It seems to work ok.  I use it on systems where the budget doesn't allow a
hardware RAID controller.  I beat up on SW RAID1 for a while on a system
built just for that purpose.  

First read the various files which comes with raidtools.  On a RH system,
it's in /usr/doc/raidtools*.

On the RH distro, you can install the system using RAID1 from the gui
install.  The text install doesn't seem to have it.  Don't skip creating
a bootdisk during the install.  

A few items which bear mention:

1. When updating the kernel, you will have to manually update the ramdisk
which gets loaded during boot (/boot/initrd*).  It will probably be
loading the SCSI controller driver and the raid1.o modules.  Copy the
updated versions from /lib/modules to the /lib directory on the initrd.  
You should also update a copy of the bootdisk you created during the
install process.

2. The command to rebuild the array after a drive failure was a bit
obscure (at least for me it was): raidhotadd /dev/md? /dev/sd<new disk>.

3. You should write a script which monitors /proc/mdstat and emails
somebody when a drive fails.  

Anyway, this should at least get you started.

rna

On Thu, 14 Sep 2000, Joel Dudley wrote:

> OK, here we go.  I am setting up a new data server here at work with bad butt SCSI 160 10, 000 RPM drives and I plan on using linux as the OS.  However, my boss indicated that he wanted a second drive inside to mirror the SCSI data disk like NT mirroring.  This, of course, is because I was not authorized to purchase RAID which is what we need.  Anyway, how would I accomplish "NT like" disk mirroring in linux.  I have a few ideas on the matter but I hope some of you have experience with this.  Should I just setup software raid?  Thanks.  I will carry this question to the meeting tonight so look for a big tall goon in a blue shirt and white shorts if you have comments on this.
> 
> - Joel
> ----------------------------------------------------------------------------
> "For the scientist who has lived by his faith in the power of reason, the
> story ends like a bad dream. He has scaled the mountains of ignorance; he is
> about to conquer the highest peak; as he pulls himself over the final rock,
> he is greeted by a band of theologians who have been sitting there for
> centuries."
> 
> - Dr. Robert Jastrow
>         Founder of NASA"s Goddard Institute for space studies
> 
> 
>