The current situtation with Linux is when my mission critical Linux server
has a bad hard disk, I have to go to the server's site, replace hard disk
then restore from backup and/or work around the partitons in question not
being there, and make excuses to my users about why the system went down.
What I would like to ideally see is the md drivers being able to do the
* 100% mirror redunancy. When a hard disk gues the way of the dodo, nothing
changes except root gets mail saying "Hard disk failure".
* I add a note to /etc/motd that there will be 20 minutes of down time as we
swap hard disks.
* I swap hard disks.
* The data from the working hard disk is automagiclly mirrored to the new
Total downtime is 20 minutes, at a preplanned time, instrad of however long
it takes between the time the hard disk dies and I am able to get to the
server room to replace the hard disk in question.
Is Linux able to do this yet?
Unique Linux information: http://linux-rules.samiam.org/linux/linux_links.html
I reserve the right to make public any mail I receive in response to this