1 and /dev/hda2 and /dev/md0 are all 
> "the
>> same thing" as long as md0 is the only one that gets written to.
>>
>> But I did run into one problem:  if you have ide master and slave, 
> and
>> you pull the master, the slave dissapears and the box/bios won't see 
> it
>> so it won't boot from it.  so I am guessing I have to make each drive 
>> the master on a seperate ide bus so that pulling one won't take out 
> the
>> other.
>>
>> Before I spend too much time trying to make this work, anyone know of 
> a
>> HowTo that outlines this?
>>
>> CarlFK
> To: ubuntu-users at lists.ubuntu.com
> In-Reply-To: <E1FM7lK-0000wT-Nf at esperanza.ubuntu.com>
> MIME-Version: 1.0
> Content-Type: multipart/alternative; 
> boundary="0-392752389-1143076396=:96754"
> Content-Transfer-Encoding: 8bit
> 
> --0-392752389-1143076396=:96754
> Content-Type: text/plain; charset=iso-8859-1
> Content-Transfer-Encoding: 8bit
> 
> 
> 
> 
> 
> 
> 
> rely on the fact that /dev/hda1 and /dev/hda2 and /dev/md0 are all "the
>> same thing" as long as md0 is the only one that gets written to.
>>
>> But I did run into one problem:  if you have ide master and slave, 
> and
>> you pull the master, the slave dissapears and the box/bios won't see 
> it
>> so it won't boot from it.  so I am guessing I have to make each drive 
>> the master on a seperate ide bus so that pulling one won't take out 
> the I haven't used software raid in some time, but wouldn't it be 
> simpler to use CSEL instead of Master and Slave in this situation, then 
> the controller wouldn't get confused with wrong jumper settings. IIRC 
> ata 100/133/66 standard 80 wire ide cables (at least oem,,,) are usually 
> intended for CSEL.
> 
> Bryann
> 
> --0-392752389-1143076396=:96754
> Content-Type: text/html; charset=iso-8859-1
> Content-Transfer-Encoding: 8bit
> 
> <br><br><br> <br> <br> <pre><tt> rely on the fact that /dev/hda1 and 
> /dev/hda2 and /dev/md0 are all <br>"the <br>> same thing" as long as 
> md0 is the only one that gets written to.<br>> <br>> But I did run 
> into one problem:  if you have ide master and slave, <br>and <br>> 
> you pull the master, the slave dissapears and the box/bios won't see 
> <br>it <br>> so it won't boot from it.  so I am guessing I have to 
> make each drive <br>> the master on a seperate ide bus so that 
> pulling one won't take out <br>the </tt></pre> I haven't used software 
> raid in some time, but wouldn't it be simpler to use CSEL instead of 
> Master and Slave in this situation, then the controller wouldn't get 
> confused with wrong jumper settings. IIRC ata 100/133/66 standard 80 
> wire ide cables (at least oem,,,) are usually intended for CSEL.<br> 
> <br> Bryann<br>
> --0-392752389-1143076396=:96754--
> 





More information about the ubuntu-users mailing list